About
We build the small, unglamorous software that keeps companies coordinated. One workspace, several apps, and a bill that only counts the people who actually open them.
It started with a launch that nearly did not happen
In the spring of 2023, two of us sat in a room watching a product launch come apart four days before it shipped. Nothing had gone wrong technically. The build was fine. The marketing was written. What had gone wrong was that a privacy review nobody remembered to schedule had surfaced on the Monday, and the person who had to sign it off was on a plane.
The post-mortem took ninety minutes and produced a spreadsheet. The spreadsheet lasted nine weeks. By the autumn there were four spreadsheets, a shared calendar nobody trusted, and a recurring meeting whose only purpose was to read the spreadsheets aloud to each other.
We had both worked somewhere that solved this properly, and we had both assumed the solution was cultural. It was not. It was a checklist that generated itself, a queue that told each reviewer what was theirs, and a calendar that everybody else could read without asking. Three small things, wired together, doing the work that the recurring meeting had been doing badly.
Taktoria is those three things, and the handful of others we kept rebuilding at every company we joined afterwards. A Q&A tool for the all-hands where the good question always came too late. Short links, because go/roadmap is faster to say than any URL and it does not rot. Objectives that survive contact with a quarter. None of it is clever. All of it is the kind of thing a company builds badly, internally, three times.
What we believe
Software should be boring where it matters
A launch calendar that surprises you has failed. We would rather ship something predictable than something interesting, and we have deleted features for being interesting.
You should pay for what you use
Per user, per app. If half your company never opens the OKR tool, you are not billed for half your company. This costs us money and we have no intention of changing it.
The work belongs to you
Every app exports. The API does everything the dashboard does, because the dashboard is a client of the same endpoints. Leaving should be a decision, not an escape.
Small teams, small surfaces
We would rather five apps that each do one thing than one app with five tabs. They share a workspace, a bill and a login, and nothing else.
How we work
- We are deliberately small, distributed across three time zones, and we write more than we meet. A decision that cannot survive being written down was not a decision.
- We ship on weekdays, in daylight, with the person who wrote it watching. Nothing goes out on a Friday afternoon, which is less a policy than a scar.
- Every change is reviewed, every test runs on every pull request, and the database migrations are read more carefully than the features. The interesting bugs are never in the features.
About the name
From taktikē, the Greek for the art of arranging things in order. We liked that it described the work rather than the product, and that nobody could tell us it was already taken.
Come and have a look
Start with one app and one workspace. You can add the rest when they earn it.