Institutional amnesia: the silent tax on small businesses

Last Thursday, your team spent forty minutes in a meeting deciding whether to switch invoicing from Stripe to GoCardless. You watched the debate unfold and got the strange feeling you had been here before. After the meeting, you went hunting through your inbox. You had. Eight months ago. Same arguments. Same conclusion. Nobody remembered.

That meeting cost you roughly three hundred euros in salaried time. It also cost you the actual decision again, because the person who had done the analysis the first time wasn't in the room this round. Eight months from now, when someone proposes Adyen, the cycle will repeat.

This is institutional amnesia. Not a glamorous problem, not one anyone is hired to fix, and easily the single largest source of slow waste at most small and medium businesses.

The cost is invisible until you count it

A SME of twenty people will, in a typical quarter, re-solve somewhere between fifteen and forty problems it has already solved before. Not big strategic ones. Small ones. Which payment processor. How to handle a chargeback. What we tell suppliers about late invoices. Who can sign off on a refund. What our standard NDA terms actually are.

Each re-solving event costs an hour of one or two people's time, sometimes more. None of it shows up on a P&L. None of it triggers a project review. The total drag is real but distributed across so many micro-events that no one ever sums it up.

A conservative back-of-envelope estimate: at twenty people and the conditions above, you are losing somewhere between thirty and seventy hours of professional time per quarter to problems you already have answers to. At a blended cost of seventy euros an hour, that is two to five thousand euros per quarter of pure tax. And that is just the easy-to-see version.

Amnesia is not the same as forgetting people

Most discussions about "knowledge loss" focus on employees leaving. That is a real problem, but it is a different problem. Institutional amnesia happens even when nobody has left. It happens because:

• The person who solved it the first time is in a different meeting today.
• The decision was made in a Slack thread that scrolled out of view in eleven days.
• The reasoning was captured in a Google Doc that nobody remembers exists.
• Three slightly different versions of the answer exist, and no one knows which is current.
• The original problem was framed slightly differently the first time, so search doesn't surface it.

The point is that information being recorded somewhere is not the same thing as information being accessible to the next person who needs it. The first happens easily. The second almost never does.

Documentation theatre solves the wrong problem

The standard response, when someone notices this pattern, is to launch a documentation initiative. Get everyone to write things down. Build a wiki. Use Notion better. Create runbooks.

These initiatives have a measurable lifespan. The first month, energy is high and the wiki grows. By month three, ten percent of pages are slightly stale. By month six, fifty percent are stale and the rest are written from the perspective of a moment that no longer reflects how the company works. The wiki, far from solving amnesia, has become a confident source of slightly-wrong answers.

The reason this happens is not laziness. It is that documentation is a job people only do when they have time, and the people most likely to know things are the people least likely to have time. The wiki captures what people can write down, which has very little to do with what the next person needs to know.

Where the answers actually live

Your company already has a near-complete record of every decision it has made. It just is not in your wiki. It is in:

• The email thread where the choice got made.
• The Slack channel where the trade-off got debated.
• The call transcript where a supplier explained why option B was worse.
• The handful of messages between two people on Discord that ratified the conclusion.

If you could read those four surfaces together, with context, you could answer almost any "have we discussed this before?" question in seconds. The information is there. The retrieval is what is broken.

This is the central insight of treating company communications as an institutional memory layer instead of as transactional noise. The amount of useful organizational knowledge already encoded in last year's email is enormous. The cost of recovering it has, until very recently, been prohibitive.

The cheap fix is not "remember more"

The instinct is to ask the team to be more disciplined. Document the decision. Pin the thread. Update the Notion page. This works briefly and then stops working, for exactly the reasons it has stopped working every other time anyone has tried it.

The non-cheap fix, which is now possible in a way it was not three years ago, is to stop relying on humans to translate their work into a parallel documentation artifact. Capture the work itself. Read across it. Let people ask "have we made a decision about invoicing before?" and get an answer that points to the specific Slack thread from eight months ago, with the reasoning intact.

This is not exotic. It is what your team would naturally do if any one person had a perfect memory of every conversation the company ever had. Software can be that person.

What stops looking like a tax

When the amnesia tax stops being paid, three small but compounding things happen.

Meetings get shorter, because at least one of the recurring "wait, didn't we already…" detours stops happening. Onboarding gets faster, because the new person can ask their way into existing context instead of waiting for it to be transferred. Senior people get freed up, because the "ask Marco, he was around for that" pattern collapses.

None of this is dramatic. No one celebrates a meeting that runs five minutes shorter. But over a quarter, across a team, the cumulative effect is the equivalent of a part-time hire whose only job is to remember things on behalf of everyone else.

Institutional amnesia is one of the cheapest problems to ignore and one of the most expensive to keep paying for. The bill is not on your balance sheet. That is exactly why it keeps growing.