Co-Sourced Accounting: The Hybrid Model in 2026

What Co-Sourced Accounting Means

Co-sourced accounting is a hybrid model where an outside accounting team works alongside your internal finance staff instead of replacing them. The internal team keeps decision rights over the numbers. The outside team supplies capacity and technical depth at the layers the internal role can’t reach, usually transaction processing, the GAAP close, and specialist reporting.

It sits between doing everything in-house and handing the whole finance function to a provider. You’ll also see it written as accounting co-sourcing, co-outsourced accounting, or a hybrid accounting model. The terms are interchangeable, and the firms that publish on the model all define it on the same axis: who holds authority over the numbers.

The cleanest way to see where the line falls is to break the finance function into its four layers.

• Bookkeeping. Transaction coding, bank and credit card reconciliation, AP and AR processing.
• GAAP accounting and close. Accruals, deferrals, revenue recognition, the month-end close, and the reviewed financial statements that come out of it.
• FP&A. Budget, forecast, variance analysis, scenario modeling, board and investor reporting.
• CFO advisory. Capital strategy, pricing and unit economics, fundraising and diligence support, board-level counsel.

Those four layers are modular. Your in-house lead already occupies part of the stack, and finance as a service is the same stack bought as a single engagement. Co-sourcing buys depth only at the layers the internal role leaves open.

None of this argues you shouldn’t have hired. The Bureau of Labor Statistics projects about 115,300 openings for accountants and auditors each year from 2024 to 2034, many of them created by people leaving the occupation entirely. Hybrid delivery, meanwhile, is now the default at scale: Deloitte’s 2025 Global Business Services Survey found that roughly 65% of organizations include outsourcers in their delivery model, with finance leading multifunctional scope at 94%. Read both numbers as context for the person you already employ. They can’t be backfilled quickly, and their week shouldn’t go to reconciliations.

Indinero delivers those four layers as one engagement from one team, which is exactly why we treat them as modular rather than all or nothing.

The Three Ways Companies Split the Work

Co-sourced accounting splits three ways in practice: transactional work out, close and compliance out, or everything below the finance lead out. Which one fits depends on what your internal hire was actually hired to do, not on what a provider happens to sell.

Pattern A. Transactional work out, review in-house

AP processing, AR and billing, cash application, bank and credit card reconciliation, payroll journal entries, and expense coding move outside. Review, approval, and sign-off stay in. Your controller reviews the reconciliations, approves the payment run, and owns the final numbers.

This is also how a thin finance team gets real segregation of duties. Where headcount can’t separate preparation, custody, and authorization, the recognized substitute is a compensating control: independent review, external accountant review, and supervisory sign-off. Moving preparation outside while keeping review inside creates that separation without adding a seat. If you’re still deciding whether the processing layer belongs outside at all, the cost side of that question sits in the comparison of outsourced bookkeeping versus an in-house bookkeeper.

Pattern B. Close and compliance out, FP&A in-house

The full month-end close moves out. Accruals and deferrals, revenue recognition, the reviewed GAAP statements, sales tax and income tax compliance, audit prep and the prepared-by-client schedules. The forward-looking layer stays in: budget, forecast, variance commentary, pricing and unit economics, the board deck, and the investor conversation.

This is the most common configuration between $1M and $20M in revenue. It fits the VP Finance who was hired to build a forecast and spends the first ten days of every month producing a trial balance instead. Under this pattern the internal lead consumes the close rather than producing it.

Keeping FP&A inside should still be a deliberate choice about proximity to the operating team, not a reflex about what feels too strategic to send out.

Pattern C. Everything below the controller out, only the finance lead internal

Bookkeeping, accounting and close, technical accounting, tax, and structured FP&A production all move outside. Layers one through three, effectively. One person stays, usually titled VP Finance, Head of Finance, or Controller, and that person owns decision rights, internal business partnering, and the relationship with the board and the bank. If the title shapes how you scope the seat, settle the difference between a controller, a comptroller, and a CFO first.

It fits venture-backed and PE-backed companies that want one accountable finance voice in the room without building a department. It’s also the pattern that concentrates institutional knowledge in a single seat, which is a risk worth naming out loud before you choose it.

Describe each pattern by what your internal person owns, not by what the provider does. The ownership list is the real agreement. Everything else is scheduling.

Where the Split Breaks Down

Accounting co-sourcing fails in predictable ways, and nearly every failure traces back to a boundary nobody wrote down. The model rarely breaks on capability. It breaks on assumption.

Shadow accounting and the reconciliation you pay for twice

The classic pathology is the internal team quietly keeping a parallel set of records to check the provider’s work. Fund administration has a name for it, shadow accounting, where the client duplicates the book of record as a control on the administrator. The cost is plain enough. The team spends its month reconciling unofficial records against the actual books instead of analyzing either one. If your finance lead is rebuilding a reconciliation to verify it, you’re paying twice for one balance, and the second payment is the expensive one, because it comes out of the most senior calendar in the department.

