Deductions · Wave 1
Get back the money that came off the invoice.
Deductions is a planned recovery platform: it collects deduction documents from retailer portals, matches remittances to invoices, classifies every short payment, and assembles the case that disputes it.
- Collect
- Match
- Classify
- Dispute
- Recover
- Stage
- Planned
- Portfolio bay
- Bay 11
- Focus area
- Quality & Assurance Lab
- Part of
- The Collective Foundry
The problem
Why Deductions exists.
Short payments are written off because disputing them costs more attention than they individually return. Collectively they are substantial, and the documents needed to contest them are scattered across retailer portals that were never designed to be read in bulk.
Who it's for
Suppliers invoicing large retailers and distributors — not only consumer goods brands — whose finance team is small enough that nobody owns deductions full time.
The opportunity
The market Deductions is built for.
Deduction management exists at both ends of the market and is thin in the middle. HighRadius serves large enterprises through a quoted sales process, with deductions sold as one module among credit, collections, cash application and presentment. Specialists such as TrewUp and iNymbus focus on deductions directly. Vividly and Promomash cover them for consumer goods brands as part of a trade promotion platform. What is missing is deductions on their own, at a published price, for a supplier who is neither an enterprise nor a consumer goods brand.
Why now
The timing behind it.
- 01
Retailers have pushed more cost back onto suppliers through chargebacks and allowances, and the volume of deductions has risen with it.
- 02
The supporting documents are now almost always available through a portal, which makes systematic collection possible for the first time.
- 03
Reading and classifying unstructured remittance documents in bulk has become reliable enough to act on.
What it does
The capabilities inside Deductions.
Collection from the portals
The supporting documents gathered from each retailer’s system automatically, rather than downloaded one at a time.
Remittance matched to invoice
What was paid reconciled against what was billed, with every difference isolated and named.
Classified, not just counted
Each deduction sorted by type and by whether the evidence supports it, so attention goes where recovery is likely.
The dispute, assembled
The case built with its documents attached and tracked through to whatever comes back.
Where we start
The smallest valuable first step.
- Who it is for first
- Finance teams of two to ten at suppliers invoicing major retailers and distributors
- The first thing it does
- One retailer’s deductions, collected automatically and disputed with the evidence attached.
- Likely commercial model
- A share of what is recovered, or a capped monthly fee once recovery is predictable
- What has to be proven
- That automated classification is accurate enough that a finance team disputes on its word.
What makes it innovative
Why Deductions is different.
Collection is the hard part
The reconciliation is well understood. Getting the documents out of each retailer’s portal, reliably and repeatedly, is the work nobody wants to maintain.
Priced on what comes back
Charging a share of recovery removes the decision entirely. Nothing is owed until money arrives.
Assembled, never auto-sent
A wrongly disputed deduction costs more than it recovers. The platform prepares the case; a person decides to send it.
The usual way
Short payments noticed at month end, investigated if there is time, and written off when there is not.
The foundry way
Every deduction collected, classified and either accepted deliberately or disputed with evidence.
How it compares
Deductions against the deduction platforms sold to enterprises and to consumer goods brands.
Every competitor cell below is drawn from that vendor's own official site on the date shown. Where a capability was not documented there, it is marked Not publicly verified rather than Not available — absence of evidence is not evidence of absence.
