How to choose a development team without regretting it a year later
In Deloitte’s latest Global Outsourcing Survey, 70% of executives said they had brought some outsourced work back in-house over the past five years. Only 25% said their vendors had delivered lower costs or better service. Most of those stories did not go wrong in month nine. They went wrong in the first two weeks, when the team was chosen.
We are a development team, so read this with that in mind. It is the advice we give founders who ask us how to compare us with others, including how to find out that we are the wrong choice for them.
Why the cheapest estimate is rarely the cheapest project
For a startup, money spent on a failed build is runway that does not come back. CB Insights’ 2026 analysis of startup shutdowns found that 70% ran out of capital and 43% never reached product-market fit. Every month spent rebuilding what a previous team got wrong is a month not spent finding customers.
Overruns are not a startup problem only. In McKinsey and Oxford’s classic study of more than 5,400 large IT projects, the average project ran 45% over budget and delivered 56% less value than planned. The research is from 2012, but anyone who has managed a software project recently will recognise the pattern: the estimate was optimistic, the scope was vague, and nobody said so early.
So the useful question is not “who is cheapest per hour” but “who will tell me early when something is going wrong”.
Red flags you can spot before signing
Most bad partnerships give warnings early. They are easy to miss because they look like confidence.
Be careful when:
They say yes to every feature on the first call
The estimate is one number, with no assumptions
You only ever talk to a salesperson
They cannot show code or a live product they built
You will not own the code and accounts from day one
Nobody asks who your users are
A good sign when:
They suggest cutting scope for the first release
The estimate lists what is in, what is out and the risks
You meet the people who will write the code
They show real work and let you talk to a past client
Repository, cloud and app store accounts are yours
They ask about your business before your tech stack
Questions for the first call
You do not need to be technical to ask these. Listen less to the answer and more to whether it is specific.
What would you cut from my first version? A good answer: two or three concrete features, with a reason for each.
Who exactly will work on this, and can I meet them? A good answer: names and roles, and a call this week.
How will I know every week where the project stands? A good answer: a demo of working software, not a status report.
What happens when the estimate turns out wrong? A good answer: when they tell you, and how the options are decided together.
How do you use AI tools, and who checks the output? A good answer: clear rules, with a person reviewing every change.
If we stop working together, what do I get? A good answer: all code, accounts, documentation, and a handover.
Test before you commit
The most reliable way to judge a team is to work with it. Before a long contract, give the team a small, real piece of work with a clear goal: one user flow, one week. You learn more from that week than from any proposal: how they ask questions, how they handle a change of mind, what working software looks like at the end, and whether you want to talk to them every week for a year.
This is also why we offer founders their first MVP week for free: it is the cheapest way for both sides to find out whether we work well together.
What to take from it
Compare honesty, not hourly rates. The team that tells you bad news early is the cheap one.
Own everything from day one. Code, cloud accounts, domains and app store accounts in your name.
Meet the people who will build it. Not only the people who sell it.
Test for a week before you sign for a year. A small real task beats any proposal.
Sign milestones, not roadmaps. Four to six weeks at a time keeps both sides honest.
Want to see how we work before committing to anything? Book a 30-minute call: https://www.commonsaltx.com/contact
