Business management · Spend Analytics & Procurement
Buying well is one thing; proving where the money is going, at what price and with what risk is another. Spend Analytics & Procurement reads orders, receipts, contracts, advances, accounts payable and bank settlements; splits spend into three stages that are never added together; separates what procurement can negotiate from taxes and payroll; and gives back concentration, pricing, opportunities, measured savings, duplicate payments to recover, three-way matching, days payable outstanding, comparison with market inflation and compliance that actually blocks. Nothing is typed in: everything is derived from what the company has already bought.
The problem and the fix
If any of these situations sounds familiar, it is because they happen in almost every graduation company, until the right system comes in.
Nobody can say how much the company spends in a specific category without spending two days building a spreadsheet.
A spend cube with its own procurement taxonomy, crossing category, supplier, cost center, class, buyer and item, with a click that opens the entries behind the number.
The same material is bought from three suppliers at different prices and nobody notices, because each purchase was looked at in isolation.
Reference price per item with last, best and average paid, and the gap turned into a savings opportunity calculated on the actual quantity purchased.
A large share of expenses reaches accounts payable without ever going through an order, quote or approval.
Direct measurement of spend that came in without an order, by supplier and by buyer, and a compliance policy that flags each case with the document in hand.
Negotiated savings are announced in meetings, but nobody can prove they showed up in cash.
Savings with a declared baseline, a calculated value and realization posted month by month, compared with the annual target. Announced savings and realized savings side by side.
An entire category depends on a single supplier, and that only comes to light when they're late delivering for the event.
Concentration index by category, single-source suppliers flagged and dependency risk calculated per supplier, with an alert before it becomes a problem.
The supplier review is a score someone made up off the top of their head, months after the delivery.
A scorecard calculated from on-time deliveries, accepted items, price against the reference and up-to-date documents, with low-volume suppliers flagged as insufficient sample instead of getting a made-up score.
The same invoice was entered twice, the boleto was paid at the bank and settled by hand later, and nobody will ever find out.
Five recovery scans across thirty-six months of accounts payable, each finding with the documents side by side and the recoverable amount calculated. In a database a few years old, that money exists.
Process indicators never add up because taxes, payroll and loans land in the same bucket as material purchases.
Total spend and addressable spend separated: the full expense shows up for the Controller's office, and procurement indicators are calculated only on what procurement can negotiate.
The policy says purchases above the limit need a quote, and the order goes out without one anyway.
Seven rules checked at the moment of issuance, with real blocking and a message that says what to do for the order to proceed.
Your basket went up 18% this year and nobody can say whether that's good or bad.
Comparison with IPCA, IGP-M, INCC or the index linked to each category, from the monthly series the company registers, with the difference translated into reais and the item that contributed most to it.
Each branch negotiates on its own and the group pays three different prices for the same item without knowing it.
Headquarters consolidated with branches, showing suppliers that serve several locations and the same item compared across them, with savings calculated from the lowest price the group itself has already achieved.
In practice
Everyday situations at graduation companies, from the problem to the result, using the modules on this page.
The company bought specialty paper from three print shops depending on how urgent the class was. Each purchase looked reasonable on its own.
The cube grouped the item across the various spellings of its description, the reference price showed the gap between the best and the average paid, and the opportunity calculated the value on the quantity bought during the year.
The negotiation happened with the number on the table, the closed price became savings with a declared baseline, and realization showed up in the following months, measured by the company's own purchases.
A third of what reached accounts payable had no order: purchases made directly by operations to handle urgencies.
Compliance broke down spend without an order by supplier and by buyer, and the split purchase policy flagged cases where several small orders added up to more than the limit that would require a quote.
The five largest suppliers outside the process were put under contract, and the percentage of spend under order became an indicator tracked every month.
All graduation gowns had come from a single supplier for years. The relationship was good, so nobody questioned it.
The positioning matrix placed the category in the strategic quadrant, with high risk and high financial impact, and the supplier's dependency risk rose to critical.
A sourcing initiative entered the pipeline with a deadline and an owner, a second source was qualified and the first supplier's price improved even before the switch.
The cube was outdated, there was uncategorized spend, savings awaiting approval and serious deviations still unaddressed.
The pipeline rebuilt the cube and recalculated prices, index, concentration, scorecard, opportunities, recovery, compliance and forecast in the correct order. The dashboard gathered the alerts and made it possible to export exactly the slice being analyzed.
The team started the day with the exception queue, knowing which analysis was up to date, what needed a decision and who had permission to take each action.
The warehouse entered the invoice through receiving, and finance entered the same invoice by hand whenever the integration failed. It became a habit, and nobody counted how many times it had happened.
The recovery scan cross-checked document numbers by supplier, compared the amount paid with the bill amount and flagged each case with both documents side by side.
The company took the cases to the supplier with proof in hand, recorded the amount actually recovered and began tracking that number as the module's return.
Classification coverage was stuck at 60% no matter how much the team classified. What was always missing was taxes, payroll and transfers between accounts, which nobody knew where to put.
Exclusion rules marked that spend as not addressable by procurement. It still showed up in total expenses and was removed from the denominator for coverage, off-process purchases and spend under contract.
Coverage started rising for real with every classification, and off-process purchasing dropped to the real number. Before, every tax payment counted as a purchase without an order.
The company knew it paid some bills late, but the interest was diluted into the bill amount and never showed up as a number of its own.
With the paid stage in the cube, interest and late fees became an entry separate from the item price, and days payable outstanding started being measured weighted by value.
The year's cost of late payments showed up on one line, with the supplier and category where it was concentrated, and the discussion about extending payment terms finally had both sides on the same screen.
How it works
Every step is a real module, and what comes out of one goes into the next without anyone typing it again.
Module by module
22 modules and 216 features on this page, all running on the same class, graduate and event records.
The company's expenses read by the tree that the buyer understands, and not by the accounting account.
The expense sorts itself out, and what's left becomes a short line instead of thousands of lines.
Who the company depends on, how much of the expense escapes the process and where the base is too fragmented.
Each opportunity is born from a deterministic rule, with a calculated value and launches that support it.
How much the company usually pays for each item, and how much this has risen without anyone noticing.
Announced savings are not realized savings. Here the two appear side by side.
Each category has an owner, position in the risk and value matrix, strategy and initiatives that deliver the goal.
Note that comes from what happened, and not from opinion: delivery on time, item accepted, price and document up to date.
The rule that the company agreed to follow, checked against what actually happened.
The year's plan by category, the forecast for the coming months and the routine that keeps everything up to date.
Money that the company has already paid and can ask for back. On a basis with a few years of bills to pay, it exists.
Order, receipt and bill count as the same purchase. When all three don't close, someone paid what they didn't receive.
How long between receiving the note and paying, how much of it became interest and how much the advance brought back.
Balance, maturity, adjustment and price inside versus outside. This is what sustains renewal.
The whole module looks at supplier and category. Here the question is who buys it.
AI suggests the category with justification and confidence. Nothing enters the cube without someone accepting it.
Your basket rose 18%. How much did the market rise? Without that second half, the first number says nothing.
Before opening the RFQ: what happens to price, concentration and risk if I change the base?
Each branch trades alone. The group's real purchasing power appears here.
Percentage spent with a local, small, certified and diverse supplier, with coverage on the side.
An alert that requires remembering to open the screen is not an alert. And the cube spreadsheet is not data output.
Every indicator is a sum that someone will dispute. The path to the document must exist.
Compare the modules and choose the setup that fits your operation.