Ask
28
@shut_it_down ·

Shut down a four year side project last month and the relief was the surprise

Four years, about 60 paying users at the end, roughly enough to cover hosting and a takeaway. Not a failure. Not a business either. It sat in the middle where it was too alive to ignore and too small to matter, which I now think is the worst place a project can be.

I had been telling myself for eighteen months that I would get back to it properly. In practice that meant fixing things when they broke, apologising for slow support, and feeling vaguely guilty every single weekend.

Emailed everyone, gave three months notice, exported everyone's data, refunded the annual plans pro rata. Half a dozen people were genuinely sad which was harder than I expected. Two asked if they could buy it, which I had not considered at all.

Six weeks later and the main thing I feel is that I got a weekend back that I had not realised was gone.

12 answers Share
Report

Answering anonymously, a moderator will review it first.

  • @sunk_cost_aware · 2w ago · 3 replies

    Going to offer the other side, gently. Three months notice and pro rata refunds is the right way to do it, and it is also expensive in a way that stops a lot of people from ever closing anything.

    I have two projects I would shut tomorrow if closing them were free. It is not free, so they limp on, and that is a real trap rather than a character flaw. Worth naming for anyone reading this and feeling bad about themselves.

    20
    Share
    Reply

    Answering anonymously, a moderator will review it first.

    Report
    • @still_paying_hosting · 2w ago · 2 replies

      Four dead projects here, all still billing me monthly, all because turning them off properly is a weekend I never have. The hosting is cheaper than the weekend, which is a genuinely stupid equilibrium to be in.

      14
      Share
      Reply

      Answering anonymously, a moderator will review it first.

      Report
      • @shut_it_down · 2w ago

        Mine took about two days total, spread over three weeks. Not nothing, but much less than the eighteen months of low grade guilt it replaced. Whatever you are imagining it costs, it is probably less.

        1
        Share
        Reply

        Answering anonymously, a moderator will review it first.

        Report
  • @eleven_users · 2w ago · 3 replies

    The two people who asked to buy it. Please tell me you followed up.

    I closed something with a similar profile and did not, and a year later I found out one of them built a near copy and is doing fine with it. Not bitter about it exactly, but I had a buyer and I threw the thing away instead, and that was purely because shutting down felt like the decision I had already made.

    26
    Share
    Reply

    Answering anonymously, a moderator will review it first.

    Report
    • @shut_it_down · 2w ago · 2 replies

      I did follow up, sort of. One went quiet. The other wanted it for free with me on call for six months, which is not selling it, that is the same job with extra steps and no money.

      19
      Share
      Reply

      Answering anonymously, a moderator will review it first.

      Report
      • @eleven_users · 2w ago

        Yeah, that is not an offer. Fair enough.

        11
        Share
        Reply

        Answering anonymously, a moderator will review it first.

        Report
  • @maintenance_mode · 2w ago

    "Too alive to ignore and too small to matter" is doing a lot of work in this post and it is the most useful thing I have read about side projects in a while.

    Everything I own is in that band. The guilt is not proportional to the work, it is proportional to the number of open loops, and a project with 60 users is one enormous permanently open loop.

    23
    Share
    Reply

    Answering anonymously, a moderator will review it first.

    Report
  • @restarted_thrice · 2w ago · 3 replies

    How did you know it was closing rather than another restart? I have restarted the same thing three times and each one felt like a decision at the time.

    17
    Share
    Reply

    Answering anonymously, a moderator will review it first.

    Report
    • @shut_it_down · 2w ago · 2 replies

      The test that worked for me: I asked what I would do if a new feature request came in from a happy user. Not "would I build it", but what my gut did when I imagined the email arriving.

      My gut sank. That was it, really. When I had restarted before, the same thought made me open the editor.

      21
      Share
      Reply

      Answering anonymously, a moderator will review it first.

      Report
      • @restarted_thrice · 2w ago

        That is unpleasantly clarifying and I already know my answer.

        1
        Share
        Reply

        Answering anonymously, a moderator will review it first.

        Report
  • @maintenance_mode · 2w ago

    Exporting everyone's data before announcing is the detail I would steal. Announce first and support turns into a month of individual export requests.

    1
    Share
    Reply

    Answering anonymously, a moderator will review it first.

    Report
  • @still_paying_hosting · 2w ago

    Sad users are the bit nobody warns you about. Sixty people is not a market but it is definitely sixty people.

    1
    Share
    Reply

    Answering anonymously, a moderator will review it first.

    Report