Sign in
Set up for Industrial distributors

AI for insights and analytics

AI analytics for industrial distributors

Industrial distribution runs on quotes, branches and contracts. AI analytics reads the quote log, the invoice ledger and the contract price file, answers questions like which customers use us to check another supplier's price, and shows the rows that say so.

Ask it in your own words

These are the questions industrial distributors ask. Each one maps to a measure our tools compute from your files. The AI model chooses the measure and explains the result; the arithmetic is done by our code, and every total is checked.

The questionThe measure behind itWhat comes back
Which customers quote us and never buy?Quote conversion, count and valueConversion by count and by value per customer, with the estimating hours spent on the ones under ten percent.
Which accounts are past their usual order gap?Dormant accounts, by prior valueA list by prior-year revenue against each account's own pattern, by branch and rep.
Where are we invoicing below contract?Price realisation against contractInvoiced price against the contracted price by line and customer, the leakage totalled.
Which key sites buy most of a category from someone else?Share of spend at key sitesOur share of estimated site spend per category, with the gap in dollars.
Who walks in and buys at the counter without an account?Counter sales without an accountCounter revenue with no account number, by branch, and the repeat buyers among it.
Which lost lines are costing us orders?Lost linesLines a customer used to buy and no longer does, valued at the prior run rate.

A conversation, with the figures

Every figure below was computed by a tool from the rows and checked before it was shown. The lines sum; the percentages match; each line opens to the rows that make it.

How much quote value did we lose last quarter, and where?

Lost quote value is $4.9m across 3 branches, against 26 percent conversion by value. Eastside branch accounts for $2.4m, 49% of the lost value.

Eastside branch$2.4m49% of the total
Riverside branch$1.6m33% of the total
Northgate branch$900k18% of the total
The 3 lines sum to$4.9m0 unexplained

Who is behind the Eastside figure?

3 customers hold $2.0m of the $2.4m lost at Eastside branch, 84% of it. Two of them have sent more than forty quotes each and converted under ten percent.

Kestrel Engineering$980k47 quotes, 4 won; 62 estimating hours
Bramley Fabrication$720k41 quotes, 3 won; 55 estimating hours
Oak Ridge Plant$310k12 quotes, 5 won
These 3 are$2.0m84% of Eastside branch

What do we do about the two price checkers?

Decide, rather than keep quoting. Either a contract price that removes the reason to shop, or a quote fee, or a slower turnaround. The estimating hours are the cost of not deciding.

Eastside branch managerMeet Kestrel and Bramley with a contract price proposalThis month
Sales managerPut quote conversion by value on the weekly branch callWeekly
Estimating leadFlag customers under ten percent conversion before quotingOngoing
Which customers quote us and never buy?Which accounts are past their usual order gap?Where are we invoicing below contract?Which key sites buy most of a category from someone else?

The measures behind the answers

Each measure has one formula, one source and one meaning. They are computed per branch and rep and in total, and every one carries an identity that must hold before it is shown.

MeasureFormulaFromWhat it tells you
Category penetrationAccounts buying the category ÷ accounts in the segment, by branch; categories held per account against the segment normLedger with category; account masterWhich accounts buy three categories where similar accounts buy six
Dormant accounts, by prior valuePrior-year revenue of accounts past k × own order gap ÷ prior-year revenueLedgerRevenue that stopped, ranked for a call
Quote conversion, count and valueQuotes matched to orders ÷ quotes decided; quoted value converted ÷ quoted value decidedQuote log; order fileWhere estimating time goes and which customers price-check
Lost linesValue of lines cancelled or unfulfilled for no stock, by branch, item and accountOrder and shipment filesSales the shelf turned away
Counter sales without an accountCash and card sales by payer, repeated visits, valueCounter sales fileWalk-in customers worth a trade account
Coverage at cadenceRevenue of accounts touched within tier cadence ÷ revenue assignedCRM; assignment file; ledgerWhether key accounts are being spoken to
Price realisation against contractInvoiced price ÷ contract or matrix price, by lineLedger; price fileContract prices not applied; overrides at the counter
Share of spend at key sitesRevenue at the site ÷ estimated site MRO spend, from headcount or stated budgetLedger; account master with site dataHow much of a plant's spend you hold
Reactivation rateDormant accounts that ordered within 60 days of a logged call ÷ dormant accounts calledLedger; CRMWhether working the dormant list pays
Contribution per customerGross margin − cost to serveLedger; delivery and order filesAccounts that earn after what they cost

Each measure is worked through, with the export it comes from and what to drop, in Sales KPIs for industrial distributors.

What the AI model does, and what our tools do

The AI model chooses the measure

From your question and the measures declared for industrial distributors, the model picks the one that answers it, and the period and comparison the question implies.

Our tools do the arithmetic

Deterministic code reads the rows, computes the measure, and checks the identities below. The same question on the same data gives the same answer, every time.

The AI model explains, and cites

The AI model writes the sentence around the result, naming the branch or account behind it. It states no figure that is not in the result, and every figure links to its rows.

The identities that must hold

Quote conversionQuotes = converted exact + converted probable + lost + open within validity
Lost linesLines ordered = shipped complete + shipped short + cancelled no stock + cancelled other + open
Category penetrationCategory revenue sums to the ledger; every account in one segment
DormancyCustomers = dormant + active + too few orders

What it reads

The exports industrial distributors already produce. Column names are mapped once and the mapping is reused. A file is the way in; scheduled delivery and connections to your systems come with the plan, and every source is listed here.

  • Ledger with category
  • Account master
  • Quote log
  • Order file
  • Order and shipment files
  • Counter sales file
  • CRM
  • Assignment file
  • Price file
  • Delivery and order files

Who owns each answer

An answer is a list with an owner and a cadence, or it is a chart nobody works.

MeasuresOwnerCadence
Dormancy; coverage; reactivationSales manager; branch managerWeekly
Lost linesBranch manager with purchasingWeekly
Quote conversion; counter sales without an accountBranch manager; inside sales leadMonthly
Category penetration; price realisation; contribution; site shareCommercial directorQuarterly

Questions industrial distributors ask about AI analytics

Why conversion by value and not by count?

Because count conversion counts a fifty-dollar quote the same as a fifty-thousand-dollar one. A branch at 41 percent by count and 26 percent by value is winning the small quotes and losing the big ones, and the second figure is the one that explains the revenue line. The tool reports both, side by side, so the difference is visible.

Can it read our quote system export?

Yes. It needs a quote file with the customer, the date, the value and the outcome, plus the invoice ledger to confirm what was actually bought. Column names are mapped once, on the data map, and the same mapping is reused for every later file.

Does it know our contract prices?

If you give it the contract price file, price realisation against contract is computed line by line: invoiced price over contracted price, by customer, with the leakage totalled and the lines listed. Without the file it falls back to list price, and says so.

How does the model avoid making figures up?

The AI model never produces a figure. It picks the measure that answers your question, our tool computes that measure from the rows and checks it, and the AI model writes the sentence around the result. If the question needs a measure the tools cannot compute, it says it cannot compute that, rather than guessing.

See it on your data

Bring a few thousand rows. The data map opens next, every column mapped once, and the first question is answered in minutes. Free, in your browser, no account.