Most fintechs don't set out to become groups. They start as one company, get a licence in a second jurisdiction to reach a new market, spin up a holding entity ahead of a raise, or acquire a smaller player and inherit its books. Each step makes sense on its own. Eighteen months later, the founder is looking at four or five entities, several currencies, intercompany balances nobody fully reconciled, and a board asking for "the group numbers" — a thing that doesn't actually exist yet.
Consolidation isn't a formality bolted on once you're big enough. It's the mechanism that turns a collection of separately-managed entities into a single, coherent business that investors, regulators, and the founder can actually see clearly. The question isn't really whether to consolidate — it's when the informal version stops being good enough, and what doing it properly actually requires.
When single-entity accounting stops working
There's rarely one dramatic trigger. It's usually a combination of signals arriving around the same time.
- More than one legal entity is trading, not just holding. A dormant holding company doesn't force the issue. Two or more operating entities — each invoicing, paying staff, holding cash — do.
- Intercompany transactions exist. Once one entity is charging another for services, recharging costs, or lending it working capital, single-entity numbers stop telling the truth about the group's real position. Revenue in one entity might just be a cost in another, netting out at group level.
- A regulator, investor, or lender asks for consolidated numbers. This is often the actual forcing function. A Series A term sheet, a banking covenant, or a group-level capital requirement will ask for something the current setup can't produce on short notice.
- The founder can no longer answer basic questions quickly. "What's our total cash position across the group?" or "What did we actually make last quarter, net of intercompany noise?" — if these take days and a spreadsheet reconstruction rather than minutes, consolidation is already overdue.
A reasonable rule of thumb: the moment a second entity starts generating its own P&L that matters to the business, start building the consolidation habit — even informally — rather than waiting until a board or regulator forces it.
What consolidation actually requires
Consolidated accounts aren't simply the sum of every entity's numbers stacked together. Done properly, three things have to happen.
1. Eliminate intercompany transactions
If Entity A pays Entity B a management fee, that's revenue for B and a cost for A — real at the entity level, but meaningless at group level, since money simply moved within the same economic unit. Left unadjusted, intercompany activity inflates both revenue and costs, and can materially distort margins.
This requires:
- A clear intercompany transaction log — not just journal entries buried in each entity's ledger, but a schedule that ties every intercompany charge to its mirror entry in the counterparty entity.
- Elimination entries at consolidation, removing intercompany revenue, cost, and balances so the group P&L reflects only transactions with genuine third parties.
- Reconciliation before consolidation, not after — a mismatch between what Entity A recorded as payable and what Entity B recorded as receivable is far easier to fix at source than to explain to an auditor months later.
2. Handle non-wholly-owned entities and minority interests correctly
Not every entity in a group is 100% owned. Where a subsidiary has other shareholders, consolidated accounts need to show the group's share of results distinctly from the minority interest's share — both on the P&L and the balance sheet. Founders who skip this, or approximate it, end up with numbers that overstate what actually belongs to the parent.
3. Apply one consistent accounting policy across the group
Individual entities — especially ones acquired rather than built in-house — often arrive with their own accounting conventions: different revenue recognition timing, different depreciation policies, different treatment of the same cost categories. Consolidation requires restating everything onto one consistent basis before the numbers can be meaningfully added together. Skipping this step doesn't just create inaccuracy — it can hide real performance differences between entities inside numbers that only look uniform.
Currency: the complication founders underestimate
A multi-jurisdiction group is rarely a single-currency group. Consolidating entities that report in different currencies introduces a layer most founders haven't dealt with before:
- Translating each entity's results into a single group reporting currency, typically using period-average rates for the P&L and period-end rates for the balance sheet — not a single blended rate, which distorts both.
- Currency translation reserves — the balancing entry that captures the effect of exchange rate movement on the group balance sheet, which is not the same thing as a real trading gain or loss and needs to be reported separately.
- Intercompany balances in different currencies, which need to be revalued consistently, or they generate phantom FX gains and losses that make individual entity performance look better or worse than it is.
Getting this wrong doesn't just create noise — it can materially misstate whether the group is actually profitable in its own reporting currency.
Choosing a structure before choosing software
Founders often reach for consolidation tools before the underlying group structure is actually clear. That's backwards. Before any system decision, three structural questions need firm answers:
- What is the parent entity, and does every subsidiary's ownership chain actually match the cap table and legal documentation? Structures drift — an entity gets set up quickly for a specific deal or licence and the group chart is never formally updated to reflect it.
- What is the reporting hierarchy? Not every entity needs to report directly to the ultimate parent; sub-groups (a regional holding company, for instance) may consolidate at an intermediate level first.
- What is the group's chart of accounts? Every entity needs to map its own ledger to a single, shared chart of accounts, or consolidation becomes a manual translation exercise every single period.
Only once these are settled does it make sense to decide whether consolidation happens in a spreadsheet (fine for two or three entities, briefly) or in a proper consolidation module within the accounting system (necessary well before it feels necessary).
Governance: who actually owns the consolidated number
A recurring failure mode in growing groups: consolidated accounts exist, but no one is clearly accountable for their accuracy. Each entity's local bookkeeper or finance person owns their own numbers; nobody owns the group view end to end.
A workable model needs:
- A single owner for the consolidation process — someone who understands every entity well enough to sanity-check the combined output, not just mechanically add the numbers together.
- A consistent close calendar across entities, so consolidation isn't waiting on one jurisdiction that reports two weeks later than the rest.
- A review step above the person who prepared it — consolidated numbers feed board reporting, investor updates, and sometimes regulatory submissions, and deserve the same scrutiny as any other externally-facing number.
The bottom line
Group structuring isn't a milestone a company hits and then finishes — it's a discipline that has to keep pace with how many entities exist, how they transact with each other, and who's actually asking to see the combined picture. The groups that get it right treat consolidation as infrastructure, built ahead of the point it's urgently needed, rather than as a scramble triggered by the first investor or regulator who asks for numbers the business can't yet produce.
The earlier that infrastructure goes in — clean intercompany records, one accounting policy, one owner for the consolidated view — the less painful every future round, licence application, or audit becomes.