Services / Financial Performance
Financial Performance
Revenue Integrity
We find duplicates, improper charges, and leaks; a person confirms every finding.
- Entry model
- Leakage Diagnostic on historical data, fixed price
- Duration
- 2 weeks (10 business days)
- Pricing
- Price on request
Does this sound familiar?
- Payers debit studies for a mis-entered order or an expired authorization and we cannot defend ourselves.
- We buy duplicate supplies and they expire without anyone noticing.
- Promotions and returns are abused and I do not know what it costs.
Who it is for
- Clinics and imaging centers defending against unjustified debits
- Multi-site groups and specialty veterinary groups
- Pet nutrition and boutique pet e-commerce brands
- Buyer: owner, administration or internal audit
How we do it
Preparation
up to 5 business daysWhat worksYour environment with the extract loaded and an initial data-quality report.Data and explicit rules
Days 1–3What worksVersion 1 rules running on your history and producing a first list of suspects.Detection and linking
Days 4–6What worksA ranked list with the reasons for each case and a review tray with the evidence.Human review
Days 7–8What worksThe review tray in use: 30 to 50 findings classified as real, explainable, or false positive, with decisions recorded.Active rules and handover
Days 9–10What worksRules and tray running on your history, with a repeatable confirmed-amount calculation.
How it starts
Leakage Diagnostic on historical data, fixed price
What works at the end
Active rules and a review tray with evidence running on your history, with an amount confirmed by your own reviewer.
Price on request
Rules engine, review tray, and evidence package, plus a success fee on the confirmed amount recovered or avoided.
Monthly monitoring: review of findings, threshold tuning, and a confirmed-amount report.
You keep every deliverable. There is no obligation to continue after the entry engagement.
How we measure value
What we measure
- Confirmed amount (never the suspected amount)
- False-positive rate according to your reviewers
- Review time per case
- Amount recovered or avoided
How you verify it
Each finding comes with the vouchers that support it, and your reviewer decides whether it is real. Nothing acts on a person automatically.
Why K2-154
- We report the amount your reviewer confirmed, not the suspected one, and we measure false positives openly.
- The rules are readable: you can read them, question them, and change them.
- No alert acts on a person. We deliver evidence and rules; the decision is yours.
Open standards and tools we build with
- PostgreSQLRelational database
- dbt CoreVersioned data transformations
- Great ExpectationsData-quality tests
- scikit-learnAnomaly detection
- SplinkRecord linkage and identity resolution
- NetworkXRelationship graphs
- Apache SupersetDashboards
Product names, logos and trademarks are the property of their respective owners. K2-154 is an independent company and is not affiliated with, sponsored by or endorsed by these projects or organizations. Listing a technology does not imply certification or partnership.
Questions
Do you accuse people?
No. Every finding is a lead for a human reviewer. Anything involving a person is an employment and legal matter that the client decides.
How much history do you need?
Six to twelve months of the chosen front, with identifiers pseudonymized at source. With fewer than 6 months, the scope is reduced.
Which fronts can we cover?
One per diagnostic: billing and unjustified debits, supplies (expiry, duplicate purchases, consumption outside the pattern), or e-commerce (promotion abuse, returns, chargebacks).
Begin the conversation.
Tell us which systems are involved. We will respond with next steps.
hello@k2-154.com
We take on no more than three engagements at a time.
Other services in this group