Frequently asked questions

An AI-native CRM and project management platform built for MSPs — everything your PSA can't do. Sales pipeline, campaigns, quote-to-project handoff, dependency-aware scheduling, time entry, and billing sync — one workspace.

In the United States. Your workspace database, the files you upload, and the backups all live in US regions, and our subprocessors process data in the United States unless we say otherwise. You'll see this stated on the signup form and in the first step of setup, and it's written into the Terms of Service and the Privacy Policy. If your organization needs its data hosted in another region, use the contact form at current.day/help/contact and we'll tell you what we can support. Regional hosting isn't available by default and is arranged case by case.

No. Current is not a PSA and it does not replace the one you run. You keep ConnectWise, Autotask, or HaloPSA for tickets, contracts, and invoicing, and Current integrates with it. What Current adds is the part your PSA never had: a real CRM, a dependency-aware scheduling engine, and an AI layer, all syncing both ways so your PSA stays the system of record.

No. It runs on top of it, and you still need it. Two-way sync keeps your PSA the source of truth for time and billing, while Current adds the CRM, the scheduling engine, and the AI layer. Nothing gets ripped out and nothing gets migrated away from it.

It means we've shipped an update and the tab you're looking at is still running the previous version. Click Refresh and the page reloads on the newest version and brings you right back to the exact page you were on — nothing you were doing is lost. If you're staff, the What's New feed also opens once after the reload so you can see what changed. The banner shows at the top of the app and at the top of the client portal, and it can't be dismissed on purpose: running an old version against the updated service can cause confusing behavior, so refreshing is the intended way past it.

Synced emails and mirrored PSA tickets carry their full body for the last 100 days. After that, a nightly job clears the stored body while keeping the subject, the people involved, every date, and a short summary — forever. Nothing disappears from the activity feed, counts and reports are unaffected, and opening an old ticket still fetches its full detail live from your PSA, which always keeps the original. Notes, meetings, calls, and quote activity are never trimmed. This keeps your workspace fast as years of mail and tickets accumulate; a tenant admin can lengthen the email window up to two years from CRM settings.

It's the full-width banner at the top of a company record for the one thing everyone must know before touching the account. Anyone who can view the company can add, edit, or clear it. On Autotask companies it syncs two ways with the Autotask Company Detail Alert, so service desk and sales always read the same warning, whichever side wrote it. Companies on ConnectWise or HaloPSA get the same banner inside Current only, and the banner labels itself accordingly.

On the company record, open the "Files" chip in the activity panel. Current lists the documents attached to that company and its opportunities in your PSA — Autotask, ConnectWise, or HaloPSA — and you can sort them by newest, oldest, name, size, or file type. Click any file to preview it in place: PDFs and images open in an expandable viewer with buttons to open full-size or download; other file types offer a download. PDF previews are drawn by Current rather than your browser's plugin, so they render on every browser including Safari — you can't select text inside the inline preview, but "Open full size" opens your browser's own viewer where you can. Nothing is copied into Current — each file is pulled from your PSA the moment you open it — and only people who can already see the company can see its files, so Partner Viewers never can. (HaloPSA file preview is still being verified against a live system; until then a HaloPSA file that can't open points you back to HaloPSA.)

Because Current now draws the preview itself instead of handing the PDF to your browser's built-in plugin, and a drawn page has no selectable text layer. That change fixed a real problem: on Safari, and on machines where Chrome is set to download PDFs rather than open them, or where a PDF extension had taken over, an inline preview came up as a blank white box while "Open full size" worked fine. Nothing in our code could detect that, so the durable fix was to stop asking the browser and render the pages ourselves — which works everywhere. If you need to select, search, or copy text, click "Open full size" and the PDF opens in your browser's own viewer, where all of that still works. This applies everywhere Current previews a PDF inline: a company's Files area, the Win Quote wizard, the quote drawer, and commission evidence.

Yes. An internal note an AE typed on a ScalePad Quoter quote — a ship-to address, a special instruction — syncs into Current and appears in the Win Quote wizard as a bold amber “Internal note from Quoter” banner at the top of the first screen. It combines the quote-level note and each line item's note (each line prefixed with its item name), and you have to tick “I've read this” before the wizard lets you continue, so a fulfilment-critical note can't be missed. If the quote has no internal note, there's no banner and no extra step. This is for ScalePad Quoter quotes only — ConnectWise CPQ and HaloPSA-native quotes don't carry it. (Field availability from the Quoter API is best-effort: notes appear when Quoter provides them, and nothing is ever invented.)

When you convert a won quote through the Win Quote wizard — into a project or a post-sale ticket — Current files the signed quote PDF onto the company in your PSA. It copies the PDF from the won opportunity to the company's own attachments and renames it to the company name, the quote/offering name, and the date you converted (for example "Acme Corp Managed Services 8-1-27.pdf"), so it appears in the company record's Files area right away — the signed paperwork lives with the account instead of buried on the opportunity. It's best-effort: if the PDF can't be filed (it's a link rather than an uploaded file, or the PSA rejects the upload), the conversion still completes and the wizard shows a non-blocking heads-up so you can file it by hand. This is live for Autotask and ConnectWise; HaloPSA filing follows once its attachment API is verified against a live system.

Autotask won't let a task run past its phase's due date, or a phase past its project's end date. When you push a task date beyond its phase on the Overview, Current now asks first — an "Extend the phase to fit this date?" confirmation — and, if you approve, moves the phase's end date (and the project's, if needed) out in Autotask so the task fits, instead of leaving you to fix it in Autotask by hand. You can also set a phase's due date directly in its header on the Overview. (This clamp is an Autotask rule; on ConnectWise, every date push is verified by read-back the same way, and a date the PSA didn't store is reported rather than silently assumed.)

That's fixed. Current had a blind spot where it could quietly record your PSA's new date in the background without actually applying it — and every later sync then read the date as "unchanged" and kept the old value, so a task moved to October in Autotask could keep showing March in Current indefinitely. Now, whenever your PSA's date disagrees with a date nobody edited inside Current, the PSA's date wins and is applied on the next sync (Autotask, ConnectWise, or HaloPSA). Dates you genuinely edited in Current are still protected — only untouched, imported dates get corrected. If an old date is still showing, run Reconcile (or Resync from Autotask on that project) from the sync health page to apply it right away.

You don't have to anymore. A task deleted in your PSA that has no logged time is now archived automatically — it drops out of every board, list, timeline, calendar, and My Day, because your PSA is the source of truth. A task that does have logged time is held in Sync health's "Removed in your PSA" list for your review, so its billing history is never removed silently: there you can Archive it (hide it, keep the time entries), Delete it, or Keep it. Deletions are detected by Reconcile, which runs nightly and on demand from the sync health page.

Autotask schedules work on business days, so a due date that lands on a weekend is stored as the next working day: it accepts the push, then quietly keeps Monday. Current recognizes the business-day move, adopts Autotask's stored date so both systems agree, clears what used to be a permanent red date-push conflict, and tells you plainly what happened — for example "stored Mon, Nov 2 instead of Sat, Oct 31." If the work must be done before the weekend, set the due date to the Friday.

You'd be told. A sync watchdog checks every few minutes that each PSA sync schedule (Autotask, ConnectWise, HaloPSA) is actually firing and that every synced record type is staying fresh. If a schedule dies or a record type stops updating, workspace admins get an in-app alert and an email, and a red or amber banner appears at the top of the Sync health page until the problem clears — a silent overnight stall can't go unnoticed anymore.

Both mean "more to come," not "broken" — the run synced what it could inside its time and request budget and resumes on the next cycle. A record type that's mid-sync and progressing under the sync watchdog's coverage shows a calm blue "Catching up" rather than amber, so an in-progress large book doesn't read as a problem. It turns amber when the watchdog flags it — no forward progress for about 2 hours, or falling behind its expected refresh window (as soon as an hour for core project data) — that's when it's genuinely worth a look. A few low-stakes record types (like holidays and time off) sit outside watchdog coverage and always show plain amber Partial while mid-sync. The note under the status says exactly why it paused (for example, which record failed to save, or that the PSA's request budget ran out) instead of a generic message. Every PSA request has a hard timeout so one stalled call can't eat a cycle, CRM records (opportunities, notes, charges, tickets) get a guaranteed slice of every cycle, and whatever has waited longest syncs first — so a catching-up row completes on its own within a few cycles. The all-systems summary chip at the top of the page counts how many record types are still catching up while everything stays green. Press Sync now to continue immediately; re-running never duplicates a record.

Search finds every task that has synced into Current — including completed ones (a done task on an active project is fully searchable by its number, name, or a fragment of either). So if a number like T20251117.0389 comes up empty, that task's row isn't in Current yet: most often its project was added to your PSA before Current started tracking it, or the task was removed here. The fix: open the task's project in Current and click the “Resync” button in the project header — it pulls every task on that project, completed included — then search again. (The Resync button shows for project managers and admins; if you don't see it, ask one to run it.) The search window now tells you this itself when a number-shaped search finds nothing.

AI briefings draft overnight, once per day — ready before your workday starts. The data underneath syncs on its own schedule: your PSA — ConnectWise, Autotask, or HaloPSA — polls every 5 minutes by default. See the sync-schedules article for the full table.

On the sign-in page, enter your email, continue to the password step, and click "Forgot password?". A one-time reset link is emailed to you (check spam); it opens a set-new-password page and signs you in. If you normally sign in with Microsoft, use your Microsoft account instead — there's no Current password to reset.

Register an app in Microsoft Entra ID, add Current's redirect URI, and paste the client ID, tenant ID, and secret into Settings → Authentication. Full walkthrough in the SSO article — including the common AADSTS errors.

Deny blocks that sign-in from reaching the pending screen again but keeps the record, so it's reversible. Delete permanently removes the login altogether — meant for a genuine orphan, like a stray login left behind by an old, torn-down test tenant. Current won't let you delete anyone who already has a real profile in your workspace. Super admins also get a one-click "Clear all stale" that sweeps clearly-abandoned logins — older than 7 days and on a domain no live workspace has claimed; denied sign-ins are never touched.