| Capability | Deductions Not built — scoped for Wave 1 | HighRadius Deductions module, quoted on invoice volume | TrewUp Cloud deduction platform, pricing not published | iNymbus Deduction management, pricing not published | Vividly / Promomash Trade promotion platform including deductions; Promomash from around $500/month |
|---|---|---|---|---|---|
| Getting the facts Automatic collection from retailer portals Published buying guidance names this as the main time-saver, and it is the work nobody wants to maintain. | Planned Depth 0/5 · target 4/5 One retailer at a time, done properly. The difficulty here is the barrier to entry as much as the cost of it, which is the reason to start rather than the reason to avoid it. | Advanced Shipped Depth 5/5 Its own comparative material cites rivals struggling with large retailer portals, which is a claim about competitors but a reasonable signal of its own coverage | Included Shipped Depth 4/5 Automated extraction of deduction data is its stated specialism | Included Shipped Depth 4/5 | Included Shipped Depth 3/5 |
| Getting the facts Remittance matched to invoice Turning a payment that arrived short into a list of specific differences with names attached. | Planned Depth 0/5 · target 3/5 Conceded. HighRadius has a decade of cash-application work behind it and we would not be close. | Advanced Shipped Depth 5/5 Deduction type identified during cash application; claims 90%+ straight-through posting, which is a vendor figure | Included Shipped Depth 3/5 | Included Shipped Depth 3/5 | Included Shipped Depth 3/5 |
| Getting the facts Each deduction classified and assessed Separating the allowance you agreed to from the penalty you should contest. Where the judgement lives. | Planned Depth 0/5 · target 4/5 Accurate enough that a finance team acts on it without re-checking by hand, which is the actual bar and the thing that has to be proved. | Advanced Shipped Depth 5/5 | Included Shipped Depth 4/5 Categorisation of deduction data is named directly in its own description | Included Shipped Depth 4/5 | Included Shipped Depth 4/5 Deduction management is a named specialism, tied to the trade promotions that caused them |
| Getting the money Dispute assembled and tracked to recovery Whether the product ends at a report or at money arriving back. | Planned Depth 0/5 · target 5/5 Assembled with its evidence and tracked to what comes back. Nothing is ever sent automatically — a wrongly disputed deduction costs more than it recovers. | Included Shipped Depth 4/5 | Included Shipped Depth 4/5 | Included Shipped Depth 4/5 | Included Shipped Depth 4/5 |
| Who can actually buy it Available without buying a platform around it Most buyers here want deductions solved, not an accounts-receivable suite or a trade promotion system. | Planned Depth 0/5 · target 5/5 Deductions on their own, for a supplier who is neither an enterprise nor a consumer goods brand. One of the two rows this is actually for. | Limited Shipped Depth 1/5 Deductions is one module among credit, collections, cash application and electronic presentment | Included Shipped Depth 4/5 Genuinely focused on deductions rather than bundled | Included Shipped Depth 4/5 | Limited Shipped Depth 1/5 Bundled with trade promotion by design, which is the right shape for a consumer goods brand and the wrong one for anybody else |
| Who can actually buy it Published price, no sales cycle A finance team of three cannot run a procurement process to recover money it is already losing. | Planned Depth 0/5 · target 5/5 Priced on what is recovered and published openly. The clearest gap in the shortlist and the cheapest one to occupy. | Not publicly verified Depth 0/5 Quote-only, no free trial and no small-business edition documented | Not publicly verified Depth 2/5 Pricing is not published | Not publicly verified Depth 2/5 One review describes competitive pricing; no list price is published | Limited Shipped Depth 2/5 Promomash publishes a starting figure around $500/month; Vividly does not publish pricing |
Who each tool is actually best for
- Deductions Not built — scoped for Wave 1
Not a buying option. Every row is target specification at an achieved depth of zero. The intended buyer invoices large customers and has nobody who owns deductions full time.
- HighRadius Deductions module, quoted on invoice volume
Large enterprises. Deduction type is identified during cash application on one data model shared with credit, collections and forecasting — depth nothing else here approaches.
Source: Deductions management for consumer packaged goods , Non-trade deductions — classification during cash application
- TrewUp Cloud deduction platform, pricing not published
Consumer goods brands wanting deductions handled directly rather than as part of a wider suite. Automates extraction, analysis and categorisation of deduction data.
Source: TrewUp — automated extraction, analysis and categorisation of deduction data
- iNymbus Deduction management, pricing not published
Suppliers with high deduction volume across several retailers who want the processing itself taken off their hands.
- Vividly / Promomash Trade promotion platform including deductions; Promomash from around $500/month
Consumer goods brands that need trade promotion planning as well as deductions. Vividly is built for small to midsize brands and states a very large volume of trade spend under management.
