Use cases
The work changes. The bottleneck does not.
Every company we work with describes the same underlying problem in a different way: the finance work grew faster than the capacity to do it. What differs is where that pressure shows up first — and therefore which part of Accruna matters on day one.
Find your scenario
- Seed to Series A — closing the books yourself
- Series A to C — volume outgrew the team
- Agencies and studios — project margin is invisible
- Ecommerce and subscription — thousands of small lines
- Multi-entity groups — consolidation and one calendar
- Pre-raise and pre-audit — surviving diligence
If volume is the problem
Reconcile
Start with line-level matching, then expand once the ledger is clean.
If control is the problem
Policy
Expense rules and approval thresholds stop leakage at the source.
If time is the problem
Close
The checklist and daily cycle pull the close into the first week.
If confidence is the problem
Trace
Audit trail and workpapers make every figure defensible.
By role
What the engagement looks like from where you sit.
The same platform, different emphasis — depending on whether you are the founder, the controller or the CFO.
You need numbers you can repeat in a board meeting.
You are not trying to become an accountant. You need to know your cash position, your burn and your margin, and to be able to defend those numbers when someone competent asks how they were produced.
- What hurts
- Month-end eats a week of your time, and the output still needs a second opinion before you trust it.
- What changes
- The ledger is maintained without you. Reports arrive reviewed, with the variance already explained.
- What you touch
- Approvals above your threshold, the monthly close sign-off, and questions you want answered.
- What you get back
- The week you were spending on reconciliation, and the confidence to use your own numbers in a decision.
Typical first 90 days
- Weeks 1–2: chart of accounts and opening position agreed
- Weeks 3–5: parallel run — you compare our output to your current process
- Week 6: handover, first close run with you watching
- Month 2: policy formalised, approval thresholds set, receipt capture live
- Month 3: first full quarter reported with variance narrative and a clean audit trail
You need the queue to stop being your entire job.
You already know how to run a close. The problem is that most of your week goes to chasing receipts, fixing coding and re-doing reconciliations that should have been routine.
- What hurts
- Repetitive preparation crowds out the review, analysis and controls work you were actually hired for.
- What changes
- Agents prepare, you review. The exception queue is scoped and pre-annotated, so your time goes to decisions.
- What you touch
- Thresholds, coding rules, policy, exception decisions and the close checklist.
- What you get back
- Capacity. The same team supporting more entities and more volume without more nights.
Controls you keep
- You define what auto-posts and what must be reviewed
- You own the approval matrix and delegation rules
- You can inspect the reasoning behind any agent decision
- You can revert any automated entry, with history preserved
- You sign off the period, not the system
You need capacity without proportional cost.
Every additional entity, currency or product line adds work that scales roughly linearly with headcount. Accruna breaks that relationship for the transactional layer, which is where most of the hours live.
- What hurts
- Hiring to keep pace with volume, while the interesting work — pricing, unit economics, capital planning — waits.
- What changes
- The transactional layer is absorbed. Your team moves up the value chain into analysis and decision support.
- What you touch
- Policy, thresholds, consolidation decisions, audit relationship and reporting design.
- What you get back
- A finance function that scales by mandate rather than by requisition.
Where the hours actually go
- Transaction codingAgent
- Receipt collectionAgent + cardholder
- Reconciliation preparationAgent
- Reconciliation reviewAccountant
- Variance analysisAgent drafts, human owns
- Policy and judgmentYour team
- Strategy and planningYour team
You need the receipt chase to stop landing on you.
In most companies under fifty people, expense admin quietly becomes the operations or office manager's job. It is thankless, it is recurring, and it is entirely automatable.
- What hurts
- Chasing colleagues for receipts, re-explaining the policy, and manually checking claims against rules that nobody can quite remember.
- What changes
- Card feeds and receipt capture do the matching. Reminders go out automatically. Policy is applied consistently, so you stop being the enforcer.
- What you touch
- Approvals within your authority, and supplier or vendor setup requests.
- What you get back
- The weekly expense ritual, and the awkward conversations that came with it.
What stops being manual
- Receipt requests and follow-ups to cardholders
- Duplicate claims caught across amount, date and merchant
- Policy evaluation with the rule quoted back to the requester
- Approval routing by amount, category and cost centre
- Reimbursement validation before it reaches payroll
Scenario 01
Closing the books yourself, without it costing you the week.
You are somewhere between seed and Series A. There is no finance hire yet, the ledger was set up in a hurry, and every month you promise yourself you will keep on top of it — then you do not.
The problem, specifically
- Transaction coding decisions were made once and never revisited, so categories silently drifted
- Receipts live across three inboxes and two card programmes
- Nobody reconciles until a board deck is due
- Investor requests turn into a scramble through statements that do not tie
- You cannot answer "what is our gross margin by product" without a new spreadsheet
What Accruna does with it
Outcome
Investor-ready numbers, without a finance hire.
- Statements that tie, with a reconciliation behind every account
- P&L by product or class, so margin is answerable
- A close report you can hand to an investor or lender
- Spend policy enforced before the money leaves, not after
- An audit trail that starts accruing now, not at diligence
Illustrative shape of the problem: transaction volume commonly grows two to six times before finance headcount moves at all.