365 days. The admin audit log holds a full year of history, and it's searchable (by actor email, entity type, or entity ID), filterable by action and date range, and paginated with Load more — click any entry for its full before-and-after detail. Entries are append-only: nobody, not even a Tenant Admin, can edit or delete them.

Yes — invite them to the free, read-only partner portal: status, timeline, progress, shared documents, and co-managed comments. Financials are firewalled at the database layer. Portal viewers are unlimited and never count toward your bill.

Open the project's Documents list, hover the document's row, and click the eye icon. A crossed-out eye means internal only; a solid green eye means it is shared. It saves straight away and the document appears under Shared documents in the portal for every partner who is a member of that project. Sharing is per document and starts off, so nothing reaches a partner until you turn it on, and clicking the eye again makes it internal again. Project managers and tenant admins have the switch. Current shares the title and the link, not the file, so the partner still needs their own access to open it.

No. Rates, costs, margins, budgets, time entries, and internal notes are blocked at three database layers — row-level security, restrictive deny policies, and financial-column-free views. Not hidden; absent.

Everyone on your team with CRM access — account executives, sales managers, and admins — sees the ConnectBooster Payments card on a company, including the amounts a customer is being charged, because knowing that is the point of the feature. Two hard limits sit under that: Current never stores or shows a full card number — only the last four digits and the expiry month, everywhere — and partner-portal viewers (your customers) can never reach the payments feed at all; it's blocked from them at the database layer, not just hidden in the interface.

Only if you turn on the Autotask bridge — it's off by default. HubSpot two-way sync keeps HubSpot and Current in step but doesn't write to your PSA on its own. A separate admin toggle on the HubSpot 2-way sync card, "Create HubSpot contacts & deals in Autotask," creates qualifying HubSpot contacts and open deals in your live Autotask (as contacts and Opportunities) and keeps edits flowing both ways. It only turns on once HubSpot sync is Dual-run and live, your CRM write mode is Live, and Autotask is connected — the toggle names anything still missing. A deal crosses only when it's open, its pipeline stage is mapped to an Autotask stage, its company is already in Autotask, and it isn't a Quoter deal; only the last 365 days of HubSpot activity is in scope, and inactive companies are skipped. A Backfill button previews how many existing open deals would be created before you commit. This bridge is Autotask-specific — ConnectWise and HaloPSA workspaces can't use it.

Your PSA first: ConnectWise PSA, Autotask, or HaloPSA, each with deep two-way sync — one per workspace, chosen at signup. Then Microsoft 365 (Teams, SharePoint, Calendar, To Do), Microsoft Entra ID SSO, HubSpot, ScalePad Quoter, ConnectWise CPQ, and HaloPSA quotes (won-quote engines), ScalePad Lifecycle Manager, CrewHu, NeverBounce (bring your own account), Handwrytten (handwritten cards), and Resend (email delivery). A set of read-only reporting connectors feed your business reviews: Datto RMM, ConnectWise RMM, Datto Backup (BCDR + SaaS Protection), Datto EDR, RocketCyber, Auvik, Addigy, and Inky. And ConnectBooster / BNG Payments (a read-only daily sync) shows which of your customers are on autopay, whether it's actually charging, and when a stored card is about to expire.

No. AI is built into the platform — briefings, SBR drafting, sequence copy — all covered by the one per-user price. No tiers, no per-module SKUs, no AI surcharge.

Yes — every workspace ships with 24 built-in delivery playbooks grouped into folders: Infrastructure, Cloud & Migration, Security & Compliance, Onboarding & Managed Services, and Managed Intelligence (AI projects you can deliver to your customers, like a Microsoft 365 Copilot Deployment or a Custom AI Assistant / Chatbot). Each comes with real phases, tasks, durations, estimated hours, owner roles, dependencies, and checklists — clone one, set a start date, and the schedule builds itself. Edit, rename, or delete them like any playbook.

One per-user price, everything included — the CRM, the project engine, every integration, and all of the AI, with no tiers, no per-module SKUs, and no AI surcharge. List price is $95 per user monthly, or $85 per user on the annual plan (billed yearly). Round 1 — the first 100 MSPs to join — lock the founding rate of $50 per user monthly ($45 annual) for as long as their subscription stays active. Every workspace has a 3-seat minimum and a 14-day free trial. Partner portal viewers are always free and never count toward your bill. See the pricing article for the full breakdown.

A member of your team — technicians, PMs, sellers, admins. Clients viewing their own projects in the partner portal are free and unlimited.

Every new workspace starts with 14 free days. A card is required up front — that's what starts the trial — but nothing is charged until day 15, and we email you the exact amount three days before. Cancel during the trial (a one-line reason is all we ask) and you're never charged. A workspace set up without that self-serve card step gets a shorter 7-day setup window, room for up to 3 members, then goes read-only until a plan is chosen — with the banner saying exactly that ("choose a plan to continue"), never a payment-overdue message it never earned.

You buy an allocation — a pool of seats, minimum 3 — and every active team member draws one. Partner portal viewers are free, unlimited, and never counted; disabled users don't count either. Adding seats charges a prorated amount immediately and co-terms to your renewal; lowering the allocation takes effect at renewal, never mid-term.

Nothing dramatic, and nothing is deleted. The card is retried for about 7 days with an email at each step. Then the workspace goes read-only — everyone can still sign in and look, nothing can change. After 7 more days it pauses entirely except the Billing page. Fix the card or pay the open invoice on Stripe's secure page and access returns instantly.

No — annual terms are never refunded mid-term. Cancelling turns auto-renew off and you keep full access through the end of the year you paid for, so you always get everything you bought. If refund flexibility matters more than the annual discount, the monthly plan is the one to be on.

An afternoon, not a quarter. Connect your PSA (ConnectWise, Autotask, or HaloPSA) and Microsoft 365 and your companies, contacts, projects, and history are already there. Write-back stays off until you deliberately turn it on.

Every SBR claim carries evidence chips that resolve to a real record. During generation each reference is checked against its source; anything unresolvable is dropped before it reaches the page.

Your own other customers of a similar size — and only ever as anonymous medians. The magazine pack gives the comparison two pages: service (median first response, after-hours share, critical-priority share, positive-survey rate) and fleet & AI (the fleet measures plus Copilot adoption), each drawn as dot-plots of the partner against the cohort median. A real peer comparison renders only when at least five similar-sized customers exist; every row states how many peers reported that measure, and a measure too few peers have data for stays off the page. No peer is ever named, and only the aggregate medians ever leave the database. With a smaller book, the pages switch to an honest fallback — the partner against their own last quarter — optionally alongside clearly-labeled general industry framing that is never dressed up as a measurement from their record.

There weren't enough similar-sized peers for a real comparison — a peer cohort needs at least five of your other customers, because below that a median is neither anonymous nor meaningful. The magazine doesn't fake one, but it doesn't just vanish either: the comparison pages fall back to "you vs. your own last quarter" trend rows, optionally with clearly-labeled industry framing. A comparison page disappears entirely only when there's nothing honest to put on it at all — no cohort, no self-trend, and no framing.

Yes — the Refresh data button on a built pack re-pulls the live numbers while keeping your picks, edited text, custom sections, and conversation choices (trend cuts recalculate). What clears edits is re-running the wizard's review step: that rewrites the text, so old overrides are removed rather than silently masking words nobody re-read.

No — by design. Any sentence is click-to-edit on a built pack (with a revert), but counts, percentages, dollar values, the scorecard, and the charts are computed from the record and can't be typed over, so a pack can never contradict its own data. If a figure looks wrong, fix the source data and hit Refresh data.

Yes — the poll interval is configurable per workspace under your PSA's connector settings (ConnectWise PSA, Autotask, or HaloPSA). Default is every 5 minutes. Pushes are immediate regardless.

The chain of dependent tasks with zero slack — if any task on it slips, the whole project's end date slips with it. Current computes it live from your dependency graph and highlights it on the Timeline and dependency map. See the critical-path article for how to read and shorten it.

Both can sync from Autotask or be entered manually; on ConnectWise workspaces both are entered manually in this release, and they drive everything identically. Holidays: one company-wide calendar, filled by Autotask sync plus any admin-added dates, drives all scheduling. Employee costs: pulled from the linked Autotask resource's internal cost (locked from editing), or entered by a Tenant Admin — and used only for the admin-only Profitability lens.

System mail (sign-in codes, invites, digests, alerts) is transactional and sends through Current's own authenticated domain so it lands instantly. Sales sequences are different — they always send from your own Microsoft 365 mailboxes, protecting your domain's reputation and keeping replies in your inbox. Campaigns send through your own verified sending domain (set up in Integrations → Email), with the "Who it's from" choices deciding whose name — and, when it's on that domain, whose real address — each email carries.

Yes — sales leadership and tenant admins can, in the composer's "Who it's from" section (this is a leadership capability, enforced at the database, so account executives keep their two options: your brand, or "you from your brand"). "Each account manager" sends every recipient's email from that company's own account manager — from their real address when it's on your verified sending domain, otherwise replies still route back to them, and any company without an account manager falls back to your brand. "Someone else" signs the whole campaign as one chosen person — pick a teammate or type a name and reply-to. The Review step shows exactly who each recipient will hear from before you send. Three sign-off tags — {{sender_name}}, {{sender_first_name}}, {{sender_email}} — fill in per recipient to match whoever the email is from.

As many as you like — enrollment isn't capped. Sending is: each Microsoft 365 mailbox sends up to 250 sequence emails a day (a hard limit that protects your sender reputation), so a big enrollment simply paces out over several business days, about 250 at a time, with the rest waiting safely in the queue. Current warns you before a large enrollment and shows how many are waiting for the next window. For a very large list — 500 or more — the warning gets stronger: because sequence email comes from your own Microsoft 365 mailbox, a list with lots of bad addresses, bounces, or spam complaints can hurt your domain's sender reputation, so verify (Enrich) the list first to strip bad addresses, and for a true one-to-many blast consider a Campaign instead of a sequence — campaigns are built for bulk outreach with unsubscribe handling. And enrolling someone into a sequence that has no email steps yet never marks them 'completed' — they show as 'Waiting for steps' until you add one.