Review that nobody performs

This is the costliest failure and the least discussed, and there’s a formal control concept sitting underneath it. Service organizations design their controls on the assumption that the client performs certain controls on its own side. The AICPA calls these complementary user entity controls, defined as controls that service organization management assumed, in designing the system, would be implemented by user entities and are necessary, in combination with the provider’s own controls, to meet the service commitments. A SOC 2 report lists them explicitly for that reason. The plain version: the provider’s control design has holes only you can fill. When each side assumes the other closed them, nobody did.

A close calendar with two owners

Two parties, one deadline, no single owner. APQC’s General Accounting benchmarking data, published in March 2018 and drawn from roughly 2,300 organizations, put the median monthly close at 6.4 calendar days from trial balance to completed consolidated statements, with the top quartile at 4.8 days or fewer and the bottom quartile at 10 or more. A split close with no named calendar owner lands in that bottom quartile by default, because the metric counts calendar days and every handoff adds a wait. Day-numbered sequencing is the fix, and the month-end close checklist for SaaS companies is a working starting point for building one both sides share.

The single internal bottleneck

Pattern C concentrates this risk, though it can appear in any of the three. Every question the external team has routes to one person, and that person has a day job that isn’t answering them. Growth then stalls at the ceiling of one calendar. The irony is that co-sourcing is supposed to reduce key person risk by moving process knowledge into a team that documents it. Designed badly, it does the opposite, because the internal lead becomes the sole interpreter between the business and the books. Watch for the tell: outside questions sitting unanswered past day three of the close.

Two more failures show up early enough to design against. Undivided system access, where both sides post to the same general ledger with identical permissions and no approval routing, means the split exists only on paper. Mismatched expectations about presence are the other, where a company expected a full-time person at a desk and bought a defined scope on a defined calendar. That one usually surfaces in month two. Scope left undefined is also where fees escalate, because every unlisted task becomes a negotiation.

Co-Sourcing vs Handing Over the Whole Function

Co-sourcing keeps decision rights inside the company. Handing over the whole function moves execution and ownership outside, usually with an external finance lead attached to the engagement.

So the question isn’t really the org chart. It’s what the internal role is for once execution moves outside. Three answers hold up.

• Decision rights. Someone inside the company signs off on the numbers and carries the consequence of them.
• Business partnering. The internal lead sits in the operating meetings and hears commercial context before it becomes a journal entry. No outside team gets that proximity by contract.
• Continuity of institutional knowledge. Why revenue is recognized the way it is, why that one customer contract is nonstandard, what the board was told last quarter.

There’s a fourth answer, and it’s the indefensible one: protecting a headcount. If the internal seat exists because removing it would be awkward, the arrangement produces duplicated work rather than divided labor. That’s cheaper to name now than to discover in month six.

We’re deliberately not rerunning the cost comparison here. That math lives in outsourced accounting versus an in-house team, and this page assumes the hire is already made.

The sharpest test of the model is the scenario nobody writes about. What happens when your internal lead resigns? Fully in-house, the resignation takes the close with it. Co-sourced, the outside team already runs the process, holds the documentation, and knows the chart of accounts, so the close keeps running while you refill the seat. That protection only holds if you avoided the shadow-book pattern and the provider genuinely owns the process rather than shadowing yours.

It matters because a fast backfill isn’t a safe assumption. Accounting degrees awarded fell 6.6% year over year to 55,152 in the 2023 to 2024 academic year, and new CPA exam candidates dropped from 42,626 in 2023 to 28,082 in 2024, according to AICPA trends data reported by the Journal of Accountancy. When you weigh co-sourcing accounting services against a full handover, continuity is the question to ask first, and headcount is the one to ask last.

What to Settle Before the First Close

Settle five things before the first co-sourced close, not after it: who reviews, approval thresholds, the calendar, the system of record, and the escalation path.

1. Who closes and who reviews, written down task by task. The standard instrument is a RACI, assigning Responsible, Accountable, Consulted, and Informed for every close task. The discipline specific to a split close is that exactly one party is Accountable per task, and your complementary user entity controls go into the same document so the review obligations are visible instead of assumed.
2. Approval thresholds and the delegation of authority. Payment release, journal entries above a dollar threshold, new vendor setup, credit memos, and chart of accounts changes. Where a small team can’t fully separate duties, supervisory sign-off and independent review are the substitute, which makes the thresholds a control rather than a formality.
3. One shared close calendar with one owner. Day-numbered, both sides on it, every task carrying a named owner and a due day. Since close cycle time is measured in calendar days including waiting time, handoff latency is the single variable separating a fast close from a slow one. The SaaS month-end close checklist gives you a sequence to adapt rather than invent.
4. One system of record and a divided permission set. Decide whose instance the general ledger lives in before anyone gets a login. In a co-sourced arrangement the people are external and the technology stays internal, so you keep continuous access to your own data and can monitor the work as it happens. Permissions should mirror the RACI, so preparers can’t approve and approvers don’t post.
5. An escalation path with response times attached. A named first contact, a named second, and an acknowledgment window at each tier. An escalation path with no time expectation is a suggestion, not a control.

