Distribution produces data faster than any team can read it: thousands of visits, orders, invoices, collections and shelf checks a week. Most of it dies in dashboards nobody opens. The point of route-to-market analytics is not more charts — it is answers at the moment of decision: which beats to re-cut, which outlets are quietly dying, where scheme money is working, who needs help before the month is lost.
One data model, so the numbers agree
Every metric here — coverage, strike rate, drop size, sell-through, scheme spend, DSO — is computed from the same records that ran the operation: visits, orders, van days, collections. There is no export step where the sales number and the finance number part ways. Dashboards and pixel-perfect reports sit directly on the operational model; Data Studio adds governed star-schema datasets with row-level security for warehouse-grade analysis and embedded BI.
The questions the route must answer
- Coverage: promised versus reached, by beat and rep — and the outlets no beat has touched in a cycle.
- Productivity: calls per day, strike rate, drop size, time-in-outlet versus time-in-transit.
- Sell-through: secondary sales against primary, stock cover by node, the divergences that flag stuffing or stock-outs.
- Money: scheme spend against the volume it bought; ageing and collection productivity by beat.
- The shelf: measured share of shelf where it is tracked, joined to the offtake it is supposed to drive.
Ask, don't build
Sense Assist sits on the same model and answers in plain language: "which outlets over sixty days outstanding were visited this week?", "compare drop size across the two new beats", "which schemes moved volume in the south last month?". It runs as the person asking — a rep sees their beats, an area manager their area — grounded in your data, with writes gated behind approval. The governance story is documented, not implied: how Sense is governed.
From answer to nudge
The step after asking is being told: nudges that surface the outlet gone quiet, the beat running chronically long, the suggested order that does not match the season — deterministic signals, delivered by AI, explained honestly (the nudge that says don't buy). Analytics that waits to be opened is a report; analytics that speaks up is an operating advantage.
Common questions
What route-to-market metrics matter most?
Coverage (promised versus reached visits), strike rate (visits that produced an order), drop size, secondary sell-through against primary despatch, stock cover per node, scheme spend per incremental case, and receivables ageing by beat. Together they answer the only two questions that matter: is the route being worked, and is the working of it making money?
Are the dashboard numbers reconciled with finance?
They are the same records — orders, invoices and collections post to the ledger the CFO closes, and analytics reads that model rather than an export of it. When the sales dashboard and the trial balance disagree, one of them is wrong; here they cannot part ways, because there is only one set of books.
Can field teams see analytics on mobile, offline?
The operational views a rep needs — their beats, outlets, outstanding, targets — travel with the offline app. Heavier analysis lives on governed dashboards on the web, and Sense answers questions on both surfaces, scoped to what each user is permitted to see.
How does Sense answer questions about my distribution data?
Sense is grounded in your tenant's own data model and runs under the asking user's permissions — it reads what they could read, nothing more, and its actions that would change data require approval. Ask in plain language; it queries the governed model and shows its working, rather than hallucinating a number.
Can we build our own datasets and reports on top?
Yes — Data Studio provides governed, star-schema datasets with row-level security, a managed warehouse for history, and embedded BI, so your analysts extend the model instead of rebuilding it in a side system.