Open the menu on its row — on Sales → Campaigns or Sales → Sequences — and choose Duplicate. You get a fresh copy called "Copy of «name»" and the original is left exactly as it was. A campaign copy carries the audience, sender, subject, and body but no recipients, no send history, no results, and no approval stamp: it is born a plain draft and walks the whole approval flow again from scratch, so copying is a shortcut through the writing, never through the review. A sequence copy carries every step in order and is created paused, so nothing sends until you switch it on. Enrollments, send logs, and auto-enroll triggers are never copied — a copy that inherited a trigger would quietly start enrolling everyone the original was set up to catch.

Open the menu on its row and choose "Save as template." Current files a copy under the Templates tab on that page, and the campaign or sequence you saved from is untouched and keeps doing whatever it was doing. To use one, click "Use template" on the template row, or pick "Start from a template" when you click New campaign or New sequence — that option appears once your workspace has saved at least one. A sequence template is always paused, holds no enrollments, and carries no triggers, so it can never send anything itself; use it to create a sequence, then set up enrollment on the sequence you created. Deleting a template only removes the saved starting point — anything already built from it keeps running.

Use the Company status filter in the audience builder — set it to Active. It reads your PSA's own active/inactive flag (Autotask, ConnectWise, or HaloPSA), so a service notice reaches only companies that are still active partners and skips ones you've offboarded — even a former company that still carries a lingering contract, which the Customers filter alone wouldn't catch. Combine Active with the Customers filter to reach current partners who also hold an active contract. One caveat: a company's status comes from your PSA on the next sync, so a company you marked inactive in your PSA becomes excludable once that sync runs; until then it still counts as active, so no one is dropped by surprise.

About 3–5 minutes per 1,000 addresses that actually need checking — and the campaign wizard states a computed estimate for your exact list size before you start. Small batches (ten addresses or fewer) finish in seconds. Addresses verified within the last six months are skipped automatically, so re-verifying a recently-checked list is usually much faster and costs nothing. When it's done, the wizard's pre-send report shows every address with its verdict — valid, accepts-all, unknown, or flagged — with suggested corrections, and flagged addresses are excluded from every send automatically.

No — by design. In the Master unsubscribe manager (Integrations → Email/Resend) you can remove an address you added by hand, but an address that landed on the suppression list from a real unsubscribe click, a bounce, a spam complaint, or a verified-invalid check is locked and can't be removed. If someone opted out, they have to opt back in themselves.

The Relationship Health tile on Sales Home ranks your customers on one churn score, computed live from your own data: survey results, the AI's read of the account, quiet-account detection, how long since anyone made contact, contracts coming due, weak renewal deals, and stalled deals — weighted by how big the account is. The AI explains a row; it never picks one. An account Current hasn't read yet is shown as unknown, never as healthy. Sales managers see the same ranking across every rep.

Current writes the to-do for you, syncs it to your Microsoft To Do, and logs an internal note on the company's timeline that partners can't see. The account then steps aside for 21 days so the next-riskiest one can surface. You get five seconds to undo before anything is created.

Yes — group reps into a sales team in Users & roles → Teams (a manager plus their reps). That manager's sales dashboard, reports, and deal views then open focused on their team by default, with "Everyone" one click away. It's a view default, not a permission change: leadership access is unchanged, and a manager without a team keeps seeing the whole CRM. A person can be on one sales team and one project team, but not two of the same kind — so it's always clear whose team a rep is on.

Because it has no close date, and every report that counts by month counts by close date. It happens when the system that marked it won sent no date with it, usually an import or a PSA sync, and Current would rather leave the date empty than invent one. Those deals are the second group on Deal review, at /sales/deal-review: "These closed, but we do not know when." That page is not listed in the Sales menu, so open it directly. Each row shows the best date on record and says where it came from. If HubSpot recorded a real close date you can accept it in one click; if the only date on record is the moment a sync touched the deal, the field starts empty and you type the day it really closed. Setting a date there changes only the date. The outcome and the stage stay exactly as they are, and until it has one the deal counts toward nothing: not won revenue, not goals, not commission. Sales leadership only, because the date you pick decides which month the money lands in.

Because Current can only prove when it closed for some deals, and it will not guess on the rest. If a deal was won months ago and only reaches the Won stage today, closing it with today's date would put commission in this month for a win that belongs in another one, and nobody would have asked for that. So Current closes the ones it can stand behind (a deal your team created and closed this month, a loss, or a deal that already carries a real close date) and sends the rest to Deal review for a person. Two things follow from that. Your open pipeline and forecast can read slightly higher than you expect, because those deals stay open until somebody works the list. And the list is short work: Deal review can accept every trustworthy date in one action, and only leaves you the ones where the date on record is an import stamp rather than a real close.

No. A one-time close-out was planned for 27 July 2026 and it never ran. The queue emptied because the everyday sync closed those deals as it caught up with them, and Current records each of those closes as coming from the sync. Commission only starts tracking when a deal is closed in Current, so a close a sync makes pays nobody, while a rep closing a deal in Current, or sales leadership closing one on Deal review, pays exactly as it always did.

No. Commission only starts tracking when a deal is won in Current, and a deal that arrives already closed was won somewhere else. That covers the HubSpot import, the HubSpot two-way sync, and the Autotask, ConnectWise and Halo opportunity syncs: when one of them closes a deal, Current records that the close came from the sync, and no commission line can ever be created for it. The deal still counts in your pipeline and your won-revenue reports, so nothing goes missing from your numbers. Won quotes are the deliberate exception and always pay in full — a ScalePad or ConnectWise CPQ quote your rep built, sent and won is a win made in Current. And if your team closed a deal in Current first, a later sync leaves it alone: the rep keeps that commission no matter how many times the deal syncs afterwards.

It is still there, at /sales/deal-review, and nothing about the page changed. The row left the Sales menu on 27 July 2026 because the queue it was built for is empty on a normal day, and a permanent menu entry for an empty list is clutter. Deals still land on that page whenever a sync marks one won without saying when, so open it directly if a report tells you a win is missing a close date. Sales leadership only, because the date you pick there decides which month the money lands in.

Only sales leadership — a tenant admin, anyone with the Sales Manager role, or anyone with the sales-manager toggle. The rep never sees them, and that's enforced at the database layer, not just hidden in the interface. A rep sees only their own sessions' shared content: the shared notes, the outcomes, and the promises. See the Sales Coaching article for how sessions, promises, and the Your Day coaching tile work.

Yes. A promise made in a 1-on-1 lands on its owner's My Day and Microsoft To Do the moment it's saved — owner-scoped, so a rep's promises go to the rep's lists and a leader's to the leader's. The next steps you type in the coaching pipeline review sync the same way. Open promises also roll forward into the next session automatically, and kept-or-missed outcomes feed the follow-through trend on the coaching roster.

No. Your assigned project tasks appear in a "Current Projects" list in Microsoft To Do as a personal, one-way view: marking a task done in Current (or your PSA) checks it off there automatically, but checking one off in To Do never completes the real task and never flows a completion to Autotask, ConnectWise, or HaloPSA. If you check off a task that's still open in Current, it comes back titled "(Still open in Current — see notes)" with instructions to finish it — mark it done in Current, or extend its due date there. Deleting a To Do item just stops it mirroring for you. (Your personal CRM to-dos in the separate "Current" list still sync both ways.)

Because it's a snapshot, on purpose. When a coaching session is finalized, its goal panel freezes as "Goals as of" the session's date — the targets and actuals exactly as they stood the day you met, not as they've moved since. That keeps the record honest: what you discussed reflects what was true then. Only the live draft session reads today's numbers.

Each offering owns a short list of match phrases, and a customer is marked covered when any of their live PSA service lines — Autotask services, ConnectWise agreement additions, or HaloPSA contract lines — matches any phrase: contains (anywhere in the name), prefix (starts with), or exact, all case-insensitive. The phrases match the service name as your PSA writes it, which is usually namespaced ("Backup - Datto BCDR - S5-6…"), so one short phrase like "datto bcdr" groups a whole product family. Coverage is only ever computed for customers, so a prospect never shows as covered.

Sales leadership can, from the Manage offerings panel on the whitespace page. Every workspace starts with a default catalog of 29 common MSP offerings, each pre-loaded with generic starter keywords, so the grid is useful on day one — your job is to refine them to match how your PSA names things. A short note at the top of the panel spells out exactly what your keywords match against for your PSA (the Autotask Service name, the ConnectWise addition's product/description, or the HaloPSA line description). As you type a phrase it shows how many of your real PSA service names it would match, so you can tell at a glance whether it's too broad or matches nothing. Add or remove a phrase and that column is rebuilt against your data within a few seconds — the grid updates itself, with the overnight sync as a backstop. A phrase can belong to more than one offering on purpose, so a bundle or a two-in-one product ticks every box it covers.

It counts the support tickets and synced emails from the last 90 days that were actually about that offering — even though the customer doesn't buy it. Twice a week (the Monday and Wednesday insights run) Current's AI reads each customer's recent tickets and emails and works out which offering each one is about, by meaning rather than keywords, so "can't get my 2FA code", "set up MFA for new hires" and "authenticator app broken" all count toward the same multi-factor offering. It reads up to the first 2,500 characters of each part of an item — a ticket's title and description, an email's subject and body — for enough context to judge it accurately. It reflects what the customer said, not what you said back: only inbound emails count, with the quoted reply thread and signatures stripped first, so a topic your engineer raised in a reply is never counted as the customer's. Hover the circle for the breakdown — "2 tickets and 6 emails mentioned this in the last 90 days" — and click it to open the drawer, where each ticket and email expands ("Show more") to the full message the AI read, then start an expansion deal. Only offerings a customer doesn't already buy light up, and only the last 90 days count. Brand-new signals appear after the next twice-weekly pass rather than instantly.

