Stop using your team as the integration layer

Spent last week with the operations lead at a mid-market firm, walking through how the team tracks its work. We counted somewhere around 270 lists. About a third of them were dead. Nobody had touched them in months. But nobody had deleted them either, because nobody could quite remember what they were for and nobody wanted to be the person who broke something.

That wasn't the interesting part.

The interesting part was watching how every single live list actually worked. Information arrived via email. Somebody manually copied the relevant fields into a list to track the work. Work happened. When it was done, somebody manually copied the result back into the system of record. Sometimes the same information lived in three places. Sometimes the address was spelled "Street" in one system and "St" in another, breaking the link entirely.

The operations lead, sharp, capable, knows the work cold, couldn't actually pull up a single customer and tell me what had happened to that account across the last 90 days without opening five systems. The data was all there, just shattered across tools that didn't talk to each other. The only thing holding the picture together was her team's collective memory and a lot of double-entry.

This is the SaaS sprawl tax. It's not the licence cost. It's the hidden labour cost of using your humans as the integration layer.

Most mid-market businesses run between 6 and 15 SaaS tools across their core operations. Each one was bought with good intent, usually to solve a specific problem the previous tool couldn't handle. Each one works fine in isolation. But almost none of them talk to each other in any meaningful way.

What happens, predictably, is that humans become the integration layer. Every time information moves from System A to System B, somebody types it in. Every time a status changes in System B, somebody updates the matching record in System A. Every time the spelling drifts, somebody eventually notices the mismatch and fixes it by hand.

The cost is invisible because it's distributed. Nobody sits down and types for four hours straight. They type for 30 seconds, forty-seven times a day, in between doing actual work. Per transaction it's nothing. Per week it's a quarter of someone's job. Per organisation it's three or four FTEs of manual integration nobody has on a spreadsheet anywhere.

Four signs you've got it

1. Your operations lead can't pull a single timeline. If they can't show you everything that happened to one customer, one project, or one asset across the last 90 days without opening five systems, you're paying the tax.
2. The same data is entered in three places. Even if it's "just two clicks" in each, multiply by every transaction and you get real money.
3. You've got a graveyard of dead workflows nobody will delete. Tools nobody uses but nobody removes are the receipt of a previous solution to a previous problem. They survive because deleting them feels riskier than ignoring them. They're a tell.
4. Your team doesn't trust the data. When somebody pulls a report, the first question is always "is this current?" That means somewhere, the system isn't being updated in real time, usually because updating it requires human effort that competes with the actual work.

The fix is rarely as scary as people think

It's not more tools. It's not even better tools. It's getting the existing tools to talk to each other so the humans stop being the API.

Most modern SaaS platforms have decent APIs. The work is mostly: figure out which system is the source of truth for each piece of data, then push it everywhere else automatically. Once you do, three things happen:

• The double-entry disappears, and capacity returns to your team
• The data quality jumps, because there's only one source
• The dashboards finally work, because you can trust what's underneath them

You also start to see things you couldn't see before. The full timeline of a customer. The pattern across last quarter. The bottleneck that's been there for two years that nobody could measure because the data was scattered across three tools.

There's a moment in every engagement where the operations lead sees their first cross-system view and goes very quiet. They've been suspicious for a long time that something wasn't quite right. Now they can see exactly what.

Stop using your team as your integration layer. They have better things to do.