We do not do big year-end retrospectives. This is a short one.
2021 was the first year that felt like we had a stable process rather than a series of projects we were figuring out as we went. That is not nothing, four years in. Here is what we saw.
What 2021 Produced
We shipped work for clients in four countries. That was new. Previously our international work had been through referrals from local clients with overseas connections. This year we had clients in Australia and Hong Kong who found us directly, and one engagement in Europe that came through a partner referral. The quality bar for international work is different - communication expectations are higher, documentation requirements are stricter - and meeting that bar made us better at our local work too.
We added managed services to our offering, which we had resisted for a long time because it felt like it would pull focus from the build work. The reality was that clients who had worked with us on a build needed reliable support afterward, and the options they had before us were either expensive or unreliable. Managed services turned out to be a natural extension rather than a distraction.
We also got more comfortable saying no. Not to clients - to scope. Earlier in our history, we had a habit of accommodating feature requests that were not in scope because the relationship felt like it required it. This year we held the line more consistently, and the relationships were better for it, not worse. Clients respect clarity about what a project is.
What We Got Wrong
We took on one project that we knew was underspecified at the start and hoped the discovery process would fix it. It did not fix it fast enough, and we spent too long in a feedback loop with a client who changed requirements mid-build. The result was delayed and more expensive than it should have been.
The lesson is not new - we have learned it before. Underspecified briefs do not fix themselves. If we cannot nail down requirements in discovery, we should not proceed to the build. We talked about this principle but did not enforce it strictly enough on that engagement.
We also spent too much time on a proposal that did not convert. Long proposals do not win more work. They take more time to write and they shift the conversation toward negotiating document contents rather than understanding the problem. We are shortening our proposals in 2022.
What 2022 Needs to Look Like
Three things.
Tighter scoping discipline. No more "we will sort it out in discovery." If the brief is genuinely unclear after a second conversation, we slow down rather than proceed.
More written process. We have good process in our heads. Some of it is in documents. More of it needs to be in documents, both for onboarding and for consistency across projects when multiple people are involved.
Fewer, better proposals. Quality over quantity. We want to write proposals we are confident will convert, not proposals we are sending as volume.
That is 2021. Four years in. Here is to the next one.