In the magazine pack, the AI's dollars live in the road-ahead chapter's AI-opportunity box, and the math is ticket-grounded only: the box appears when a category of the partner's own human-raised tickets is big enough to matter — at least 15% of their human tickets — and its paragraphs print the factors behind every figure so the arithmetic is checkable. When no category clears that bar, the box drops entirely; there is deliberately no generic AI pitch, no assumed adoption rate, and no invented headcount.

Whatever your own catalog says. The required Email seats offering (Sales → Whitespace → Manage offerings) holds License pools rules — includes and excludes — that decide which of your ScalePad SaaS license pools count, and that definition drives the pack's seat figure, the AI per-person math, and the peer-comparison size bands. Out of the box the rules cover Microsoft's mailbox-bearing base licenses (E3/E5, Business plans, Exchange Online), and add-ons are excluded so nobody double-counts: a person on E3 plus Teams Phone is one seat. An MSP on Google Workspace just edits the rules — no product update needed — and the seat label in the pack is the offering's own name.

Because it defines what a managed email seat means everywhere Current states one — the SBR's seat figure, the AI at-scale math, and the peer-comparison size bands. Deleting it would leave those numbers with no definition, so it wears a lock badge (Required · defines email seats) and deletion is refused at the database layer, not just in the UI. Everything else is yours: rename it (the seat label follows), deactivate its whitespace column, or rewrite its rules entirely.

No. The Datto RMM connector is read-only: every request goes through a guard that only allows specific GET reads (sites, devices, open alerts). Resolving alerts, moving devices, running jobs, and resetting API keys are unreachable — Current reads Datto RMM and never writes to it.

Almost always because API access isn't switched on yet, or the platform is wrong. In Datto RMM go to Setup ▸ Global Settings ▸ Access Control and Enable API Access, then generate the Access + Secret key under Setup ▸ Users, and make sure the platform you pick in Current (pinotage, merlot, concord, vidal, zinfandel, or syrah) matches the region you log into.

Each Datto site (identified by its siteUid) is auto-matched to a Current company by normalized name and linked only when exactly one company matches. Unmatched sites appear on the card's mapping panel to link by hand; a manual mapping is never overwritten. Unmapped sites still sync — the data just doesn't surface on a company until it's mapped.

No. The Datto Backup connector is read-only: it calls only GET (read) endpoints — the device list, each appliance's protected-systems list, the Direct-to-Cloud endpoint list, and the SaaS Protection customer roster with per-customer seats — and every request is checked against that allowlist in code before it's sent. Current never writes back to Datto, and the key-management actions (regenerate, deactivate, delete) live in the Partner Portal, not in Current.

In the Datto Partner Portal, go to Admin ▸ Integrations ▸ API Keys ▸ Create API Key. Datto issues a Public API Key and a Secret API Key — the Secret is shown once, behind Reveal, so copy it before leaving the page. It takes a Partner Portal administrator. Paste both into the Datto Backup card on Current's Integrations page. One pair reads appliances, SaaS Protection, and Direct-to-Cloud.

Datto identifies customers by company name, not a stable id, so an appliance surfaces on a company only once that name is mapped. Open the Datto Backup card, work the Unmatched list, and pick the Current company for it. Unmapped appliances still sync — they just don't appear on a company until linked — and a manual mapping is never overwritten by the auto-matcher.

Some Datto Partner-Portal API keys are product-scoped, so a key created for another integration may read appliances but not SaaS Protection or Direct-to-Cloud. When Current detects that, the Datto Backup card shows a note naming the fix: create a NEW API key from Admin ▸ Integrations ▸ API Keys, paste the new Public/Secret pair into the card, and press Test. Reconnecting is non-destructive — every appliance, agent, seat, and mapping already synced is kept.

No — never. The Datto SaaS Protection connector is read-only. The one write in the whole Datto Partner API is a bulk seat change (pause/unlicense seats), and it is walled off in Current's code: the request layer only allows the three read endpoints and throws on anything else, so the seat-change endpoint is unreachable. Connecting Datto to Current cannot change, pause, or remove a single seat.

In the Datto Partner Portal: Admin → Integrations → API Keys → Create API Key, as a Partner Portal administrator. You get a Public key and a Secret key; paste both into Integrations → Datto SaaS Protection → Manage in Current. It's the same key family as Datto BCDR, so one pair can power both connectors.

Seat and backup data only appears on a mapped company. Datto identifies each customer by a SaaS customer id; Current auto-matches by name where it's unambiguous, but the rest need mapping by hand on the Datto SaaS Protection card's Unmatched list. Once the customer is mapped, its coverage shows on the company record within a sync cycle.

No. Datto EDR's LoopBack API exposes full create/update/delete and response actions (isolate, scan) on most models, but the Datto EDR connector is read-only: every request goes through a guard that only allows a strict list of GET reads (endpoints, sites/organizations, alerts). Isolating a host, running a scan, or closing an alert is unreachable — the API allows those changes, Current's code physically cannot issue them.

Datto EDR API tokens expire exactly one year after they are created, with no auto-refresh, so the most common cause is simply that the token lapsed (a 401). Create a new token in Datto EDR under Admin ▸ Users & Tokens ▸ API Tokens, paste it into the card, and re-test the connection. Set a calendar reminder ~11 months out so the next one is replaced before it expires.

Your Datto EDR instance is a per-partner subdomain like https://yourcompany.infocyte.com — find it in the address bar when you're logged in. Current needs it with /api on the end (https://yourcompany.infocyte.com/api); the trailing /api is mandatory, and leaving it off makes authentication and company loading fail.

No. RocketCyber's Customer API is entirely read-only — it has no write endpoints at all — and on top of that every request Current makes goes through a guard that only allows four specific GET reads (accounts, agents, enabled apps, and incident counts). Current reads RocketCyber and never writes to it.

Per-customer value signals for your executive and CIO business reviews: the count of managed endpoints and enabled security apps, how many incidents are open right now by severity, and 90-day incident activity (raised, resolved, and the most recent incident). It stores rolled-up counts per customer, not a raw log of every incident.

Each RocketCyber account (identified by its accountId) is auto-matched to a Current company by normalized name and linked only when exactly one company matches. Unmatched accounts appear on the card's mapping panel to link by hand; a manual mapping is never overwritten. Unmapped accounts still sync — the data just doesn't surface on a company until it's mapped.

Yes — strictly. Current only reads from Auvik (device inventory, network summaries, and open-alert counts) and never writes back. Auvik's one write endpoint, dismissing an alert, isn't wired into Current and can't be reached; the connector's allowlist contains only three read paths.

Three things from your Auvik dashboard: your region/cluster code (us1, us2, eu1… — read it from the dashboard URL), an API username (an Auvik user's login — ideally a dedicated service account), and an API key generated for that user. Paste them into the Auvik card under Integrations and press Test connection. Best practice is a service-account user so the key survives staff turnover.

The customer's Auvik tenant probably isn't mapped to that Current company yet. Current auto-links Auvik tenants to companies by exact normalized name, and leaves the rest for you to map on the Auvik card. Unmapped tenants still sync in the background — their devices and summary are stored — but only appear on a company record once you link them.

No. The Addigy connector is read-only. It mirrors device inventory, macOS version, FileVault/encryption and compliance status for reporting only. MDM commands (lock, wipe, restart, key rotation), the managed-admin password-reveal endpoint, and all create/update/delete calls are blocked in code — they aren't on the connector's allowlist, so Current cannot call them at all.

No. Addigy locks an account for 24 hours if you exceed roughly 1,000 requests in 10 seconds. Current budgets far below that — it waits at least a second between every request (about 1% of the limit), walks only a small bounded number of pages per run, and never retries into a rate error. A large fleet fills in over a few of the scheduled 6-hour syncs instead of all at once.

Each Addigy organization (a child organization under your MSP org) is one end-customer. Current lists all of them by name and auto-matches each to a Current company — first by the Autotask account id Addigy stores on the organization's policy (an exact match, on Autotask workspaces; on ConnectWise the name match carries the load), then by name, including lightly decorated policy names: a suffix like "Northwind Traders - Production", a site prefix, an environment tag, or an encoded character all pair to the right company as long as exactly one company fits. Anything genuinely ambiguous is listed as unmapped on the Addigy card for you to pick the company by hand. Mappings you set by hand are never overwritten by the auto-matcher, and unmapped devices still sync — they attach within seconds of you mapping the customer.

Most likely your Addigy account runs everything under one master organization and keeps customers as policies instead of child organizations — a common single-org setup. Current lists each policy as a customer in the mapping drawer, exactly like a child organization (hybrid setups with both work too). Check the Sync detail line on the Addigy connector card: it reports what the last sync found — how many child organizations and how many policies — and if your API token can't read policies, it shows the exact error so you can re-create the token with the right access.

Almost always because the match is ambiguous: two of your companies resolve to the same name — say "Acme" and "Acme Inc" — so auto-pairing would be a guess, and Current never guesses on mapping. Direct and lightly decorated names pair automatically; anything genuinely ambiguous waits on the Addigy card for you to pick by hand, on purpose. Map it once — a hand-set mapping is never overwritten, and its devices attach within seconds.

Because Inky has no pull API. Its only way to send data out is the Custom SIEM Feed, which POSTs events to a URL you provide. So Current hosts a receiver and gives you a secret feed URL to paste into Inky (Settings → Integrations → Custom SIEM Feed) — the reverse of every other connector, where you paste a key and Current reads the vendor.

No. Current keeps threat signals only — event type, verdict, threat level and category, timestamp, and the sender and recipient domains. Email bodies, HTML, subjects' contents, and attachments are stripped the instant an event arrives, before anything is written to the database, and are never stored.

Nothing yet — Awaiting feed means the receiver is enabled but Inky hasn't sent an event. Confirm the exact feed URL from the Current card is saved in Inky's Custom SIEM Feed with the JSON format, that any event filter isn't excluding everything, then use Inky's test-send. The card turns to Receiving within seconds of the first event.

