Signs You’ve Outgrown QuickBooks (and What Comes Next)

Last updated: August 7, 2026

By Colin Boothe, CIO at Foretopia

You have outgrown QuickBooks when the number of people required to answer one business question keeps climbing. Not a revenue number. A headcount-per-answer number. The ledger stays accurate the whole time. It just stops being the place anybody looks first, and the spreadsheets quietly become the real system.

A CEO asked a question in a Thursday meeting. “Did we make money on the Restoration order?”

Four people wrote it down.

Nobody answered. Writing it down is a different activity, and by the following Tuesday there was a file in my inbox called Restoration_Margin_v4_FINAL_JS.xlsx. Four days. Three people touched it. The number in cell F22 was off by about nine thousand dollars because a freight credit had landed in the wrong period, and that only surfaced because the controller opened an AP aging for an unrelated reason on Wednesday.

QuickBooks was fine, by the way. That’s the part people get backwards.

It did exactly what it was built to do, which is keep an accurate general ledger for a company that has one of most things. One entity. One warehouse. One way of making money. The moment you have three of any of those, the ledger stays correct and stops being useful, and those two failures get filed under the same complaint by everyone who has never had to separate them.

Where I’m standing, so you can weight the rest accordingly. I run an ERP and business systems consulting group. The back half of this article describes work my company gets paid to do, which is a real conflict that I’m not going to wave away with a disclaimer sentence. So instead I’ll be specific about the cases where the answer is don’t switch. Those are the ones I get wrong in my own favor when nobody is watching, including me.

Worth naming too, since you’re on a staffing company’s blog. KORE1 doesn’t sell software and didn’t buy this post. Their ERP recruiters fill a typical search in about 17 days, and 92 percent of those placements are still in the seat a year later. Hang onto that. The sentence “we need a new system” turns out to be the sentence “we need a person” with better lighting more often than my industry likes to admit.

The Signs Everyone Lists Are Lagging Indicators

Outgrowing QuickBooks means the system can still record what happened but can no longer tell you what is happening. The books close. The decisions move to spreadsheets. Those spreadsheets become the real system of record, and the software you pay for turns into a very expensive filing cabinet with good search.

Search the phrase and you’ll get the same eight bullets from a dozen accounting firms. Slow close. Spreadsheet sprawl. Multi-entity pain. Inventory you can’t trust. Reporting that needs a human translator.

All accurate. All useless.

Every one of those is a lagging indicator. Polite way of saying it reports on a line you crossed eighteen months ago. If your close currently takes eleven days, the interesting question isn’t whether you’ve outgrown your accounting system. You have. The better question is what you were supposed to have noticed in month four.

Three things happen before any of that shows up in a close calendar. Somebody starts exporting the same report every week and fixing it by hand. A second person builds a workbook the first person doesn’t know about. And then, this is the tell, a third person asks the first two which version is right, and gets an answer that starts with “well, it depends what you mean by.”

That’s the moment. Everything after it is cleanup.

The Ceilings Intuit Publishes and Almost Nobody Reads

Here’s something that comes up in maybe one evaluation out of ten, which is strange, because it’s published, free, and takes four minutes to check against your own account.

QuickBooks Online has hard caps. Not performance degradation, not “it gets slow.” Caps. Intuit publishes them, and those four numbers decide more architecture than most CFOs realize.

Forty combined classes and locations is the one that quietly wrecks people on Plus. Six warehouses and eight product lines, and that’s fourteen gone before anybody in the room has said the word department. When the space runs out, the workaround is always identical. Fold two dimensions into one, name the result something like WH3-Retail, and accept that nobody pulls those apart again without rebuilding it by hand. Which somebody then does. Every week. In Excel.

Two hundred fifty accounts sounds generous until a controller who came up in manufacturing starts encoding department and product line into the account number, at which point you burn through it in a quarter and start reusing accounts for things that are not the same thing.

Advanced drops the account and dimension caps entirely and takes users to 25. Genuinely. If a ceiling is your only real problem, upgrading the tier is a far better trade than a nine-month implementation, and I’d rather you put that money into the subscription than into me.

Desktop is a different conversation with a date attached. Intuit’s service discontinuation policy ended support for the 2023 versions after May 31, 2026. That deadline is behind us now. Your file still opens, so nothing feels broken, but payroll tax tables, bank feeds, payments, and security updates are gone, and the security updates are the part that should bother you rather than the bank feeds. Intuit also stopped selling new Pro Plus and Premier Plus subscriptions to U.S. customers back in 2024. Enterprise wasn’t included in that change, which is a distinction that gets lost in roughly every article on the subject.

Five Things That Actually Mean You’re Done

Not eight. Five, in the order I’d weight them.

Two Systems Disagree and Your Team Has a Favorite

I could have stopped at this one. It’s the whole article. When your ecommerce platform says one revenue number and QuickBooks says another, and the finance team has a settled opinion about which one to trust, you no longer have a system of record. You have two systems and a folk tradition. Somebody has to pick. Nobody ever does.

