Seven SaaS subscriptions into one internal platform: how to know when consolidation is worth it

Cancelling subscriptions pays for the build. That's the easy math. The reason it actually sticks is something else entirely — and it's the part most consolidation projects get wrong.

A construction contractor was running nineteen systems. Seven of them overlapped — each doing roughly half of what the next one did, none of them matching how the crews actually worked.

We replaced all seven with one internal platform. The subscription savings covered the build. But that's not why it worked.

The math everyone leads with, and why it's the weak argument

Seven subscriptions at a few thousand a month each is real money, and it's the number that gets a project approved. It's also the least durable justification, because a vendor can always discount to keep you, and someone will eventually point that out in a meeting.

If subscription cost is your only case, you'll lose the argument the moment a rep offers 30% off.

The argument that actually holds

The cost isn't the subscriptions. It's the seams between them.

In that business, a single job touched: intake, estimating, scheduling, field production, safety documentation, fleet, timekeeping, purchasing, AP, and finally invoicing. Seven systems meant six handoffs, and every handoff was a person re-keying data or a spreadsheet acting as glue.

That produces costs nobody puts on a line item:

Reconciliation labor. Someone spends days each month making two systems agree.
Decisions on stale data. If the field system and the finance system disagree, leadership is choosing between two wrong numbers.
The bus factor. One person understands how the spreadsheet in the middle works.
Errors that surface late. A mis-keyed quantity found at month-end close, not at the job site.

You can't get a clean invoice for any of those, which is exactly why they persist.

When consolidation is worth it

I look for all four:

The systems overlap significantly. Not "we have a lot of tools" — genuine functional overlap where the same entity exists in several places.
There's manual glue. A person re-keys, exports, or reconciles between them. This is the strongest single signal.
The workflow is unusual enough that no vendor fits. If an off-the-shelf product genuinely matches how you work, buy it. Consolidation is for businesses whose process is their advantage.
The data model is stable. If the business is changing shape monthly, you'll build the wrong thing.

Miss #3 and you're rebuilding commodity software badly. That's the most common consolidation failure — a company builds their own CRM and discovers why CRMs are complicated.

When it isn't

Anything regulated with a certification you'd inherit. Payroll tax filing, for example. Let a vendor own that liability.
Commodity functions with no process advantage. Email, storage, video calls, accounting ledgers.
When the real problem is process, not tooling. If the handoffs are broken because nobody agreed who owns a step, new software encodes the confusion at higher speed.

What made this one stick

One login. One permission model. One data layer.

The platform spanned finance and treasury, field production, safety, fleet, expenses, intake, and hiring — wired into QuickBooks Online, Bill.com, Samsara, Fleetio, Paylocity, KPA and Microsoft 365 so data moved on its own instead of being re-keyed.

The result the client actually talks about isn't the cancelled subscriptions. It's that finance, the field, and the back office finally run on the same source of truth. When the CFO and the operations lead look at a job's margin, they see the same number. That was never true before.

How to scope it so it doesn't become a two-year project

Don't replace seven systems at once. Sequence by pain:

Start where the manual glue is worst — usually the handoff someone complains about most.
Ship that in weeks, not months. Get one seam closed and running in production.
Keep the old system live in parallel until the new path has run a full monthly cycle.
Cancel one subscription. Now the project has paid for something and you have political capital for the next one.
Repeat.

The version of this project that fails is the one that tries to launch all seven replacements on the same Monday.

The takeaway: consolidate for the seams, not the subscriptions. And sequence it so each step cancels a bill before you start the next.