Yes — completely. Unchecking a feed in the review's Data feeds panel doesn't just hide a chart: its numbers leave every page of the pack, its telemetry chapter drops out, and the AI never sees that data either — so no finding, callout, or recommended action can quietly reference something you removed. Re-check it and it comes back on the next build.

The Data feeds panel greys a feed you can't use and tells you which of three reasons applies: Not connected (set the integration up under Settings → Integrations), Connected but unmapped (the integration is on but this partner isn't matched to a company in it yet — map it), or Mapped but no data (everything's wired, but this partner genuinely has nothing there this period). Only the first two are things to fix.

Tenant admins and account executives / sales leadership. Branding lives in Settings → Branding (a logo and two brand colors), and the people who present reviews can edit it so a seller doesn't have to file a ticket to fix a logo before a QBR. Everyone else sees the branding applied but can't change it. It applies to the Strategic Business Review (the cover, the PDF, and Live mode), the partner portal header, and the emails Current sends on your behalf — campaigns, booking confirmations and reminders, partner project digests, invites, and partner notifications, which use your logo (favicon as backup) and brand colors on headers, buttons, and links.

Yes. Tenant-sent emails — campaigns, booking confirmations and reminders, partner project digests, invites, and partner notifications — carry a small, muted "Powered by Current" link at the bottom, on by default. A Tenant Admin can hide it with the "Show 'Powered by Current' on emails" toggle in Settings → Branding; flip it off and the line disappears from every branded email your workspace sends.

For readability. Brand colors are chosen for logos, not for pages of text, so a color that would make text hard to read (too little contrast against the surfaces it's used on) gets nudged just enough to clear a legibility bar — with a short note telling you it happened and why. It's a contrast guardrail, not a style opinion: a perfectly legible but unusual pick is left exactly as you set it.

No. Integration alerts are open-once and self-resolving: a rule opens one alert when it trips (not a new one every time the condition repeats), and the alert resolves itself automatically when the underlying problem clears — the backup succeeds, the detection is closed, the phishing surge subsides, compliance recovers. Nothing to dismiss by hand, and no stale alerts lingering after the fact. They're in-app only today, and outbound email/Teams notifications are off by default.

No. The three telemetry chapters read as themes, never vendor names — the partner sees "your security posture," "your fleet and network health," and "your backup and continuity picture," not the brands of your endpoint, monitoring, or backup tools. It keeps the review about their business, and it means changing a vendor never changes how the pack reads.

The same magazine the PDF prints, restaged for a room. Every chapter in the pack becomes one full-screen slide on a dark stage in your brand colors — the quarter on a page, the priced risk cards, the budgetable road-ahead table, the team close — drawn from the same frozen numbers and inline edits as the printed pack, so the deck can never disagree with the document you hand over. The honesty rules hold on screen too: a chapter with no data has no slide, exactly as it has no page, and nothing is padded with a zero. Move with the arrow keys or by clicking either side of the screen; a chapter rail along the bottom shows where you are, and clicking any chapter jumps straight to it. Reviews built before the magazine redesign present in the same deck.

Everywhere it makes a review sound like yours. The tagline lands on the cover, the founding story and core values in a "how we work" band, your three uniques on the AI page, your backup methodology at the top of the backup chapter, and your account manager's sign-off on the closing team page — in both the printed PDF and the matching Live slides. The AI that drafts the review is handed your story too, so it can reference your values and uniques by name instead of inventing generic ones. Set it once in Settings → Branding → Your story (tenant admins and account executives / sales leadership can edit it); leave it blank and Current uses neutral fallbacks rather than making anything up.

From records, with the arithmetic visible. A priced risk card shows where its price came from: a price from a real quote wears a QUOTED chip, an account manager's figure wears ESTIMATE, and where no honest number exists the column reads "By quote" or "Included" — never a fabricated dollar amount. The value ledger's dollars are computed from your workspace's own constants (Settings → Branding): minutes per monitoring catch, per quarantined phish, and per outage-class save, plus a downtime rate — with the factors printed beside every line. And the road ahead's total sums one-time and per-year costs separately, so nothing is inflated by mixing the two. Your internal costs and margins are never part of any of it.

Yes — Current partners get 10% off Handwrytten with the code CURRENT. Enter it in the "SignUp Code (Optional)" field on Handwrytten's sign-up form (the last box before the Terms and Conditions checkbox) before you press Sign Up. It's easy to miss because it's marked optional. Sign up at https://www.handwrytten.com?ref=otfimjl — the Handwrytten article shows exactly where the field is.

No. Test mode is on by default, and while it's on cards are accepted and processed by Handwrytten but never printed, mailed, or charged — so you can try the whole flow for free. Only when a Tenant Admin flips the Sending mode switch off test mode do real cards actually mail and bill to your Handwrytten account. See the Handwrytten article for setup and going live.

Yes. As soon as a lead arrives — from a website form or typed in with the New lead button — Current creates its company and its contact in your PSA and marks the company a Prospect, so your team finds the account where they already work. Current matches before it creates: if the lead's email domain already belongs to a company you have, the lead is linked to that company and nothing new is pushed, because an account that already lives in your PSA should never be re-created or overwritten. The contact is still created under it, so a new person at an existing company lands properly. Two things to know. This follows the CRM write switch in Settings → Integrations: while CRM writes are off nothing is pushed, and the company and contact wait in Current marked pending until writes are switched on. And on HaloPSA the client is created without a type, because Halo has no prospect field Current can set safely — the push result says so plainly rather than implying the mark was made. Leads imported from a saved list are never pushed, since those people and their companies are already in your CRM.

On Sales → Leads, click "Import from list," pick any list from Sales → Lists, and choose some or all of its people. They land on the same leads board under the Lists source filter, in the first open stage, alongside your website and manual leads. Anyone who already has an open lead is shown pre-unticked and labelled before you commit, and the result tells you what happened — for example "12 imported, 3 skipped (already in the funnel)" — so nobody is skipped quietly. A lead you converted or discarded earlier is finished business and doesn't block a fresh import of the same person, and a dynamic list resolves its filter at import time so you get who is on it now. Imported leads arrive unrouted with no response timer, because they're people your team chose to work rather than an enquiry waiting on a reply; the clock starts when someone routes one. Nothing is pushed to your PSA either — these contacts and their companies are already there.

Into the list view on Sales → Leads, not out of Current. The funnel no longer keeps Converted and Disqualified columns — it has an outcome rail beside the board instead, and a lead you drop on "Drop to disqualify" leaves the board so the funnel only shows leads you can still work. Switch Funnel to List and two chips split them by age: Discarded holds the last 30 days, and Archived holds the older ones. Both are equally restorable. Nothing is ever deleted, and there's no cleanup task to run. The quickest route is the quiet caption under the grey zone on the funnel — "5 discarded · View" — which opens the list with that filter already applied. The green zone has the matching "3 converted · View".

Open Sales → Leads, switch to the List view, and pick the Discarded chip for a lead disqualified in the last 30 days or the Archived chip for an older one. On the row, Restore replaces the Convert and Disqualify buttons — click it and the lead goes back to the top of the funnel, ready to work again. You can also do it from the lead's drawer: click the row to open it and Restore sits in the action bar at the bottom. Age changes nothing about whether you can restore a lead; the 30-day line only decides which chip it sits behind.

No. A note you add in the lead drawer stays in Current, on that lead, and nowhere else. It doesn't post to the company timeline, it doesn't show on the company record, and it never reaches Autotask, ConnectWise, or HaloPSA. That's deliberate: qualifying notes are working scratch, and notes about a lead nobody has qualified yet shouldn't land on an account your whole team reads. Once the lead converts, write what matters on the company or the deal, where it belongs. Adding a note does move the lead's Last touch, so a worked lead rises to the top of the list.

Click a column header. On Sales → Leads, switch Funnel to List and every header sorts: Lead, Company, Status, Source, Rep, Age, and Last touch — click once to sort, click again to flip the direction. It opens on Last touch, newest first, so whatever you worked most recently is at the top. Last touch moves when the lead's status changes, when it moves stage, when its rep changes, or when someone adds a note. Age sorts on when the lead arrived. Both sort on the real timestamps rather than the words on screen, so "today" and "3d" order the way you'd expect.

Because the integration isn't connected. Two columns — Handwrytten and Follow up — appear on Sales → Leads (between Qualified and Meeting set) only once a Tenant Admin connects your Handwrytten account in Integrations → Sales & marketing. If it's connected and they still aren't there, the card service may not be deployed in your environment yet — the connection card will say so. The Send a card button on company records follows the same rule. Don't have a Handwrytten account yet? Sign up at https://www.handwrytten.com?ref=otfimjl and use the code CURRENT for 10% off, then connect it. Full details in the Handwrytten article.

A lead you sent from moves itself into the Follow up column, and Current creates a to-do for you due about five business days later — roughly when the card should have landed — that rides your Microsoft To Do sync like your other to-dos. The card is also stamped on the lead and logged on the company timeline. Cards sent from a company record do the same (the to-do and the timeline note) with no funnel column involved.

About $3.75 plus postage per live card, or whatever your Handwrytten plan pricing is — billed directly to your Handwrytten account, not through Current. The send dialog shows the price before you send, and in test mode nothing is charged at all. A Tenant Admin can also set an optional monthly card cap. See the Handwrytten article for the details.

The ones that hold real customer work — and not the four kinds of machine noise: alerts and monitoring, RMM automation, backup and BDR jobs, and spam or internal-IT chatter. Whether they're Autotask queues, ConnectWise service boards, or HaloPSA ticket types, Current highlights those with a "Recommended off" badge, but it only ever recommends — it never disables one for you. Muting a queue or board hides its tickets immediately so they stop distorting support history, sentiment, SBR stats, and AI insights; nothing is deleted in Current or touched in your PSA. Flip it back on and the tickets Current already synced reappear instantly, with new activity syncing again right away — tickets created while it was muted only return after an admin with CRM access runs a full ticket re-sync. You set this in the setup wizard's Mapping & noise step or later in your PSA's Manage drawer under Integrations; tenant admins can manage it with no CRM access. See the ticket-queue-noise article.