The Close Has Become a Project With a Name

APQC’s Open Standards Benchmarking for general accounting puts the median monthly close at 6.4 calendar days. Top quartile lands at 4.8 or less. Bottom quartile needs ten days or more. Double the top. That data is drawn from thousands of organizations across every industry and it is not QuickBooks-specific, so treat it as a rough shape rather than a target, but the shape is useful. Ten-plus days means your finance team spends a full third of every month producing numbers about a month that already ended, and the strategic work everybody says they want from finance happens in whatever is left.

Somebody’s Job Title Is a Workaround

Every company has this person and can name them in about two seconds. There is somebody whose actual daily function is moving data between two systems that were supposed to talk to each other. Their title says Staff Accountant or Operations Analyst. Their job is being an API. And they’re fast at it, which is exactly the trap, because competence hides the risk right up until the week they take PTO and the numbers stop moving. Single point of failure, and nobody ever wrote it down anywhere.

You Cannot Answer “Did We Make Money on That”

Per job. Per SKU. Per customer. Per entity. Pick whichever unit your business actually runs on, ask the question cold in a meeting, and time the answer. Under an hour, you’re fine. Four days and a spreadsheet with a version number in the filename, you already know.

Consolidation Happens in Excel

Two or more entities, separate QuickBooks files, and a monthly workbook that combines them with intercompany eliminations typed by hand. Everybody in this position knows it’s fragile. Nobody schedules the fix. Almost nobody moves until an auditor, a lender, or a diligence process makes it somebody else’s problem, and that timing is unfortunate, because you end up buying an ERP under deadline pressure from whoever answers the phone fastest.

Three or more of those and the conversation is no longer about whether. Two, and you have some room. One, and you probably have a process problem that a new system would inherit and make more expensive.

When You Have Not Outgrown It

My industry is bad at this section, so let me be useful.

Every article you read on this topic quotes a revenue range. Ten million to fifty million. Fifteen to thirty. Somebody will tell you five. The ranges are all describing the same thing, which is who tends to buy, and none of them are describing who needs to. Revenue is a terrible predictor. A $60M services firm I know runs clean on QuickBooks Advanced right now. One entity, one revenue model, a chart of accounts somebody actually maintains. Meanwhile a $12M ecommerce brand carrying three entities and inventory in two 3PLs was genuinely underwater at a size where the internet says it should have been comfortable.

Complexity is the variable. Revenue is just correlated with it, loosely, the way shoe size correlates with reading level in elementary school.

So here are three cases where I’d tell you to keep what you have.

Your chart of accounts is a mess and nobody has cleaned it in four years. Fix that first, before anything else gets a budget line. A rebuilt chart of accounts with real dimensions costs you a controller’s attention for three weeks, and roughly a third of the time it dissolves the reporting complaint entirely. Migrating a bad chart of accounts into a new platform is how you pay six figures to be confused in a nicer interface.

Second case. One department is loud and everyone else is fine. If operations is miserable and finance is not, buy operations a tool. Just buy it. A real WMS, a real project management system, a real billing platform. Point solutions get an unfair reputation from consultants who’d rather sell the platform, and there is a legitimate version of the bolt-on strategy that buys you two or three good years. The seams are the catch. Every bolt-on adds one more place for two systems to disagree in public, and that failure mode has its own rules, which is the entire subject of unifying disparate systems without point-to-point wiring.

Third, and this one is about timing rather than fit. You are eight weeks from a funding round, an audit, a peak season, or an acquisition. Don’t start. An implementation during any of those is a coin flip, and I’ve been on the losing side of that flip and had to sit in the room afterward.

What Comes Next, and What It Actually Costs

Three real paths. One of them gets written about constantly and the other two get dismissed as unserious, which is backwards.

On the third path, the candidate list is shorter than the marketing suggests and you already know most of it. NetSuite, Sage Intacct, and Microsoft Dynamics 365 Business Central cover the overwhelming majority of the mid-market. Intacct is strong in finance-first, non-inventory businesses. Business Central shows up where there’s already a Microsoft stack and a partner relationship. NetSuite is where I live, and I’d tell you that’s exactly why you should get a second opinion from somebody who doesn’t. If you want to run that comparison without a salesperson in the room, KORE1 built an ERP comparison tool covering NetSuite, SAP, Dynamics, and Epicor, and there’s a NetSuite implementation cost calculator that will get you closer to a real budget than the first quote you receive.

What it takes is less mysterious than the sales cycle makes it feel. We rebuilt the stack at Just Ingredients while they were growing through $150M in sales on systems that had stopped talking to each other, and the integrated environment now supports a business north of $250M with actual visibility across ecommerce and wholesale. About seven months. No magic. Not five years either, and definitely not the twelve weeks somebody will promise you in a demo.

Let’s graduate our tech stack from the 90s to at least the 2000s. That’s the whole ambition here, and it stays out of reach for most companies not because the software is hard but because the process arguments underneath it never get finished. Six weeks of a project like that is software configuration. The rest is four people in a room deciding what a job cost is, and if that argument never resolves, no platform on that list saves you.

