AI that answers from your own numbers.
Every figure Dexor speaks is derived from your real data. The AI narrates; it never invents. It sees only the subtree you're entitled to see. The deterministic baseline is free, with no model, no latency and no per-use cost.
The baseline insights cost nothing to run.
Ranked at-risk flags and plain-language findings (critical below 60% attainment, warning to 85%, ahead at 100%+), plus forecast-driven risk and statistical anomaly detection, all computed deterministically from your team-performance tree. Immediate, defensible, and free. The AI layer is optional and sits on top.
Ask about your team in plain language.
“How is Ayesha’s territory tracking against target?” Ask Dexor answers synchronously and streams as it thinks. It gathers its own data through internal tools (dashboard snapshot, forecast outlook, org roster), looping until it can answer, then grounding every figure. History is kept for 90 days. And the caller can only ever see their own subtree, structurally.
Precise questions with auditable answers.
A structured ask (scope, subject, period, optional metric) is validated, authorized to your subtree, queued, and answered into your inbox. The figures are the deterministic summary scalars, verbatim; the model only writes the prose around them.
- Answer: rep 2026-10Tyler Morgan made 38 visits in 2026-10, 41% of his monthly visit target. 7 of his 17 targeted doctors are on track, plan adherence is 64% and there are 3 no-show days. The 10 doctors who are behind are where the month can still be recovered.09:14
- An expense claim needs your approvalA 2026-10 expense claim of 1845000 USD is waiting for your decision.08:05
- Indianapolis is heading for 41% of its quarter targetProblem: Indianapolis is projected to reach 41% of the quarter target, with $2,310,000 sold against a monthly target of $5,630,000. Reason: attainment momentum is down 9% month on month, doctor coverage fell to 41% and no-show days rose to 3, so fewer targeted doctors are being called on. Recommendation: re-plan Tyler Morgan's next two weeks around the 10 doctors who are behind, and schedule Joint Field Work in Indianapolis this week.17:22
Search your team's insights and sales summaries.
Hybrid keyword and semantic search across your team’s insights and per-rep sales summaries. Finding the records involves no language model; a short answer is then written from the retrieved hits only, and the hits are always shown alongside. Search is off by default, because its index is refreshed by a reindex run rather than on every change.
Indianapolis is the territory furthest behind. It has sold $2,310,000 against a target of $5,630,000, which is 41% of target, and at the current run-rate it is projected to reach 41% of the quarter target. Tyler Morgan is the rep, and coverage and no-show days both moved the wrong way this month.
| Record | Severity |
|---|---|
| Indianapolis is at 41% of target | critical |
| Indianapolis is at risk of missing quarter target | critical |
| Sales, Tyler Morgan: sold 2310000 vs target 5630000 (41%), 2026-10 | info |
The weekly narrative, already written.
One grounded, 2 to 4 sentence executive summary per manager, generated on a weekly schedule and waiting in the Insights view, with no live model call on read. Managers with nothing to report are skipped.
The team is at 85.6% of target and the quarter projects to 86%. New York South and Washington are carrying the month, while Indianapolis is the only territory below 60%, with coverage and no-show days both worse than September. Priority this week: a coverage push in Indianapolis with Tyler Morgan.
Alerts that name the problem, the reason, and the fix.
A statistical anomaly detector surfaces outliers as insight items. On top, the notice engine decides if something matters, what to say (in a strict Problem / Reason / Recommendation format, grounded in real figures), and who to notify, targeting exact people on the roster. Deduplicated over 24 hours, pushed live, dismissible.
- Indianapolis is heading for 41% of its quarter targetProblem: Indianapolis is projected to reach 41% of the quarter target, with $2,310,000 sold against a monthly target of $5,630,000. Reason: attainment momentum is down 9% month on month, doctor coverage fell to 41% and no-show days rose to 3, so fewer targeted doctors are being called on. Recommendation: re-plan Tyler Morgan's next two weeks around the 10 doctors who are behind, and schedule Joint Field Work in Indianapolis this week.08:40
- An expense claim needs your approvalA 2026-10 expense claim of 1845000 USD is waiting for your decision.08:05
- Answer: rep 2026-10Tyler Morgan made 38 visits in 2026-10, 41% of his monthly visit target. 7 of his 17 targeted doctors are on track, plan adherence is 64% and there are 3 no-show days. The 10 doctors who are behind are where the month can still be recovered.17:22
Right-sized models that stay up.
Dexor sets each account’s model from a curated list: an efficient default for everyday questions, with stronger models where depth matters, and an answer can step up one level mid-way when the question needs it. Inference runs on a primary provider with automatic failover to a second provider on outage or throttle. Failover covers the full agentic loop, not just one-shot answers. Streaming and prompt caching keep latency and cost down.
Indianapolis is the bigger problem. It is projected to reach 41% of its quarter target, below the 60% line, and the outlook flags falling momentum, a coverage decline and more no-show days.
Chicago North is projected at 63%. Its only open signal is a coverage decline, and most of its gap sits in one SKU, CRD-10, so a focused push there closes more of the distance.
Cache read 6,140
4 model calls in this answer
Escalated mid-answer:
haiku 4: 4,730 tok
sonnet 4: 6,694 tok
AI that can't blow your budget.
You buy AI as a predictable pool of prepaid Compute Credits, never an open-ended metered bill. Dexor manages the cost controls, and your admins see the credits used and what each team has reserved, so there are no surprises.
Reserving credits guarantees a team its share. What is left over is shared by everyone without a reservation. You cannot reserve more than your allowance.
| Team | Used | Reserved | |
|---|---|---|---|
| North region | 74,200 | 150000 | Save |
| South region | 51,900 | 120000 | Save |
Every feature in Dexor intelligence.
Each one has its own page.
Ask Dexor
Ask Dexor answers questions about your team from your real dashboard, forecast and roster, and shows how it got there.
Read moreAsk your data
A structured question about any rep in your line, answered into your inbox with the figures exactly as Dexor computed them.
Read moreSearch your data
Hybrid keyword and semantic search over your team's insights and per-rep sales summaries, with an answer grounded only on what it found.
Read moreWeekly executive summary
Every Monday Dexor writes each manager two to four sentences on their team, from the week's findings, and leaves it at the top of Insights.
Read moreAnomaly alerts
Dexor watches the forecast for real trouble, decides whether it matters, and tells the right leaders the problem, the reason and what to do.
Read moreDocument exports
Generate PDF, Word, PowerPoint or Excel reports from the dashboard, Insights or a single Ask Dexor answer, and download them from Documents.
Read moreCommand palette
One search box for doctors, chemists, SKUs, distributors and every screen, with Ask Dexor one keystroke away.
Read moreResilient AI
Dexor starts every answer on an efficient model, steps up only when a question needs it, and is built to fail over to a second provider.
Read moreTargets and forecasting
The deterministic team-performance tree the AI narrates: attainment, coverage, and the quarter verdict.
Targets and forecastingPlatform and admin
Per-account AI configuration, seat governance, and the tamper-evident audit trail behind the intelligence layer.
Platform and admin