When you convert a won quote, the Win Quote wizard counts the labor hours off the quote's lines automatically into a "Professional-services hours sold" figure (each charge line has a Counts as sold hours checkbox, labor ticked by default), and the AI drafting the plan is told that number and your working-day window so it fits. A banner then compares the plan's task hours to what you sold, and Scale to match lines them up in one click — quarter-hour steps, never below 0.25 hours on a task, milestones untouched. It's guidance, not a lock: the project manager always has the final say.

Yes — as a first-class PSA, at the same depth as Autotask. ConnectWise PSA syncs two-way (companies, contacts, agreements, opportunities, projects, project tickets, members, and time entries), ConnectWise RMM feeds device inventory and patch posture into your business reviews, and ConnectWise CPQ won quotes flow through the same win-quote engine as ScalePad Quoter. You pick your PSA in the setup wizard, or connect it later from Integrations.

ConnectWise PSA, Autotask, and HaloPSA — one per workspace, chosen at signup and switchable later from Integrations. Current is not a PSA and does not replace yours: it expects one, and your PSA stays the system of record for tickets, contracts, and billing. If you don't run a PSA yet you can still start, but the PSA-driven features like billing sync and the whitespace grid wait until you connect one.

Yes — as a first-class PSA, at the same depth as Autotask and ConnectWise. HaloPSA syncs two-way (clients, contacts, opportunities, projects, milestones, tickets/tasks, contracts, and time), Halo stays the source of truth for time and billing while Current owns scheduling, and won Halo quotes flow through the same win-quote engine as ScalePad Quoter. HaloPSA support was added in July 2026. You pick your PSA in the setup wizard, or connect it later from Integrations.

No — one PSA per workspace (ConnectWise, Autotask, or HaloPSA), enforced at the database layer, not just the interface. Every synced record carries exactly one external identity, so Current never has to guess which system owns a company, project, or time entry. Switching is deliberate: disconnect the current PSA (a typed confirmation — nothing already pulled is deleted, and its synced projects freeze safely), then connect another from its own card.

No. The ConnectWise RMM connector is read-only twice over: the API credential is created in Asio with read-only scopes, and every request Current makes passes through a guard that only allows specific read endpoints. Automations, scripts, policy changes, and credential management are unreachable — Current reads sites, devices, and patch posture, and can never change anything in your RMM.

By authenticated polling, not a webhook — ConnectWise's callbacks carry no signature Current could verify, so Current polls ConnectWise every few minutes for won quotes and opportunities sitting at your configured Ready status-and-stage pair. Wins land in the same "Won quotes — ready to convert" queue as ScalePad Quoter wins, and the same wizard turns each into a project or post-sale ticket with charges in ConnectWise.

In Halo, sign in as an administrator and go to Configuration → Integrations → Halo API → Authorise a new application, choose "Client ID and Secret (Services)", note the Client ID and Client Secret, and grant the scope (Halo's own advice is "all" if unsure). Then in Current, open Settings → Integrations → HaloPSA, paste your Halo instance URL plus the Client ID and Secret, and run the test connection. Current works out your Halo auth and resource server addresses from the instance URL, and supports both hosted (yourcompany.halopsa.com) and on-premises instances. See the connect-halopsa article for the full walkthrough.

Clients become CRM companies (with Halo's native account manager filling the account-manager field), Halo Users become contacts and Agents become your team, opportunities become deals, and Halo projects — tickets with project features on — become Halo-synced projects with their milestones as phases and their child tickets or to-do items as tasks. Halo stays the source of truth for time and billing; Current owns scheduling (task start/end dates, criticality, and milestones), which it pushes back. Time logged in Current becomes a Halo action for billing, holidays and agent leave feed the scheduler, and client contracts drive MRR. See the HaloPSA field-mapping reference for the field-by-field detail.

Yes. Halo logs time as an action on a ticket, and once an action is on an invoice Current treats it as final and never edits or deletes it. Because logging billable time against a prepaid (Pre-Pay) contract draws down the client's balance in Halo, Current never re-sends a time entry blindly — every push is claimed once and confirmed by Halo's own response before it's considered done, so a retry can't create a duplicate charge or over-draw prepaid hours.

By authenticated polling — Halo's callbacks carry no signature Current could verify, and Halo has no "changed since" filter on quotes, so Current polls Halo for quotes that reached your "Accepted" quote status (the won bucket) and works out what's new by comparison. Wins land in the same "Won quotes — ready to convert" queue as ScalePad Quoter and ConnectWise CPQ. And yes — quoting is now PSA-agnostic: a won ScalePad Quoter quote converts into whichever PSA your workspace runs (Autotask, ConnectWise, or HaloPSA), building the project and pushing charges into that PSA.

Yes — Halo client contracts are read-only inbound, powering each company's MRR, customer status, and the whitespace grid. A contract's billing period (monthly, quarterly, six-month, or annual) is normalized to a true monthly figure. Halo stores that period as a numeric code whose meaning varies by version, so Current reads the human-readable period label; an unfamiliar period is left blank (which safely defaults to monthly) rather than divided by the wrong number, so MRR is never written wrong.

Task dates and status, task comments (as Halo actions), time entries, CRM company/contact/note/deal writes, and won-quote conversion (project build plus charges) all flow back to Halo — but never silently. Three gates stand in front: only HaloPSA-synced projects push at all (a Current-native project makes no Halo calls); the two-way sync switch controls task-field pushes; and CRM write mode (off by default, with a test mode that only touches "ZZ-Current Test" companies — older "ZZ-Cascade Test" names still count) controls every CRM write. Contracts, invoices, and the product catalog are read-only in every direction. See "What Current writes back to HaloPSA (and when)" for the full list.

Halo has no separate update call — you POST a record, and Halo creates or updates it based on whether the record carries its id. So every update Current sends includes the mirrored Halo id it stored on the first create or pull, which makes Halo update the existing record instead of quietly making a second copy; a create returns the new id and Current stores it immediately, so even a retry can't duplicate. For nested lists (a project's milestones, a ticket's to-do checklist, a Workday's holidays), Current reads the current list, merges your change onto it, and writes the whole list back — so items you didn't touch never disappear.

Yes — after every push Current reads the record back from Halo and compares what landed against what it sent. A value Halo accepted but quietly adjusted (a clamped date, a defaulted field) is reported as a residual rather than swallowed, and a failed push surfaces Halo's real error message, never a generic "sync failed." On time specifically, an invoiced Halo action is treated as final and is never edited, deleted, or retried.

A couple. HaloPSA is Current's newest PSA, so the post-sale ticket-path options (the queue/status/procurement picklists) and the ticket-sync diagnostic aren't wired for Halo yet, and Autotask project templates build only on an Autotask workspace. In each case Current shows an honest "not available for HaloPSA yet" message instead of a fake error — quote conversion into projects, charges, time, and the CRM writes all work now, and the deferred pieces light up as they ship.

As soon as it's won. The Commissions module reads your won deals automatically, so a deal you close lands on your next monthly statement with no data entry — it's assembled for you from your won deals, linked quotes, and goal attainment. The line starts near the bottom of the recognition ladder (Quote won) and climbs as the work becomes real. When it becomes fully Earned depends on your plan's trigger — on quote-won, on kickoff or a post-sale ticket (the default), or on invoice. So a deal can appear on your statement immediately while its line still reads Pending until the trigger is met.

Because your plan earns on invoice, and Current hasn't yet seen the deal invoiced in your PSA. Current checks Autotask, ConnectWise, or HaloPSA and moves the line to Earned once the deal is billed — you don't have to do anything. Some deal-and-reference combinations can't be matched to an invoice automatically; those stay Pending invoice honestly rather than being guessed, and a sales leader can mark the line invoiced with one click once they've confirmed the billing. A "Pending kickoff" line is the same idea for plans that earn on kickoff: it waits for the kickoff or post-sale ticket.

No — once you submit, you can't pull it back or change it yourself. That's on purpose: a submitted statement is one you stood behind. If something needs fixing, your manager can return the statement to you with comments, and only then does it reopen for you to edit and resubmit. This keeps the review clean on both sides.

Someone else. Current will never let anyone approve their own statement — a sales manager's statement has to be approved by another sales leader or a tenant admin. It's a hard rule enforced in the database, not a setting, and it's the segregation of duties that keeps commission honest.

Every earned line carries a 6-month clawback watch from the deal's close date — a heads-up that the commission is still inside the window where a customer could cancel or refund. Current doesn't automatically reverse your pay behind your back; if a cancellation happens, a sales leader adds a clearly-labeled negative adjustment line on a later statement, with a note explaining it. After 6 months the money is settled and the watch clears.

A Margin % component pays a percent of the gross margin — quote sell price minus quote cost — so it needs a linked quote to read those numbers from. If the deal has no quote linked, Current can't calculate the line and leaves it for your manager rather than guessing an amount. Link the quote to the deal and the line fills in on the next assembly. A TCV % line works the same way when a recurring deal has no term set — link the quote and set the term.

A sales leader can split a line between two reps — either by percentage (the shares must add up to 100%) or as a flat dollar amount off the top, where a fixed amount goes to the assisting rep and the owner keeps the remainder. The split shows on both reps' statements, clearly attributed, so everyone sees their share and the totals reconcile. Like every manager edit, a split requires a disclosed note both people can see, and an edited line carries a MODIFIED badge — there are no silent changes to anyone's pay.

When you ask your manager (or nudge a rep) about a commission line, the message goes out as a real email from your own mailbox until your organization turns on Teams messaging — then the same asks arrive as genuine Teams 1:1 direct messages. The upgrade is a one-time Microsoft admin step: grant the Current app the Chat permission in Azure / Microsoft Entra ID and re-approve (grant admin consent) for the whole organization. After that, Current detects the new capability and switches from email to Teams automatically — nothing to change inside Current.

