The Integration Tax

Written by Dave Linderman | Jun 5, 2026 6:00:10 PM

Every disconnected system in your business has a tax attached to it. The tax is paid by someone on your team, in time, every day, and nobody is putting it on the P&L.

Spend some time with any growing SMB, and you'll find the same picture. Sales lives in a CRM. Finance lives in an accounting system. Operations lives in a project management tool. Support lives in a ticketing system. HR lives in a payroll platform. Each of these systems is competent on its own. None of them talks to the others.

Between them is the integration tax. It is paid by the people who copy data from one system into another. The people who reconcile the numbers when two systems disagree. The people who keep a spreadsheet on the side because neither of the official tools gives them the view they actually need, or don't trust the output. The people who spend the first hour of every Monday morning making sure the things that should match actually match.

You are paying this tax whether you measure it or not. The only question is how much.

A 40-person services business closes a sale. The salesperson updates the CRM. The deal needs to get into the accounting system so the customer can be invoiced. There is no integration, so someone in finance manually creates the customer record and the invoice, copying fields from the CRM. The project gets staffed. Operations sets up the project in the PM tool, manually copying the deal information from the CRM again. The customer signs the SOW. Someone has to update the CRM with the signed status, the accounting system with the actual contract value (which may have changed from the deal stage value), and the PM tool with the kickoff date.

That is one sale. The same person's name has now been entered into three systems by three people, each with three opportunities to make a typo. Two weeks later, a junior accountant notices the customer's name is spelled differently in the accounting system than in the CRM. They send a message to find out which is correct. Three people look up. Forty-five minutes of three people's time gets spent verifying a name.

Multiply that by the number of customers you process a year. Welcome to the integration tax.

The tax has three components, and you are paying all three.

The first is labor. People spending time copying data instead of doing actual work. Industry estimates put this at 30 to 90 minutes per knowledge worker per day in companies without integration. For a 40-person company, that is two to six full-time equivalents' worth of work going into copying rather than producing.

The second is errors. Manual data movement is the highest-error part of any operation. A typo in an address means a shipment goes to the wrong place. A typo in a routing number means a wire goes to the wrong account. A misclassified customer means the customer is receiving the wrong tax treatment. These errors are not free. They are just hard to attribute.

The third is decision lag. When the systems do not talk, you cannot ask a question like "how profitable is this customer" without someone manually pulling data from four places and reconciling it. By the time you get the answer, the question will be two weeks old. You either decide on incomplete data or wait for clarity. Both cost you.

What frustrates me is how this gets framed in the market. Vendors sell integration as a technology problem, with a technology product, with a technology price tag. But the actual problem is not technology. It's that nobody is responsible for the data flowing between systems. The salesperson is responsible for the deal. The accountant is responsible for the books. Nobody is responsible for the place in between, and that's where the tax is collected.

Map the data flow. Draw the actual diagram of what information moves from which system to which other system, and by what mechanism. Most SMBs have never done this exercise. It takes a few hours and reveals everything. The places where information is moved by human hands are the places with the highest tax.

Then prioritize the highest-traffic flows for actual integration. Not all of them. The ones where the same data is being typed into multiple systems multiple times a week. The CRM-to-accounting handoff is almost always the top of this list. Pick the tool by volume. Deploy native integrations where they exist. Use lightweight automation for simpler flows. A dedicated integration platform when volume justifies the investment. And consider custom development for the unusual cases. It doesn't have to be an all-or-nothing proposition. But you have to begin somewhere.

And once you've picked your starting point, name a person responsible for the data flow. Not the systems individually. The flow. Someone whose job includes making sure the customer name is the same in all four systems, and that when a deal closes, the right thing happens automatically in accounting and operations within 24 hours. This person does not have to be senior. They have to be empowered to fix things when the flow breaks.

Disconnected systems are not a feature of small businesses. They are a phase. You either move out of that phase intentionally, or you keep paying the tax until it becomes the thing limiting your growth.

Connect the systems, or keep paying the people who connect them by hand. One of those options scales. The other one does not.