Before you talk to anybody selling, run a free ERP readiness assessment on yourself. It takes about ten minutes and it will tell you whether you’re buying a system or outsourcing a decision.

The Chair Nobody Budgets For

Here’s where these projects actually die, and it isn’t in the software.

A new system needs an owner. One name. Not a committee, not a vendor, not the CFO adding it to eleven other things. A person whose name is on it, who understands both the accounting and the configuration, and who is still there in month fourteen when somebody wants to add a subsidiary. Employee or consultant is a genuine fork, and I’m the wrong guy to ask. KORE1 took an honest swing at it in NetSuite consultant versus in-house team.

$81,680 was the median wage for accountants and auditors in May 2024, per the Bureau of Labor Statistics, which also has the occupation growing 5 percent through 2034 on roughly 124,200 openings a year. That’s a stable, competitive market for exactly the people you’re about to need, and the systems-literate slice of it is much smaller than those numbers suggest. A controller who can close books and a controller who can own a dimensional chart of accounts inside an ERP are not the same hire, and the second one is who you’re actually recruiting against.

My part of this ends at go-live. After that it belongs to somebody on your payroll for the next six years, and right now that person is either underqualified or hypothetical.

So the hire matters more than the platform choice, and I say that as someone whose revenue depends on the platform choice. If you’re not sure the role is permanent yet, don’t guess. Put it on contract staffing for two quarters and find out what the job actually is before somebody writes a job description for work nobody in the building has done. If the gap is above that line, a fractional CIO for ERP and digital transformation is a real option and a cheaper one than most people assume. And if you want to see what mid-market companies are actually running before you commit, KORE1 published a 2026 report on mid-market ERP and AI adoption built from five independent datasets.

What People Ask Me Right Before They Switch

Is there a revenue number that forces the decision?

Wrong unit of measure. There is no revenue threshold that forces a move off QuickBooks. Entity count, inventory complexity, and the number of dimensions you need to report on are what force it. The $10M to $50M range you keep reading describes who tends to buy, not who needs to, and I’ve seen $60M companies run fine and $12M companies drown.

How long does the move actually take, start to finish?

Five to nine months for a mid-market company doing it properly, and the software is not what makes it nine. Data cleanup and unresolved process decisions are. The projects that land in five months are the ones where somebody did the chart of accounts work before the partner showed up. The ones that hit fourteen months are almost always waiting on a decision nobody wants to own.

Can we keep QuickBooks for one entity and run an ERP for the rest?

Hybrid setups work for about a year and then quietly stop working. Mechanically it holds up, and you consolidate through a reporting layer or a manual journal. Culturally it’s a slow leak, because the entity left behind gets a different close calendar, different dimensions, and a different definition of gross margin, and reconciling those two worlds becomes somebody’s permanent job. Do it as a transition state with an end date written down. Not as an architecture.

What actually moves over, and what dies in the migration?

Less than you’re picturing, and that’s usually good news. Open transactions, customer and vendor masters, item masters, and trial balances by period move. Full transactional history typically does not, and you keep the old file in read-only mode for lookups. Every client asks for ten years of detail. Almost none of them open it after month three, and paying to migrate it is one of the easier line items to cut from a quote.

We’re on QuickBooks Desktop 2023. Does the cutoff force our hand?

May 31, 2026 already passed, so this is a live problem rather than a planning exercise. The file still opens and nothing looks broken, which is precisely why it gets ignored. No security updates is the real exposure, not the bank feeds. It forces a decision, though, not a destination. Upgrading to a supported Desktop version or moving to QuickBooks Online are both legitimate answers, and neither one requires you to buy an ERP this quarter.

How do I tell whether a partner is scoping this honestly?

Short answer: ask them what you should not buy. A good partner will name a module you don’t need yet and tell you which of your problems is a process problem their software will not fix. If every answer is yes and every requirement maps neatly to something on their price list, you’re talking to a salesperson with a technical vocabulary. Ask what happens in month eight when the person who learned the system leaves. The quality of that answer tells you most of what you need.

Count the Hands

Try this at your next leadership meeting. Ask a real operating question, something like what your margin was on your third-largest customer last quarter, and then count how many people would have to touch something before you get a number you’d repeat to a lender.

One person, one system, same day. Nothing here is your problem. Good news.

Three people and a workbook, and you have your answer, and you also have the beginning of a business case that doesn’t require anybody to say the phrase digital transformation out loud. Four days is not a software preference. It’s a tax. Four days a month you’re paying for twice, once in salary and once in the decisions that got made without the number.

Nobody sends a notification when you cross this line. That is the actual problem with it, and it’s why the question is worth asking on a random Thursday rather than during an audit.

Run the test and hit me up on LinkedIn with what you counted. I’ll tell you which of the three paths I’d take and whether I think you need anybody at all. If it turns out the thing you’re missing is a controller or a systems owner rather than a platform, bring that one to the KORE1 team. Hiring is the part I’m genuinely no help with.