From your existing Sales Goals — the same "rock" your team already chases on the goals dashboard. Accelerators, goal bonuses, and President's Club all read those goals directly, so there's one quota number everywhere: change a goal in one place and every commission calculation that leans on it updates too. Commissions never keeps a separate, parallel set of targets.

Yes. President's Club is an optional leaderboard and it's off by default — sales leadership turns it on, and can turn it off any time, from the Club page. When it's off, reps still see the Club page but it shows a short teaching state explaining that leadership hasn't set it up yet, so the nav never dead-ends. Turning it off doesn't affect anyone's pay — President's Club is recognition, not commission.

Yes, by describing it. Dashboards is its own page in the left navigation, directly under Companies, and quick search finds it if you type dashboard or wallboard. It used to be the first tab on Reports, and an old link to that tab still brings you to the page. Click Add a tile and type the number you want to watch in plain words (tickets we closed this week, won revenue this quarter by rep, hours logged this week by person) and Current draws it as a gauge, a leaderboard, a trend, a donut, a bar, a table, or a single big number. Type your own request and it always asks three short questions about it, each with three suggested answers to click plus a box to type your own, and answering is optional on every one. Then it draws up to three finished tiles side by side with your real numbers so you pick the one you want, and a last step lets you set its name, shape, queues, target and colour bands with a live preview beside you. The starter chips skip the questions, because those are already worked out. One timeframe bar at the top drives every tile: Today, Week, Month, Quarter, Year, trailing 7, 30, or 90 days, or any custom range. Every tile has a Refine with AI action, so you change it by saying what's wrong with it. See the ai-dashboards article.

Yes. There is a sun and moon control in the dashboards toolbar, on the right next to the dashboard name, and it switches that page between light and dark. It changes the dashboards page only: your sidebar, projects, the CRM and every other screen stay as they are. The choice follows you rather than the machine, so a board you set to dark on your laptop is dark when you next open it somewhere else. Wall screens are already dark without you doing anything: Present mode and TV links both open dark, since a bright white board in a dim office is the first thing people ask us to turn down. In Present mode the same sun and moon control sits in the bar at the bottom, so whoever is driving can switch that session to light without changing their own desk. A TV link has nobody standing at it, so add ?theme=light to the end of the link to keep that screen light; anything else on the end is ignored and the screen stays dark, so a typo can never leave a lobby TV blank. None of the numbers change with the theme, and colours were rechecked so arcs, bars and chart labels stay readable in both. See the ai-dashboards article.

Because by default Current decides that by size, and the tile got big enough to become a hero tile: a deep ink card with the figure in large type and its change and trend line underneath. Make the tile smaller again and it goes straight back to an ordinary card. If you would rather it stopped changing when you resize it, open the tile menu, choose Edit tile, and set Tile style to Standard card to keep it an ordinary card at any size, or Always hero to keep the dark look at any size. Hero tiles look the same whether the board is light or dark, and About this number still sits underneath them exactly as it does on every other tile. See the ai-dashboards article.

Yes. Open the tile menu, choose Edit tile, and look for Tile style. Automatic is the default and is what tiles do today: make one big and it wears the dark hero look, make it small and it goes back to an ordinary card. Always hero keeps the dark look whatever size the tile is, and Standard card keeps it an ordinary card however big you make it. Once you pick either of those two, resizing never changes the look again. It only changes which card the figure is drawn on: the number, the timeframe, the queues it counts and the explanation under it are all untouched. The setting appears on single-number tiles and gauges only, because a leaderboard, a table or a chart cannot be drawn on the dark band. On a very small tile Always hero falls back to an ordinary card, since the hero look needs room to read. Your choice also survives a Refine with AI, the same way your queue and timeframe choices do. See the ai-dashboards article.

It moved to its own page. Dashboards is now in the left navigation, directly under Companies, so a board you check every morning is one click away instead of two clicks inside a reporting page. Quick search finds it too: type dashboard, dashboards or wallboard. Nothing you built changed, and nothing was lost. Every board, tile, arrangement and TV link is exactly where you left it, and the page still remembers the timeframe and the dashboard you had open. Any link you saved to the old tab takes you straight to the new page. Reports now opens on Overview, with its seven lenses working exactly as before.

No, and it does not any more. A dashboard's arrangement is shared, but what each person can see is not: a tile whose numbers your role does not reach is hidden from you. Those hidden tiles now hold their slot, so you see an empty space where one sits, and nothing you drag can move it. The gap is deliberate. Closing it on your screen meant your own edits were then saved against the tidied-up version, so making one tile slightly taller could quietly re-order three tiles you had never touched, and simply opening and closing the board could rewrite the arrangement for everyone. A visible gap is a much smaller price than that. See the ai-dashboards article.

Below laptop width the board is one honest column rather than a twelve column grid, so there is no drag and there are no corner handles. Each tile carries its own controls underneath instead: a move up button, a move down button, and a height bar you drag or nudge with the arrow keys. Moving a tile there rearranges the whole board and keeps the widths you set on a laptop, so a tidy-up on a phone does not flatten your desk layout. On a laptop, grab the grip in the tile's top left corner to move it and pull any of the eight edge or corner handles to resize. See the ai-dashboards article.

You should not any more, and if you do it is worth telling us. Every window on a dashboard, on a TV board and in Reports is now cut in your workspace's time zone: today, this week, this month, this quarter all mean the same thing to everyone, whichever country they are sitting in and whichever clock their laptop is set to. Each tile prints that calendar under the number, beside the window it covers, so the two of you can compare the same words instead of guessing. The Reports header, every printed report and every TV board say it too. Your zone lives in Settings, under Integrations, in the Workspace basics section. It starts on the zone Current already used for notification timing and the lead response clock, so there is nothing for you to set. Some numbers did move the day this shipped, and here is exactly which: the forecast tile and the Reports Pipeline, Forecast, MRR movement and AE scorecard lenses used to cut their months and quarters at UTC midnight, so on any workspace outside UTC a deal that closed in the last hours of a month could sit in the wrong month. It now sits in yours. Totals in the middle of a window did not move.

Your workspace's, not your laptop's and not UTC. There is one zone for the whole workspace rather than one per person, deliberately: a number two people cannot compare is a number they can only argue about. So today, this week, this month and this quarter mean the same thing to everyone in the workspace, and to the TV board in the lobby, which used to count on the clock of whatever screen it was playing on. You can always see which calendar a number was counted on: every dashboard tile prints it under the number beside the window it covers, the Reports header and every printed or exported report carry it, and a TV board prints it along the bottom. It starts on the zone Current already used for notification timing and the lead response clock, so there is normally nothing to set, and a tenant admin can change it in Settings, then Integrations, then Workspace basics. Two limits, said plainly. Quarters are calendar quarters, so a financial year that does not start in January is not supported yet. And a sales goal's progress still measures its month, quarter or year at UTC edges, because the database side that produces those actuals does, and moving one half without the other would be a worse bug than the one it fixes: a deal closed in the last few hours of a month can therefore land on the next month's goal bar while Reports counts it in the month it closed. Which period the goals page treats as the current one does follow your zone.

Settings, then Integrations, then Workspace basics. Only a workspace admin can change it, the card shows you the date and time in the zone you have picked before you save, and every change is recorded in the audit log. Changing it re-cuts every window on a different midnight, so counts right at the start or end of a window will move and so will the daily points on a trend. Totals in the middle of a window do not move. A TV board that is already playing follows the new zone on its next refresh, which the card warns you about before you confirm. Quarters stay calendar quarters for now: if your financial year does not start in January, tell us and we will add it.

Because Current worked that number out itself instead of using a metric from its vetted catalog, and it will not pretend otherwise. The footnote reads "AI-computed metric. Verify this number against your source before relying on it." Check it against your PSA or the matching Reports lens for the same date range, then use Mark verified on the tile menu: the footnote then names you and the date you checked it. The asterisk stays on purpose, so anyone reading the dashboard later still knows the difference between a number an engineer wrote and a number the AI assembled. Catalog tiles have no asterisk at all.

Only if you publish it. Every dashboard starts personal, visible to you alone, and you can keep as many as you like. Project managers and tenant admins can also share one with the team, in which case it has a single layout that its editors control. Sharing never widens what anyone can see: a tile whose data sits behind a wall a viewer isn't through (cross-person hours for a non-manager, CRM numbers for someone without CRM access) simply isn't shown to them, and the database enforces that, not the interface. Partners never see Dashboards or Reports at all.

Connect the integration it named, then ask again. Current builds only from systems you actually have connected, so a satisfaction tile needs CrewHu, a phishing tile needs Inky, and a warranty tile needs ScalePad Lifecycle Manager. Rather than drawing an empty tile, it names the one it needs and links you to Settings → Integrations, where a tenant admin connects it. Give it one sync cycle to land (most reporting connectors refresh every 6 hours, your PSA every 5 minutes), and the sample prompts and everything Current will build widen on their own. This is different from Current telling you it does not hold a number yet: that one is not fixed by connecting anything, and Current says so rather than sending you shopping.

Because it would not be true, and Current would rather say so than draw it. There are three different reasons and the wording tells you which one you have hit. The number needs an integration nobody has connected yet, in which case Current names that integration and points you at Settings → Integrations. The number sits behind a wall your role is not through, such as cross-person hours or CRM figures, in which case it says that kindly and never builds a substitute. Or Current genuinely does not hold the number yet, whatever you connect: technician utilization, first contact resolution, escalation rate, quote win rate, blended gross margin, revenue per technician. In that last case Current gives you one plain sentence on what is missing and offers the nearest numbers it can really draw as buttons you can click, already checked against your permissions and your connected systems. Reactive hours per endpoint used to be in that last group and is not any more: Current mirrors your PSA's service ticket time now, so ask for RHEM and you get the real gauge. See the ai-dashboards article for the full list.

