Time lost switching between systems

Switching between systems costs far more time than the few seconds of changing a tab, because every switch costs attention and overview. You fix it by getting your systems to talk to each other, so AI takes over the retyping and searching and people only make the decision. Below you'll read where it goes wrong, what AI does in practice, roughly what it costs and when it isn't worth it yet.

Why does switching between systems cost so much time?

An employee at an SMB rarely works in one system. The email is open next to the CRM, the accounting runs in a separate tab, WhatsApp Business pings in between and an Excel file is still kept for the planning. Every time someone jumps from one application to another, something gets lost: attention, time and sometimes data.

The switching feels small, but it's the build-up that hurts. Research on task switching shows that people need time after every interruption to get fully back into a task, and that productivity drops noticeably as a result. The exact figures differ per study and per role. But on a working day with several systems and lots of interruptions, there's a lot less time left for the work that really matters.

Why does this hit SMBs extra hard?

Large organisations often have specialists per system. At an SMB, one person often does everything: makes quotes, sends invoices, answers customer questions and keeps the planning up to date. That person doesn't switch between systems twice a day but dozens of times. The problem is then usually not too little capacity but too much fragmentation.

Which hidden costs do you not count?

Nobody puts switching costs in the budget. There's no line for time lost searching in the wrong system. Yet you see the effect in other figures: longer lead times for quotes, more errors in invoices because data is entered twice, and customers waiting longer for an answer because an employee first has to search three systems.

There's a quality side too. Work that keeps getting interrupted is often more superficial and more error-prone than work you finish in one go. For administrative tasks like invoicing or updating customer details, that's a real risk of mistakes that take time to fix later.

What does AI actually fix here?

AI doesn't solve the switching problem by adding another tool. It does it by getting systems to talk to each other, so people have to switch less often. Three uses:

• AI agents that move data across automatically. A new lead in your email is created as a contact in the CRM, with no copying and pasting.
• An AI assistant that consults several systems at once. Instead of searching three systems for a customer's status, an employee asks a question and the assistant pulls the answer from CRM, email and accounting.
• Workflow automation between existing applications. A quote request automatically sets up a task, a draft quote and a calendar appointment, with nobody switching between four screens.

Automation like this often lives in AI agents that run in the background and only step in when a person has to make a decision. You can read more about connecting systems in the integration layer: connecting systems in SMBs.

Important: AI isn't set up separately from your existing systems. It's built on top of them or in between. Most SMBs don't want to replace their CRM, accounting package or planning tool. They want those systems to finally talk to each other. That's fundamentally different from adding another dashboard, which in the long run creates more switching moments.

How do you go from switching around to one workflow?

Most companies don't start with a big rebuild of their systems. They first map where the most switching happens and then pick up one connection or automation at a time.

Step 1: map the switches

Have an employee note for a whole day which systems they switch between and why. It often turns out that most switches revolve around three or four fixed combinations, for example email to CRM, CRM to accounting and WhatsApp to planning.

Step 2: connect the systems that are used together most

Step 3: let AI take the in-between steps, not the decision

The goal isn't for AI to take over every decision. It's for AI to do the switching work, so a person only has to do the last, substantive part. That difference matters for buy-in: people accept automation sooner when they keep control.

Not sure where to start? A short AI scan shows which system switches cost the most time.

What does it cost to get systems working together?

The cost depends on how many systems you connect and how messy the source data is. The amounts below are indicative.

These are estimates based on typical SMB projects, not a guarantee. Custom work, outdated systems without an API or lots of manual exceptions can push the cost up a lot.

When isn't it worth it yet?

Not every company benefits from this approach.

• The volume is too low. If you switch between two systems twice a day, the time you save is probably smaller than the investment.
• The systems are about to be replaced. A connection to a system that disappears in six months is money thrown away.
• Processes aren't fixed yet. Connections work best on processes that have been stable for a while. If they change every week, the automation is out of date sooner than it pays off.
• There's no API or export option. Some older systems don't allow a connection without expensive custom software. Then you need a system replacement first, not an AI project.

If the switching is occasional, the best fix is sometimes just a better routine.

Automation isn't a one-off snapshot either. A connection that fits perfectly now may need adjusting in a year because a system gets an update or a supplier changes. So besides the build cost, leave some room for upkeep.

What's a good next step?

Want to know where in your organisation the most time is lost to switching? Start with a conversation about your current systems and work processes. Book an intro call, or first read how AI consultancy works.