A Nordic fintech closes the books in two days.

The company provides payment cards and spend management to small and mid-sized companies in Norway, Sweden, Denmark and Finland. It runs seven legal entities, settles in four currencies and processes card transactions through its own ledger, with subscription revenue billed through Stripe and booked in Xero.

Closing the month meant reconciling three versions of the truth. The internal ledger in Postgres recorded every transaction, Stripe recorded billing and refunds, and Xero held the accounting view. The finance team exported all three, matched them in spreadsheets line by line, chased differences with engineering and only then started on consolidation and FX. The whole cycle took nine business days, which meant management saw last month’s numbers when this month was already half over.

In the previous year, two board packs had to be corrected after they were sent, both because of manual reconciliation errors. For a regulated fintech preparing for a funding round, that was not a minor irritation.

Because ledger data had to stay in the company’s own EU environment, the team installed Kimo Bridge next to a read replica of its Postgres ledger: queries run through an outbound-only, mutually authenticated tunnel and no ledger rows are stored on Kimo’s side. Snowflake, Stripe, Xero and HubSpot sync into a dedicated Kimo Cloud tenant with SSO and role-based access. Instead of reconciling at month end, the team built a reconciliation model that matches ledger entries to Stripe charges and Xero journals continuously, every hour, with tolerance rules for FX rounding and timing differences.

Unmatched items surface in an exceptions dashboard owned by the financial controller, each with the entity, amount, age and a suggested reason. Most are resolved the same day while the context is fresh, rather than three weeks later. A second model handles consolidation: entity mapping, intercompany eliminations and FX at the Norges Bank month-end rate, all defined once in the semantic layer and versioned.

The board pack itself became a Kimo dashboard: revenue by product and country, net revenue retention, gross margin, cash runway and headcount. Because every number traces back to its definition and source rows, the auditors were given read-only access and stopped asking for screenshots.