Combines what only Redbrain can see — invoicing history, payment terms, current CPC exposure — with what the rest of the world can see, so finance, commercial and leadership all read from the same dashboard.
As a CSS, Redbrain pays Google for clicks now and invoices the merchant for those clicks 30 to 60 days later. A merchant that fails owing a billing cycle of spend takes a real bite out of the P&L.
The CSS working-capital cycle is brutal: cash out daily to Google, cash in monthly from merchants. Standard credit-insurance products don’t price CPC receivables cleanly. Begbies Traynor and Red Flag Alert publish risk scores but they don’t know which merchants are ours, what they owe us today, or whether their payment behaviour has changed.
A merchant rarely fails without warning. The signals are mostly there, but scattered: in the finance team’s aged-debt report, in commercial’s account notes (“they’ve asked to switch to monthly billing”), in trade press, in Companies House filings nobody reads until it’s too late. The work this system does is collect them in one place and surface the ones that matter.
Bad debt from merchant failure is the most preventable hit on the P&L. It only takes us by surprise because the signals were never in the same room.
New merchant pitching to join the CSS programme. Should we take them on at all, and on what terms? Today: a manual check. Tomorrow: a score on day one, with the external signals doing the work the commercial team can’t in 20 minutes.
Existing merchant up for renewal, or a step-change in their spend. Are 30-day terms still appropriate, or should they move to weekly billing? Should we cap their daily spend until trust builds? The score plus exposure data answers it directly.
The hard one. Existing customer, things changing. DSO drifting up. A profit warning in trade press. A CFO resignation at Companies House. The system pings finance and commercial in the same alert — before the merchant becomes a problem.
The unique advantage is that Redbrain has data on these merchants that nobody outside the company can see. The external signals are valuable, but they’re a layer on top.
Days Sales Outstanding by merchant and over time. Late-payment patterns. Term-change requests (“can we move to quarterly?”). Payment-method changes. This is the strongest single-source signal in the system because it’s direct evidence, not inference.
Outstanding CPC spend by merchant, billing-cycle stage, days-to-next-invoice. Combined with the risk score, gives finance an exposure-weighted view: a 90/100 risk score on a merchant with £50 outstanding matters less than a 60/100 score on one with £80,000.
Statutory filings, charges, officer changes, winding-up petitions, late accounts. Free, structured, legally definitive. The earliest external signal — sometimes weeks before press coverage.
Retail Gazette, Drapers, The Grocer, Retail Week, plus mainstream business press. Profit warnings, CVA rumours, store closures, layoffs. LLM-classified for severity and merchant relevance.
Site uptime, careers-page change (job listings drying up), discount depth (sustained 50%+ off across the catalogue). Operational tells from the retailer’s own front end.
Glassdoor on unpaid expenses, Trustpilot on order non-delivery, Reddit on customer service collapse. Noisy and processed last — but corroborating when other signals are already firing.
All six sources feed into the same risk score and the same exposure view. The two internal sources are weighted more heavily because they’re behaviour, not commentary. A merchant whose payments are slipping today is a stronger signal than a merchant whose trade press is wobbly.
Finance, commercial, and leadership ask different questions of the same merchant book. Each gets a view that answers theirs first.
Sorted by exposure-weighted risk. The merchants that could hurt us, ranked by how much they could hurt us for.
Sorted by merchant relationship. Heads-up before a difficult conversation; context for term-tightening discussions.
Sorted by total book exposure. Where the concentration is, what’s moving, how the portfolio looks today versus a quarter ago.
Below is what the finance dashboard would look like with a small sample of merchants. Data is illustrative, not live — but the columns, sort order and action recommendations are exactly what v1 will ship.
Honest accounting. Every module is built and deployable. What’s pending: wiring real Redbrain finance data into the CSV pipeline, a commercial scraping provider for Glassdoor/Reddit, and the first week of tuning against live signals.
All eleven modules are now built. What remains is wiring real Redbrain data into modules 05 and 06 (replacing the seeded CSVs with the live finance feed), adding a commercial scraping provider for Glassdoor/Reddit in module 10, and the inevitable first-week tuning. See build plan for the one-day deploy.