How to Evaluate a Software Studio Without Getting Burned

Why this is a hard decision to get right

Picking a software studio is one of the highest-stakes decisions a non-technical founder makes. Get it right and the next twelve months are about product growth. Get it wrong and the next twelve months are about damage control.

I have been on both sides of this. As a founder who has hired the wrong studio. As the studio that has been hired to clean up after another one. There is a pattern to how good evaluations go and a pattern to how bad ones go. This post is the playbook I wish someone had given me the first time I had to make this call.

The signals that matter

Most founders evaluate studios by looking at the wrong things. Logo reels. Generic case studies. Number of employees. Office locations. None of those tell you whether the team you are about to hire can actually deliver your product.

The signals that matter are quieter and harder to fake.

Who is actually on your project

Ask, in writing, who will be on your team. Names. Roles. Their LinkedIn. Their GitHub. The seniority spread.

If the studio is reluctant to commit to specific people, that is the answer. They are selling you a brand and assigning whoever has capacity that month. The studios who will name their team are the studios who actually have a team to name.

How the senior people use their time

Most studios will put a senior on the pitch. The real question is what that senior does once you sign.

Ask: how many hours a week will the most senior person on this project spend on it? If the answer is less than ten, you are buying the brand of the senior and the work of the juniors. That is not the same product.

What happens after launch

The classic dev house contract ends at handoff. You get the code. You get a 30-day warranty. After that you are on your own.

Studios that are confident in their work stay around. They will offer ongoing support. They will commit to a SLA on critical fixes. They will tell you what observability they leave you with.

If a studio cannot answer "what does month three look like" cleanly, that tells you what their model is built for. And it is not your long-term success.

How they handle scope changes

Every project changes scope. Every single one. The question is whether the studio handles it through honest renegotiation or through the change-order industrial complex.

Ask how they handle the case where you need to pivot a feature mid-build. The answers separate them quickly. The good ones describe a weekly checkpoint where scope drift gets caught early and either traded against time or pushed to a later phase. The bad ones describe a change-order process that sounds suspiciously like a profit center.

What their internal tooling looks like

This is the question almost no one asks, and it is one of the most predictive.

Studios with serious internal tooling, deployment pipelines, monitoring, AI-augmented workflows, in-house design systems, ship faster and cheaper than studios without it. Their cost structure is just better. They can pass that on or pocket it. Either way, you should know which kind you are talking to.

Ask: walk me through your internal stack. The studios who have invested in their tooling will light up. The studios who have not will give you a vague answer about "best practices."

How they answer hard technical questions

If you have a CTO or technical advisor, get them on a call with the studio's lead. Ask the hard questions. Why this database. Why this framework. What happens if traffic spikes ten times. How do you handle data residency. What is your eval strategy for the AI features.

You are not testing whether they have a perfect answer. You are testing whether they can think on their feet, admit what they do not know, and reason through tradeoffs in front of you. The studios that fake confidence on hard questions are the studios that will fake confidence on hard incidents.

The red flags

Some signals are immediately disqualifying. Walk away from any of these without a second thought.

They will not put you in touch with past clients. Three references. Real names. Real conversations. If they cannot produce that, do not hire them.

They lead with rate. Studios that compete on price compete on price for a reason. The reason is rarely "we are very efficient." The reason is usually "the work is junior, the support is thin, and the references would not back us up."

They cannot show you working software they built recently. A live URL. A demo account. Something you can poke at. Slide screenshots are not proof of work.

They commit to your timeline before they have scoped your project. This one shows up constantly. A studio that promises six weeks before they have understood the requirements is a studio that will quietly extend to twelve and blame the change orders.

They do not have an opinion. A studio with no opinion on your stack, your architecture, your roadmap, is not a studio. It is a contractor for hire. Sometimes that is what you want. Mostly it is not.

The questions worth asking on the first call

If you only get an hour, here are the questions that buy you the most signal.

1. Tell me about the last project you killed mid-build. (Tests honesty and judgment.)
2. Walk me through a hard technical decision you made on a recent project and why. (Tests technical depth.)
3. Who, specifically, will be on my team? (Tests the bait-and-switch risk.)
4. What does month three after launch look like? (Tests post-launch commitment.)
5. What is your internal tooling stack and what do you build versus buy? (Tests their operating model.)
6. What is the engagement you would not take? (Tests their integrity.)

The wrong studios struggle with most of these. The right ones answer them in their sleep.

A note on price

I am going to say something that sounds self-serving and is just true. The cheapest studio is almost never the right answer.

Software is the kind of work where the cost of getting it wrong is multiples of the cost of getting it right. The studio that costs forty percent less and takes three times as long to ship something half as good has cost you the same money and a year of your runway.

Pay for senior. Pay for ownership. Pay for studios who use AI to actually compress cost rather than studios who pretend AI is a magic wand.

The math always works out the same way.

How to make the final call

Once you have done the diligence, the final call almost always comes down to one question.

If this project goes sideways at week six, do I trust this team to tell me the truth about it.

That is the call. Not the price. Not the timeline. Not the brand. The trust.

The studios you should hire are the ones who would rather lose the deal than lie to win it. Those are also the studios most likely to ship what you actually need.