The tax nobody voted for

Monday, 8:40 a.m. The finance director of a mid-sized distributor opens her laptop and, before her first coffee, already has seven tabs open: the ERP, the bank portal, the tax authority, payroll, the sales CRM, the BI dashboard and a spreadsheet called, without irony, “Reconciliation_final”. None of those systems knows the others exist. She's the only one who knows them all.

The scene is a composite, but anyone who has worked in a company of more than fifty people recognizes it. It's the everyday face of a problem that rarely gets named in board meetings: fragmentation.

Nobody decided this

No company sets out to fragment. It arrives by accumulation, and every step makes sense. The ERP comes in to organize accounting. The CRM, because sales needed follow-up. The help desk, because complaints were getting lost in email. BI, because leadership wanted numbers. And spreadsheets, for everything none of those tools quite solved.

The industry has a name for the result, “SaaS sprawl”: applications bought one at a time that nobody designed to work together. The problem isn't in any of them. It's in the gaps between them.

A fragmented company doesn't have a system. It has several, and people acting as the cables between them.

The cost hiding in the hours

Every license has a visible invoice. What appears nowhere is the work needed to make those licenses behave like one operation: someone exports from the CRM, matches it against the ERP, fixes it in a spreadsheet and loads it back. That work has no owner, no budget and no place on any org chart. But it's paid every month, in the salaries of skilled people doing tasks that don't need their judgment.

And it has second-round costs, which are the most expensive. Hand-off errors show up late, once they've reached a leadership report or a customer's invoice. Decisions wait for someone to “pull the information together”. Every new hire needs months to learn five systems and, above all, the unwritten shortcuts between them.

Putting a number on it without hiring consultants

You can estimate a first baseline with a sheet of paper and an afternoon. Pick a process that crosses systems, like month-end close, and record who takes part, the hours spent moving or reconciling data and the cost of that time. Track duplicate licenses, errors and rework separately so you don't count them twice. The total shows resources committed, not guaranteed savings: automation has costs too. The calculator on our home page helps start the estimate.

The point of the exercise isn't to produce a dramatic figure. It's to bring previously separate costs together and compare them with the cost of a concrete improvement.

The way out isn't another tool

One option is another integration tool. It can help if it reduces work and has an owner, or add complexity if nobody governs it. Replacing everything at once can also be costly and risky. The decision needs a scope, a baseline and a migration plan.

What works is less epic and more effective: looking at operations as a whole. Connect the systems that do their job well, build the processes that live in the gaps between them today and retire what only adds friction. One process at a time, measuring every step. It's the idea that runs through this season.

The scenes in these episodes are composites of common situations, not customer cases. Any figures in them are illustrative.