By default yes, and you can change that. Average first response and average resolution time can be measured on plain elapsed time or on the hours your desk actually works, and every workspace starts on elapsed time so nothing moved under you. Elapsed time is why an 8x5 desk looks bad against the 60 minute benchmark on that tile: a ticket that lands at 5pm on Friday and is answered at 9am on Monday scores 3,840 minutes, when on an eight to five desk it took one working hour. Every published response time figure in this industry, including that one, is measured on open hours. Set your desk's hours in Settings, then Integrations, then Workspace basics: when you open, when you close, which days you work, or that you work around the clock. Holidays come from the list you already keep under Holidays, the same one the project scheduler skips. Switching to working hours makes those two numbers smaller, usually two to three times smaller for an eight to five desk, so tiles that read red can turn green. That is the correction, not a break, and each of those tiles says under its number which of the two clocks it counted on, so you never have to guess. Nothing else changes: ticket counts, open ticket aging and the trend charts are not on this clock.

Almost certainly a tenant admin switched your workspace from plain elapsed time to your desk's working hours, in Settings, then Integrations, then Workspace basics. On working hours only the hours the desk is open count, so nights, weekends and the days on your Holidays list stop counting, and for an eight to five desk average first response and average resolution time usually read two to three times smaller. A ticket that arrived at 5pm on Friday and was answered at 9am on Monday now scores one working hour instead of 3,840 minutes. Tiles that read red can turn green, and that is the correction rather than a break: every published response time figure in this industry, including the 60 minute mark on the first response tile itself, is measured on open hours. You can tell which clock a tile is on without asking anyone, because both of those tiles say under their number which clock they counted on, in the app, in Present mode and on a TV board. Nothing else moved: ticket counts, open ticket aging, the opened versus closed trend, hours logged and SLA compliance are not on this clock. If you need to know exactly when and who, the change is in your audit log, and a workspace can be put back on elapsed time in the same place.

It sorts itself now, and rank 1 is your fastest responder. On a number where a lower figure is the better one there is only one honest order, so average first response and average resolution time always rank fastest first when you break them down by person, queue or partner, and the tile says so under its number. There is nothing to set, and asking Refine with AI to reverse it gets you a sentence saying the board already reads that way rather than a tile that quietly changes nothing. Two things came with it. Every row says how many tickets it is an average of, and a row built on fewer than five tickets sits below the ranked rows without a rank, because one overnight ticket out of three is a fact about a ticket rather than about a person. And a board showing eight of twenty five now tells you how many it is not showing.

Because until you set your desk's hours they were measuring with two different clocks, and both were reading their source correctly. SLA compliance is your PSA's own verdict on each ticket, worked out against its own due times, which already follow the partner's SLA calendar, their priority tiers and their exclusions. First response time was Current subtracting two timestamps, which counts nights and weekends. Set your desk's hours in Settings, Integrations, Workspace basics and the two tiles start telling the same story. Current never re-works out the SLA verdict itself, deliberately: if it did, your dashboard could disagree with the SLA report your partner gets, and then neither number is worth anything.

Because too few of your resolved tickets carried an SLA verdict for a percentage to mean anything. The tile now prints its own denominator, in the form based on 14 of 3,902 resolved tickets, and when fewer than twenty carried a verdict it hides the percentage and tells you so. A green 100 percent off three tickets is not a fact about your desk, and on a wall screen nobody can question it. Widen the timeframe, or check that your PSA is setting SLA targets on the tickets you expect. Only Autotask sends SLA verdicts into Current today, so a ConnectWise or HaloPSA workspace is not offered the tile at all. Broken down by partner, queue or person, the same floor applies to each row, and the rows it leaves out are counted on the tile face.

Ask for it by name, or pick the starter suggestion called Reactive ticket hours per endpoint. It divides the hours your team logged on service tickets in your PSA by the workstations and servers your RMM manages, and reports that as a monthly rate. Two things have to be true first: your PSA has to be one whose ticket time Current mirrors, and a tenant admin has to pick which RMM counts endpoints in Settings, under Integrations, because counting two RMMs would count the same machine twice. Until that pick is made the tile says so instead of showing a zero. On the tile you will always see which RMM was counted, how many endpoints are not matched to a partner yet, and how much of the time was entered on the day the work happened, because below about 90 percent same day entry the number is a floor rather than a measurement. Reactive means service tickets people asked for: tickets your monitoring raised automatically, and queues an admin marked as project work, are left out.

Because somebody worked those hours and Current holds no cost rate for them, a subcontractor or somebody without a Current login for instance, and pretending their time was free would flatter the number. The tile counts those hours, names them as unpriced hours, and says the margin reads higher than the truth because of them. Cost rates sync from your PSA, so the fix is to make sure everyone who logs service time exists there with a cost rate. Two more things worth knowing: managed services margin covers reactive labor only, not tools, licenses or project work, so it is not a full gross margin, and both money tiles need a one month timeframe because recurring revenue is a monthly figure. Tenant admins only.

Because a zero there would be a claim, and Current will not make a claim it cannot back. Backups failing, EDR alerts open, SOC incidents and phishing caught used to show a reassuring zero both before the product had ever sent Current a record and while a read was failing, which is the worst possible place to be wrong. Now each tile says which situation you are in: waiting on the first data from that product (it is connected but has sent nothing yet, and it fills in with the first sync), could not read this number (a tile that already had a number keeps it, says it could not refresh, and lets the Updated stamp keep aging), or the number would be incomplete (only part of a large fleet could be read, so check that the product is syncing under Settings, Integrations). A genuine zero over a product that has delivered records is still shown as a zero, because a quiet week is real. Devices compliant goes further: if any connected RMM cannot be read it refuses the whole percentage rather than publishing the RMMs that did answer, prints how many devices actually reported, and says when it is mixing Windows patch compliance with Mac policy compliance.

Yes, and they now say so under the number. Every ticket tile counts machine-raised alert tickets alongside the ones people asked for, exactly as your PSA stores them, so a workspace whose RMM opens hundreds of alert tickets a month sees counts that read higher than the support load feels. Nothing about your numbers changed when that note appeared: choosing which queues a tile counts is being built, and Current will ask you which queues to include when you build a ticket tile. Until you choose, your existing tiles count precisely what they counted before, so no wallboard changes under you.

Because two things about a ConnectWise service ticket were being read wrongly, and both are now fixed. Current used to treat a ticket as closed only when ConnectWise had stamped a close time, so every ticket ConnectWise considered finished without one stayed open forever: Tickets open now and all four aging buckets read too high, Tickets closed read too low, and the opened versus closed chart showed a backlog that only ever grew. Current now trusts ConnectWise's own closed flag, and where there is no close time it uses the ticket's last update as the closing moment and says so on the tile, since a resolution time built that way is an upper limit rather than a measurement. Second, Tickets closed by person used to measure the ticket owner, which ConnectWise desks often never set, so that leaderboard was mostly Unassigned; it now measures the person the work was dispatched to and takes the name from your own resource list, so bars move and some names appear for the first time. Current re-reads the last 24 months of ConnectWise tickets once, a chunk at a time across your normal syncs, so the numbers settle over a few hours to a few days. The same correction reaches the Tickets chip on a company, the ticket figures in a Strategic Business Review, the AI company insights, whitespace signals and the renewal dossier, so a review exported last month will not match one exported today. Autotask and HaloPSA workspaces are not affected.

Because time your PSA voided and time a manager rejected stopped counting as delivered work on 26 July 2026. A voided entry is one your PSA deleted during pre-invoice review, so nobody was ever billed for it. Rejected time was turned down as not worked. Counting either one as delivered hours overstated what your team actually did, so both are now out of Hours logged, Hours by person, My hours, Budget used, the count of projects over budget, and the Effort, Resources, Budget and Profitability lenses in Reports, along with the CSV those lenses export. Expect lower hours totals, a slightly flatter burn-rate trend and a shorter list of projects over budget: a project that was over budget only because of voided or rejected hours stops being named, which is the correction rather than a loss. Nothing was deleted. The entries stay exactly where they were on your Time page and in the task panel, marked and struck through, with the amount that is not counted printed beside the total, so you can always see what happened. Two places are deliberately different. The Budget % column and over-budget filter on the Portfolio page are the percentage your PSA reports, not Current's hours math, so they did not move. And where a figure is your PSA's own worked hours, which is the case for estimate versus actual on a synced task and for the close-out pack, it already left voided time out but can still include rejected time, because Current does not re-cut a number your PSA reported.

The number is the last one Current managed to read, and the tile says so rather than blanking or pretending. When a read fails, times out or is refused, Current keeps the number that was already on screen and prints a line in amber under it: Couldn't refresh this. Showing the last number Current read. The Updated stamp in the tile header keeps aging so you can see how old the number is, the tile retries by itself about every 90 seconds, and the line clears as soon as a read succeeds. A tile that never had a number says it could not read this one instead of showing a zero. That line now appears in the app, in Present mode and on a TV screen, so a wallboard can no longer show a frozen number silently. You will never see database wording on a tile face either: a read that takes too long says it took too long and Current will try again in a moment, and the technical detail goes to the browser console where an engineer can find it. If the line stays up for more than a few minutes, open Settings, under Integrations, and check that the product behind that tile is still syncing.

Because a ticket gauge is only as good as what your PSA sends Current, and not every PSA sends the same things. Tickets opened, tickets closed, tickets open right now, tickets closed by person, average resolution time, open ticket aging, and opened versus closed all need Autotask or ConnectWise. Average first response and SLA compliance need Autotask, because ConnectWise does not send a first response time or an SLA verdict and Current will not guess one. On HaloPSA no ticket gauge is offered yet: HaloPSA syncs your projects and time, but its service tickets do not reach Current, so those tiles would sit on zero forever and a zero on a wall screen reads as a quiet week rather than a missing feed. Nothing else is affected, since projects, hours, tasks, budget, capacity and the whole CRM are Current's own numbers and work whichever PSA you run. HaloPSA ticket sync is being built, and the gauges appear on their own once it lands, with nothing for you to switch on.