[case] CRM for B2B sales · boiler equipment manufacturer · PASK
CRM for a boiler equipment manufacturer with AI tender analysis — PASK case study
Custom CRM + Prozorro integration + AI tender analysis · PASK
[01] challenge
What the project had to solve
PASK is a boiler equipment manufacturer: block-modular boiler houses (BMBH) and water-heating boilers. Sales run through three channels at once — direct deals, a dealer network of 23 partners across 18 regions, and Prozorro tenders. Each channel lived on its own. Deals and leads — in spreadsheets, with no single funnel. The dealer network — without oversight: who's hitting plan, who's at risk, visible only manually. And someone reviewed Prozorro tenders by hand every day: scrolling through hundreds of listings by CPV codes, eyeballing a pricing strategy. A tender is a complex decision for a manufacturer: does the model fit the spec, what starting and minimum price to set so the margin isn't blown, will we slide into a price war. It all rested on one manager's experience and took half a day per serious listing.
[02] solution
What we did — and why
Deal funnel. Six stages — new lead, qualification, proposal, negotiation, contract, shipment. Cards with the product (TP-500 boiler, BMBH 1.5–2.0 MW), amount, region, manager, and priority. You see the whole pipeline and where the money is stuck. Dealer network on a map. 23 dealers on a country map with tiers (silver / gold / platinum), plan attainment, volume, and status (active / risk / attention / inactive). You see who's carrying the network and who's lagging. Sales breakdown by channel. The dashboard splits volume into direct, dealer, and tender sales — the owner sees which channel actually brings in the money. A single overview. Active deals and their value, tenders in progress, won this year, revenue this month, number of dealers — on one screen. AI inside the system — on tenders. AI Prozorro monitoring automatically pulls listings by the client's profile CPV codes and scores each for fit (as a percentage) with a verdict: recommend / consider / price-war risk / don't recommend. AI tender analysis breaks down the spec and says whether the model fits, analyzes competitors and the average price reduction, computes a pricing strategy (starting and minimum price for a target margin), flags risks, and gives a probability of winning. From the analysis screen — "Build a proposal" and "To deals" buttons: the tender becomes a deal in the funnel without double entry. How we built it: a CRM from scratch for B2B sales (funnel, dealers, tenders, channel analytics), a Prozorro integration by CPV codes, AI modules for tender scoring and pricing strategy, a dealer network map. This is a sales CRM, not an accounting system — no inventory or bookkeeping. Development was led by our developer paired with AI-assisted development.
project frames
[03] result
→ Three sales channels — in one system: direct deals, dealers, and tenders, not scattered spreadsheets → Tenders arrive on their own with a score: AI monitors Prozorro by CPV codes, the manager doesn't scroll listings by hand → A tender decision — in a minute: AI breaks down the spec, a margin-based pricing strategy, and a probability of winning instead of half a day of analysis → The dealer network under control: you see who's hitting plan and who's at risk
Dealers
23
Regions
18
Sales channels
3
Tenders
AI + Prozorro
Need a project like this in your niche?
We'll show how we'd solve it for your business. The estimate is free.
← All cases