What is a discovery phase?
A discovery phase is a short, focused piece of work that happens before anyone writes code. Its job is to turn an idea into a plan you can price, schedule and trust.
Software ideas usually start vague. An HVAC shop owner says, "We need an app." A week of good questions often finds the real problem: jobs get lost between the phone call, the whiteboard and the invoice.
Discovery is where that gets named. You look at how the work moves today, who touches it, where it breaks and what a fix is worth. Then you decide what to build and, just as important, what to leave out.
The same pattern holds for a home services company, a property manager or a job shop. Understand the work first, then size the software to fit it.
What should you get out of discovery?
A discovery phase should end with documents your whole team can read, not only developers. At a minimum, you should walk away with these:
- A problem statement in plain words: what is broken, who it hurts and what fixing it is worth.
- The people and roles who will use the software, and what each one needs to do.
- The workflows and rules, written down: how a job, order or request moves from start to finish.
- A scope line: what is in the first version and what is deliberately left out.
- A right-sized solution design: the main screens, the data and the systems it connects to.
- A data plan: where your current data lives and how it moves into the new tool.
- A fixed quote for the build, or at least a firm number with its assumptions written out.
- Risks and open questions, named honestly, with a plan for each one.
If discovery ends and you still can't explain what will be built and what it will cost, it did not do its job. A slide deck of ideas is not a design.
Ask who owns these documents before you start. They should be yours to keep, whatever you decide next. The same question applies to the code later, which the guide on who owns the code covers.
Why is discovery paid?
Discovery is real work. Someone interviews your team, reads your spreadsheets and forms, and maps each workflow. They check what your other systems allow and then design the solution.
A free estimate can't include that work. Without it, any number is a guess. Guesses tend to be padded for safety or too low to hold, and you find out which one during the build.
Paying for discovery also buys an honest answer. When the developer is paid for the thinking, there is no pressure to make every project look like a build. "You don't need custom software" becomes an answer they can afford to give.
And you keep the design. A paid discovery should hand you a problem statement, a design and a quote that you can take to any developer, or build in-house.
Think of it as a small, fixed cost to learn before you commit to a larger one. If the numbers don't work, you stop there with a clear picture and nothing half-built.
How long does a discovery phase take?
For a small or mid-sized business, one to four weeks is a sensible range. The length follows the scope: how many roles, workflows and integrations are involved, and how much data has to move. As a rough guide:
| Scope | Example | Discovery length |
|---|---|---|
| One focused tool | A job board for a machine shop that replaces a spreadsheet and paper travelers | About one week |
| Several roles or an integration | A maintenance request portal for tenants, vendors and the office, linked to accounting | About two weeks |
| A larger system, built in stages | Scheduling, visit notes and billing for a home-care agency | About four weeks |
What stretches discovery is rarely the design itself. It is waiting: for a decision-maker, for sample data, or for another vendor to explain what their system can share.
Be careful with a discovery that has no end date. A fixed length and a fixed price keep everyone focused on decisions. For the build itself, see how long custom software takes.
How the discovery credit works.
At Lagom Studio, Discovery is a fixed-price engagement of one, two or four weeks, set by scope. It starts at $1,500. It produces a clear problem statement, a right-sized solution design and a fixed quote to build it.
If you sign a Build within 30 days, the whole Discovery fee is credited against it. Not part of the fee: all of it. If you go ahead, the thinking simply counts toward the build.
Builds start at $8,000, with fixed scope and fixed price agreed before any code is written. See full pricing for the details.
Why credit it at all? Discovery should be a decision point, not a sales cost you have to win back. We are paid for the work either way, so we have no reason to push you toward a build you don't need.
Why 30 days? The design and the quote describe your business as it is now. After a month, priorities can shift, and the quote deserves a fresh look.
If you don't build, or you build with someone else, the design is still yours to keep.
When does discovery tell you not to build?
A good discovery is allowed to end with "don't build". That is not a failure. It is the cheapest good news you will get.
Here are honest reasons to stop:
- A better spreadsheet will do. A five-person office tracking a few hundred records may only need one shared file, drop-down lists and locked formulas.
- An off-the-shelf tool fits. If your process is standard, such as booking, invoicing or basic field service, a mature subscription product may do nearly everything you need.
- The problem is a process, not software. Sometimes the fix is a clear owner for each step, or one form instead of three.
- The numbers don't add up. If the build costs more than the problem does over a few years, wait or choose something simpler.
Custom software tends to earn its keep when the workflow itself is what makes you different. It also fits when you pay per seat for several tools that don't talk to each other, and your team retypes data between them.
If subscriptions are the question, add them up with the subscription cost calculator. If spreadsheets are the question, read signs you've outgrown spreadsheets.
One useful question for any developer: when did your discovery last end with "don't build"? A team that never says it may be selling, not advising.
How do you prepare for discovery?
You get more from discovery when you arrive ready. None of this needs to be polished.
- Name one decision-maker. Someone who can say yes or no to scope questions within a day or two.
- Gather real samples. The spreadsheets, paper forms, reports and emails your team uses today, with real data where you can share it.
- List the people who touch the process. The office, the field and the customer. Plan a short conversation with each.
- Write the pain in your own words. What goes wrong, how often, and what it costs you in time, errors or lost work.
- List the software you already pay for. Include accounting, scheduling and anything the new tool may need to connect to.
- Sort must-haves from nice-to-haves. The first version should solve the main problem well, not everything at once.
- Know your dates. A busy season, an audit or a subscription renewal can shape the plan.
If you're comparing prices, the custom software cost guide explains what moves a quote up or down.
Not sure discovery is the right next step? Book a free problem review with Lagom Studio. It is 30 minutes with no pitch deck, and the goal is an honest view of your options.