Source: Vividly — trade promotion and deduction management for small to midsize CPG , Promomash — forecasting, deduction management and budget allocation
Considered and excluded
- Card chargeback tools (Chargeflow, Chargeback Gurus) — A different problem that shares a word. Those contest consumer card chargebacks through the card networks; this contests short payments by business customers against invoices. Comparing them would confuse a reader rather than inform one. Check current status
How this comparison is made
- Comparative claims follow ACCC guidance: they must be accurate, fair and supportable on reasonable grounds, judged on the overall impression rather than line by line.
- A competitor capability is only marked as present where its official site or documentation describes it. Where it was not documented the status is “Not publicly verified” — not “Not available”. This is a desk review of published material, not a hands-on benchmark.
- This is not an empty market. HighRadius scores 5/5 on three rows and we concede matching outright. The opening is narrow and specific: deductions sold on their own, at a published price, to a supplier who is neither an enterprise nor a consumer goods brand.
- Lock-in here is real but conditional — open disputes hold a customer and closed ones do not. Charging on what comes back removes the buying decision entirely and matches how the recovery agencies in this market already charge, with the difference that a system is left behind.
- Depth scores (0–5) are our own analytical assessment of how deep a capability runs. They are not vendor-published figures, and they are separate from the evidence status beside them.
- Deductions is not built. Every one of its cells shows delivery “Planned” and an achieved depth of zero. A target beside it is the specification being built toward, never a claim about today.
- Targets are published only for our own column and only with a stated reason. Remittance matching is set at 3/5, below HighRadius, because a decade of cash-application engineering is not something a first version matches honestly.
The approach
How we build it.
When a large retailer pays a supplier, the amount that arrives is often smaller than the amount invoiced. The difference is a deduction, and it may be a legitimate allowance, a penalty for a delivery that missed its window, or an error. Working out which is slow, manual and usually skipped.
Deductions is a planned platform that does that work. It collects the supporting documents from the retailer portals, matches remittances back to invoices, classifies each short payment, and assembles the dispute with the evidence attached — then tracks what actually comes back.
The outcome: money that was quietly written off becomes money that was argued for.
The outcome
What changes when it's in place.
- Deductions are reviewed rather than absorbed.
- The evidence for a dispute is already gathered when it is needed.
- Recovery is measured rather than assumed.
What could go wrong
The risks, named.
-
Retailer portals are inconsistent, undocumented and change without notice.
One retailer at a time, properly. The difficulty is the barrier to entry as well as the cost of entry.
-
Suppliers may prefer an agency taking a percentage to software on a subscription.
Charge on recovery rather than on seats. The buying decision disappears when nothing is owed until money arrives.
-
A deduction wrongly classified as invalid damages a retailer relationship worth more than the claim.
Nothing is disputed automatically. The platform assembles and recommends; a person sends.
The open questions
What would need to be true.
-
Enough deductions are genuinely invalid to pay for the work.
How we would know: Run a sample of historical remittances and measure the proportion that would have been successfully disputed.
-
Portal collection can be kept working as retailers change.
How we would know: Track breakages per retailer per quarter, and how long each takes to repair.
-
Suppliers will act on an automated classification.
How we would know: Measure how many recommended disputes are actually sent without being re-checked by hand.
A scenario
What this could become.
$9M–$21M
at $3M ARR, on 3–7× ARR
A scenario, not a valuation, a forecast or an offer. It states what the venture might be worth IF it reached the revenue shown, at multiples other private SaaS businesses have transacted at. Nothing here has revenue today.
Published private-SaaS transaction benchmarks, checked 2026-10-13: private deals broadly transact at 2–7× ARR, with medians around 3–5×. These come from advisory firms and marketplaces rather than audited transaction data.
How it fits the foundry
A venture, not a department.
Deductions moves through the whole foundry — discovery, design, engineering, quality and launch — like every venture we build. It's most visible in Quality & Assurance Lab, where its business purpose is sharpest.
- Discover
- Define
- Design
- Build
- Assure
- Launch
- Learn
- Scale