Scenario 02
The volume tripled and the team did not.
You are past Series A. The business added products, channels and headcount, and every one of those added lines to the ledger. Your finance team is competent and completely saturated.
Symptoms
- Close drifting past the second week
- Reconciliations done late and at summary level
- Analyst time spent on data cleanup
- Approvals bottlenecked on two people
What Accruna absorbs
- Coding across multiple channels and processors
- Line-level reconciliation on every account
- Receipt capture and duplicate detection
- First-pass variance analysis for review
Where the capacity comes from
Not from asking your team to go faster. From removing the preparation work that sits upstream of their review, so the same people review more without preparing more.
Absorbed
Preparation
Coding, matching, collection, first-pass analysis.
Retained
Review
Judgment, policy, approvals, sign-off — your team.
Freed
Analysis
Margin, pricing, planning and decision support.
Scenario 03
You know revenue per client. You do not know margin per client.
Agencies, studios and consultancies sell time and manage it badly in the ledger. Revenue lands in the billing system, costs land in payroll and cards, and the relationship between them is maintained by hand — if at all.
Where project accounting usually breaks
- Time is tracked in one system and never reconciled to billed revenue
- Contractor and freelance costs are coded to overhead rather than to the project
- Retainers are recognised on receipt instead of over the service period
- Work in progress is estimated rather than calculated, and never trues up
- Pass-through costs are billed without a matching cost record
What correct project accounting makes visible
Margin per engagement
Revenue against its real cost base, including freelance and pass-through spend.
Utilisation against plan
Billable time reconciled to what was actually invoiced, with the gap explained.
WIP that is real
Unbilled work calculated from time and milestones, not from a partner's estimate.
Client profitability trend
Which accounts are getting more expensive to serve, and when it started.
| Engagement | Revenue | Direct cost | Margin |
|---|---|---|---|
| Northwind — retainer | $96,000 | $54,200 | 43.5% |
| Halden Retail — project | $68,400 | $51,900 | 24.1% |
| Presto Labs — retainer | $42,000 | $18,600 | 55.7% |
| Meridian — pass-through | $21,300 | $20,980 | 1.5% |
Halden Retail includes $14,200 of unbilled freelance cost this quarter. Meridian's pass-through is billed at cost with no handling margin — flagged for repricing.
Outcome
Pricing decisions made on evidence.
- Time, cost and revenue joined at the project level
- Retainers and milestones recognised over the service period
- WIP calculated and trued up rather than estimated
- Margin trends visible before they become a cash problem
Scenario 04
High volume, low value, endless to match.
Ecommerce and subscription businesses generate the exact profile of work that is worst for humans and easiest for agents: thousands of small, similar, rule-driven lines.
What a month actually contains
A single processor payout is a net figure that hides gross sales, refunds, chargebacks, processing fees and reserve movements. Manual reconciliation either takes days or gets summarised into an average that hides the detail you need for margin analysis.
| Component | Volume per month | How it should be treated |
|---|---|---|
| Gross sales | Thousands of lines | Revenue, recognised at the transaction date |
| Refunds | Hundreds | Contra-revenue, matched to the original sale where possible |
| Chargebacks | Dozens | Contra-revenue plus a separate dispute cost |
| Processing fees | One per payout | Cost of revenue, allocated by period |
| Reserve movements | Periodic | Asset movement, never revenue or cost |
| Payout to bank | Daily or weekly | Clearing movement, not income |
How agents handle it
- Line-level ingestion of processor transaction reports, not just payouts
- Refund matching back to the original sale, with cross-period refunds handled explicitly
- Chargeback separation so dispute costs do not distort product margin
- Fee allocation to the period and channel that generated it
- Clearing account discipline so the payout never touches revenue
- Deferred revenue maintained for subscription and prepaid orders
Outcome
Gross margin you can trust, by channel.
- True gross revenue separate from net cash received
- Refund and chargeback rates visible as trends
- Fee drag measurable per channel and per market
- Reconciliation that finishes daily rather than monthly
Scenario 05
Five entities, four currencies, one close calendar.
Growth brings structure: a new subsidiary for a market, a holding company for the raise, an acquisition that came with its own ledger. Consolidation stops being an accounting exercise and becomes a project management problem.
Where it goes wrong
- Entities close on different calendars, so consolidation is always provisional
- Intercompany balances do not agree and the difference is plugged
- Exchange rates are applied inconsistently between periods
- Transfer pricing is undocumented and challenged at audit
- Nobody can produce a consolidated cash position on demand
How Accruna runs it
- Every entity on the same ordered close checklist
- Intercompany matching before elimination, not after
- Rates approved once per period and applied consistently
- Translation differences held in their own reserve
- Consolidated statements produced from the same ledger data
The elimination test
Intercompany balances should net to zero. When they do not, the difference is either a timing item, a missing entry, or an error — and each has a different fix. Accruna matches the balances first and shows you the residual with its cause, rather than posting an elimination plug.
