The Margin Roundtable · Tuesday 19 January 2027, London · finance, procurement and commercial leaders, one room, no slides Ask for a seat →
What we hear from a chief financial officer

“Our suppliers have ten names and our parts have three codes. Nothing adds up.”

Each ERP, site and buyer spells suppliers its own way and codes the same part differently. Every report starts with someone cleaning data by hand, and prices cannot be compared across sites.

The film · 2:08
You are not the only one asking

What else we hear from finance, procurement and commercial teams.

“The same part costs five prices because it has five codes.”
Category manager
“I spend the first week of every project cleaning names.”
Data analyst
“We cannot say what we spend with our biggest supplier group.”
Chief procurement officer
Where it costs you

Why it is hard to fix today.

The same thing has ten names

One supplier appears under a dozen spellings across your ERPs. One part has three codes and five descriptions. Totals look right and are wrong.

Prices cannot be compared

One site buys by the box, another by the unit, a third in another currency. Until the basis is the same, nobody can say who pays more.

What changes for you

Cleanse, normalise and enrich all your data, so every figure joins up.

Today

  • A supplier under a dozen names
  • One part with several codes and descriptions
  • Prices in different units and currencies
  • Data cleaned by hand before every report

With Spendkey

  • One parent company per supplier and customer
  • One identity per part and SKU
  • Every price on one unit and currency
  • Clean data kept clean, every change recorded
One identityfor every supplier, customer, part and SKU
One basisunits, pack sizes and currencies normalised
Every change keptwith who approved it and when
How it works

Spendkey's master data agent cleans your supplier, customer and item records before any figure is shown. It groups names into parent companies, finds duplicate suppliers and verifies VAT and bank details, matches the same part or SKU across systems and descriptions to one identity, puts units and currencies on one basis, and enriches each record from public filings and standard codes. Every change is kept, and you approve what matters.

  1. Read every record Supplier, customer and item records are read from every system you run.
  2. Cleanse and group Duplicates are found, names grouped into parent companies, and VAT and bank details verified.
  3. Normalise Parts and SKUs are matched to one identity, and units and currencies put on one basis.
  4. Enrich and keep Records are enriched from public filings and standard codes, and every change is kept for approval.

What it works from: Your supplier, customer and item master data, accounts payable and receivable extracts, and invoices.

Do we have to clean our data before we start?

No. Cleaning, normalising and enriching your data is the first thing the agents do, and it keeps running as new data arrives.

Does it change the records in our ERP?

No. Your ERP stays as it is. Spendkey keeps its own clean copy, and you can send corrections back if you want them.

Client result · Traffic technology company, Europe€2.03bn

Every supplier, every invoice, agreed to the euro

9,919 supplier names cleaned to 8,046 parent groups, with every dashboard total agreed to the euro.

See every customer story

More in Across everything

Back to the platform
  1. AI agents that do the workAsked by the finance director
  2. Cassian, the AI assistantAsked by the finance director
  3. Contract readingAsked by the head of procurement
  4. Supplier Control Tower: everything on one supplierAsked by the head of procurement
  5. Customer Control Tower: everything on one customerAsked by the commercial director
  6. Margin tracked to EBITDAAsked by the chief financial officer
  7. Opportunity pipeline: savings and price risesAsked by the chief financial officer
  8. Savings and initiative trackerAsked by the chief procurement officer
Finance director asks
“We have the dashboards. What we don't have is the people to act on what they show.”

AI agents that do the work: they find the money in every deal, order and invoice, research outside events, draft claims and negotiation emails, hold what breaks a rule and chase each owner until it is done.

See it at work · film 2:48
Finance director asks
“I just want an answer without waiting a week for an analyst.”

Ask Cassian anything about your spend, contracts, market and margin, in your own language, and get an answer with its source, a report you can export, or an action you confirm.

See it at work · film 2:19
Head of procurement asks
“The answer is in a contract somewhere. It takes a day to find it.”

Every supplier contract and customer contract, scans and amendments included, read in any language and turned into terms you can search: prices, index clauses, rebates, obligations and renewal dates.

See it at work · film 2:13
Head of procurement asks
“I walk into supplier meetings with half the facts, and the supplier has all of theirs.”

Everything found on a supplier in one view: spend, supplier contracts, renewal and notice dates, performance, rebates, signals and open actions, with a brief for your next negotiation.

See it at work · film 2:12
Commercial director asks
“My account managers walk into price talks without the numbers the contract gives them.”

Everything found on a customer in one view: margin by customer contract and deal, cost pass-through, renewals, unbilled charges and open actions, with the facts for your next price conversation.

See it at work · film 2:33
Chief financial officer asks
“EBITDA is down on plan, and it takes a month to find out why.”

Your margin tracked continuously against EBITDA, on both sides of every deal: what price, volume and currency did to it this month, against plan, and every saving, price rise and recovery seen as it lands.

See it at work · film 1:46
Chief financial officer asks
“Nobody keeps the pipeline up to date, and our price rises aren't in it.”

Every saving and price rise the agents find, on the supplier side and the customer side, put into one pipeline automatically, ranked by value, with an owner, and moved on as the evidence arrives.

See it at work · film 2:05
Chief procurement officer asks
“I've got savings in five spreadsheets and no idea which ones are real.”

Each savings initiative run as a project, with an owner, milestones and sign-off at every stage, followed from agreed to banked in EBITDA. An add-on for teams that report savings.

See it at work
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.