Run these through a phased transition with a parallel period and defined exit criteria, commonly across the first 90 days, so the first live close isn’t the test. Clear scope is also the cheapest cost control you have, since undefined work is what turns a fixed arrangement into a variable one. If you’re building the budget alongside the operating agreements, the cost anatomy of a full finance engagement breaks down what each layer typically carries.

How Indinero Co-Sources Around an In-House Lead

Indinero gives you the full finance function, bookkeeping through CFO advisory, without managing three separate vendor contracts and timelines. On a co-sourced engagement you buy those layers at the depth your in-house role leaves open, not as a package that swallows the seat you already filled. The same four layers, scoped to the gap.

• Under Pattern A, we run the bookkeeping layer and your controller reviews, approves, and signs off. Preparation moves, authority doesn’t.
• Under Pattern B, we run bookkeeping, the GAAP close, and tax on a monthly cadence. Your VP Finance keeps the forecast, the variance story, and the board conversation.
• Under Pattern C, we run layers one through three with CFO advisory available on tap, and your finance lead holds decision rights and the outside relationships.

Three facts matter more than the rest for a finance leader dividing labor with an outside team. The work is CPA-led and GAAP-first, which is what makes the review layer defensible when your auditor asks who prepared a schedule and who approved it. Indinero is SOC 2 compliant (2026), which speaks to the system access and control questions that decide whether a split close is auditable. And with continuous operations since 2009, 500+ regular customers, and 100+ years combined team experience, the process keeps running when your internal seat turns over.

The rhythm is year-round rather than seasonal. A monthly close, advisory on tap between closes, and one team behind all of it, so you’re dividing work with a single counterparty instead of coordinating a bookkeeper, a CPA firm, and a fractional CFO across three calendars. You can see the full scope on our accounting services page.

Keep your finance lead where they’re strongest. We’ll take the layers underneath at whatever depth the role leaves open, and a free consultation is the place to work out where that line sits.

Frequently asked questions

Here are the questions finance leaders ask most often when they’re weighing a co-sourced arrangement against keeping the work in-house or handing the function over entirely.

Is co-sourced accounting cheaper than a fully outsourced engagement?

Typically no, co-sourced accounting costs more in total than a fully outsourced engagement, because you keep a salaried finance seat alongside the external scope. The reason to co-source is control, proximity to the operating team, and continuity of institutional knowledge, not savings. With indinero you buy bookkeeping, accounting, tax, and CFO advisory only at the depth your in-house role leaves open, so the external scope stays sized to the actual gap.

Who owns the month-end close when the work is co-sourced?

In a co-sourced arrangement one named party owns the close calendar, and every task on it carries exactly one accountable owner. Split closes drift into the bottom quartile of close speed when nobody holds the calendar, because cycle time counts calendar days and every handoff adds a wait. On an indinero engagement we run the close on a monthly cadence against a day-numbered calendar both sides share, with your finance lead reviewing and signing off.

Does co-sourcing work if our only finance hire is a bookkeeper?

Yes, though the split inverts, with your bookkeeper keeping transactions in-house while an outside team supplies the GAAP close, tax, and advisory above it. The caveat is decision rights, since a bookkeeper doesn’t carry authority over revenue recognition or the reviewed statements, so a founder or CEO ends up reviewing what comes back. Indinero covers accounting, tax, and fractional CFO advisory in one CPA-led engagement, so you can add the layers above the bookkeeper without hiring for each one.

How do you stop two teams from doing the same reconciliation twice?

Assign one accountable preparer and one reviewer per reconciliation in writing, so neither team rebuilds work the other already owns. The failure pattern is shadow accounting, where the internal team quietly keeps parallel records to check the provider’s work, and you end up paying twice for one balance. Indinero works from a shared day-numbered close calendar and a permission set that mirrors it, so preparers can’t approve and approvers don’t post.

Can we co-source now and hand over the whole function later?

Yes, and a co-sourced start usually makes the later handover safer, because the outside team already owns the process and holds the documentation. The move is usually incremental, transactional work first, then the close and compliance, then everything below the finance lead. It works only if the outside team genuinely owns the process rather than shadowing yours. Indinero runs all four layers under one engagement, so widening scope is a change in depth, not a new vendor search.

Who answers the auditor when the work is split across two teams?

Name one party as the auditor’s single point of contact for the request list before the audit starts, usually the internal finance lead. The prepared-by-client schedules can still come from the outside team, but one named person routes the requests, tracks what’s outstanding, and answers for the response. Indinero’s work is CPA-led and GAAP-first, and we’re SOC 2 compliant (2026), so when the auditor asks who prepared a schedule and who approved it, the trail holds.

What does an in-house finance lead do once the close moves outside?

An in-house finance lead keeps decision rights over the numbers, partners with the operating team, and carries the institutional knowledge no contract transfers. Under an indinero engagement the lead consumes a reviewed close instead of producing one, which frees the calendar for forecasting, variance commentary, the board deck, and the bank relationship. If the seat exists only to protect a headcount, you get duplicated work instead of divided labor.