Decide
What actually has to exist for this to be worth doing, and what gets cut to a later phase. This is the part that saves the money, and it is the part a vendor has no incentive to do properly.
This is the technical half of an operating partnership: deciding what software the business actually needs, making sure it gets built properly, and staying responsible for whether it holds up — with the build itself handled by The Creative Collective, the studio I co-run.
Most software fails at the handoff, when the people who understood the decisions leave and the people who inherit them don't. That doesn't happen here, because I'm not leaving — I'm in the business. I make the call on what gets built and what gets cut, the studio builds it, and I'm still the one answering for it a year later.
What actually has to exist for this to be worth doing, and what gets cut to a later phase. This is the part that saves the money, and it is the part a vendor has no incentive to do properly.
The work gets specified and priced with the studio, and I sit on your side of that conversation rather than theirs. My stake is in the business, not in the size of the build.
The Creative Collective builds it — nothing subcontracted, no junior handoff. I stay close to the decisions that come up mid-build, because most of the expensive ones do.
Something working goes live early, and then it is mine to live with. I am not handing this over and leaving; I am one of the people it has to keep working for.
The Creative Collective, a studio I co-run. I am not a development shop and do not pretend to be one — what I bring is ownership of the decisions plus a team I can put on the work without you running a procurement process. If you already have a team you trust, they build it and I do the same job alongside them.
It would be if I were paid for the build, so I am not. My upside is equity or revenue share in your business, which means an unnecessary build costs me and a cheaper answer pays me. I will tell you when the fix is fewer steps and no new software — that is exactly the advice a vendor cannot afford to give you.
Whatever the job needs rather than one house stack. Front end, usually React or Vue in TypeScript, with Next.js, Nuxt or Astro on top depending on how much of the page has to be dynamic. Back end, Node or Python, with Postgres, MySQL or Redis behind it. Deployed on Cloudflare, Vercel or AWS. The constraint I hold to is maintainability: nothing gets built that the business cannot afford to run.
Both, and the thing that matters is being honest about which one you are getting. An MVP is a question you are asking the market and should be cheap enough to throw away. Production software is something the business will depend on and should be built to be maintained. The expensive mistake is paying for one and receiving the other.
I co-build the business model, the metrics that matter, and the first ninety days of execution — then stay through the quarter that proves it works. Sometimes that starts from a napkin. More often it starts from chaos.
Positioning that survives contact with a real market, an identity system your team can run without me, and the campaigns that carry it. I build the thing and hand you the keys — I don't rent you a dependency.
I map what's actually happening — not what the org chart says — cut the steps nobody defends, and leave behind systems that hold under load. Unglamorous work. Usually where the fastest margin is hiding.
A conversation about what you are building and what is blocking it — no charge, no deck. If there is a fit we scope a defined first phase with a clear deliverable and an end date.