- Intercompany matrix reconciled entity pair by entity pair
- Residual differences categorised before any elimination posts
- Elimination entries approved by a named accountant each period
- Consolidated and standalone views generated from one dataset
| Entity | Currency | Revenue | Status |
|---|---|---|---|
| Accruna Client Inc. (US) | USD | $1,284,000 | Closed |
| Client UK Ltd | GBP | £412,600 | Closed |
| Client GmbH | EUR | €508,900 | Closed |
| Client APAC Pte | SGD | S$318,400 | Closed |
| Intercompany eliminations | — | ($186,240) | Matched |
- Reporting currencyUSD
- Rates usedPeriod average + closing, approved Sep 30
- Translation reserve movement($14,820)
- Unmatched intercompany$0.00
Scenario 06
When someone else is about to read your books closely.
A raise, an audit or an acquisition does not create accounting problems. It reveals them, on a deadline, in front of people whose opinion matters.
Raising
Diligence without the scramble
Investors ask for the same things every time: statements, reconciliations, revenue support, and an explanation of anything unusual.
- Statements that tie to the ledger, period by period
- Revenue schedules with contract support
- Reconciliation history for every account
- Anomalies already explained, in writing
Audit
Workpapers that already exist
The expensive part of a first audit is reconstructing evidence. If it was produced as the work happened, the audit becomes a review rather than a rebuild.
- Reconciliations dated and signed within each period
- Journal register with approvals and rationale
- Support attached at the transaction, not the batch
- Access for auditors scoped read-only
Acquisition
A file that survives inspection
Buy-side diligence looks for quality of earnings. The questions are about consistency and support, not about whether the numbers are round.
- Consistent coding policy across periods
- Clear separation of recurring and one-off items
- Revenue recognised on a documented basis
- Full history retained if you leave
Industries
Where the fit is strongest.
We are deliberately not generalists. These are the sectors where the volume, structure and reporting needs match what agents do well.
Fit
Where we are the wrong choice.
Being clear about this saves everyone a wasted conversation — including the ones where the answer is no.
When to look elsewhere
- You need a statutory audit opinion — that requires a licensed audit firm
- Your ledger sits in a bespoke ERP with no export or API route
- You want to prepare the books yourself and only need a review at year end
- Your volume is genuinely low and a spreadsheet is working
- You need consumer tax filing as a one-off, unconnected to a ledger
- You are looking for a software licence with no service relationship
Where we do our best work
- Volume has outgrown the people available to process it
- Multiple systems hold financial activity that does not reconcile itself
- Investors, lenders or a board need numbers on a reliable schedule
- You are preparing for a raise, an audit or an acquisition
- Finance capacity, not finance expertise, is the constraint
- You want an accountable team rather than a tool to operate yourself
Scenario questions
Questions these situations raise.
The ones that come up most often when a finance team recognises itself in one of the scenarios above.
Not necessarily, and usually not immediately. There are three common shapes. Accruna can take over the transactional layer entirely; it can work alongside a bookkeeper who keeps client-facing and administrative work; or it can sit under a controller who reviews and owns the outputs.
What we will be direct about is duplication. If a bookkeeper continues preparing the same ledger independently, the value of the engagement largely disappears, and we would rather say that at the start than take a fee for it.
Seasonality is one of the cases where continuous processing helps most. A December peak is absorbed as it happens rather than arriving as a backlog in January, and accruals for seasonal cost patterns are based on actuals rather than an annualised estimate.
For fee purposes, an engagement is scoped against annual volume rather than the peak month, so a heavy quarter does not trigger a scope review on its own.
Usually yes, and unexpected models are often the most interesting part of onboarding. What determines feasibility is not how unusual the model is, but whether the underlying transactions can be observed and whether the recognition policy can be written down clearly.
If your model needs a policy judgement that has no obvious answer — a multi-element contract with uncertain delivery, for instance — we will flag it during discovery and agree the treatment with you before any period is processed under it.
Often, yes. With low transaction volume and no external reporting obligation, a spreadsheet and a part-time bookkeeper is a legitimate answer, and we will say so rather than sell you an engagement you do not need yet.
The point at which it changes is usually when the first external reader appears — an investor, a lender, a grant body — or when spend spreads across enough tools that reconciliation stops being trivial.
Then that system determines feasibility, and we will assess it before contracting. The four routes are a documented API, a scheduled export, an SFTP file drop, or a bounded manual intake — in that order of preference.
If none of those work, the honest answer is that we cannot serve you well yet. We would rather decline than run an engagement propped up by an unreliable manual step. See the connection routes.
Local statutory reporting and group management reporting are different outputs from the same ledger, and both are supported. Each entity keeps its own books in its functional currency, and the consolidated view is generated from the same data with approved rates and eliminations applied.
Where a local filing obligation needs a specific format, that is scoped explicitly — we produce the schedules, and confirm who prepares and files the return.
Which of these sounds like your month?
Tell us where the pressure shows up first. We will tell you whether Accruna fixes that specific problem, and what it would take to start.