Paid discovery, one week, $1,500, and you keep the document
A paid discovery phase is a short engagement that converts an estimate into a commitment. Ours runs one week for $1,500, credited in full against the project, and produces a fixed fee, a fixed date, a deliverable list and a named owner. If you do not proceed, the document is yours to take to another vendor.
By HubReven
The market offers two options and both are bad.
Free discovery is a sales call wearing a costume. Nobody profiles your data for free, so the "scope" that comes out of it is a template with your logo on it, and the number attached is padded to cover everything nobody looked at.
The other option is a multi month paid strategy engagement that produces a slide deck and no code.
There is a third shape, and it is boring: one week, a fixed small fee, and a document specific enough that a different vendor could build from it.
What the week costs and what happens to the money
$1,500. One week. Credited in full against the project if you proceed.
The credit is the important part. If you go ahead, the scoping week is free in net terms and the only thing it cost you was a week of calendar. If you do not go ahead, we got paid for the work we did and you got a document you own outright.
That last clause is not a formality. You can take it to another agency and get competitive quotes from a real specification instead of from a paragraph. That has happened, and it is a fair outcome.
What actually gets done in the week
Not workshops. The week is mostly us reading things.
We read the code, if there is code. Architecture, dependency health, test coverage, the parts that will fight a change. A description of a codebase and a codebase are different objects.
We profile the data, if data is involved. Duplicate rate, association integrity, date format consistency, whether a stable external ID exists. This is the single biggest driver of cost in migration and integration work, and it is knowable in two days.
We audit the systems that have to talk. What integration surface each one actually offers: a documented API, a nightly file, a database replica. The answer changes the architecture and the price.
We map the process as it runs today. Including the steps people do by hand that are not in any document, which is usually where the actual requirement is hiding.
We agree what is explicitly out of scope. A list of things we are deliberately not building. This is what makes a fixed date possible.
What you get at the end
Five things, in writing:
- A fixed fee. Not a range. The number that does not move unless you change the scope in writing.
- A fixed date. With the sequence that produces it.
- A deliverable list. Specific enough to argue with.
- A named owner. A person, not an account.
- The findings. What we found in your code, your data or your systems, including the parts you will not enjoy reading.
The findings section is frequently the most valuable page. A data profile that says your contact database has a 31% duplicate rate is worth knowing whether or not you hire anybody.
Why one week and not four
Because most of what a longer discovery produces is not decision relevant.
A week is enough to read a codebase, profile a database and map a process. It is not enough to build consensus across eleven stakeholders, and it is not supposed to be. If the blocker is internal agreement rather than technical unknowns, a paid discovery will not fix it and you should not spend money on one yet.
It is also short enough that the cost of being wrong is small. Two weeks of calendar and $1,500 is a cheap way to find out that the answer is buy rather than build.
When you should skip it
Three cases where we will tell you not to bother.
The work is genuinely tiny. A single landing page or a one field mapping change does not need a scoping week. We will just quote it.
You already have a specification. If someone has already profiled the data and written the mapping, we will read it and quote from it.
You are not ready to decide. If budget is not approved or the requirement is still being argued, the document will go stale before you use it. Wait.
How this compares to the alternative
The real comparison is not against free discovery. It is against the cost of starting without one.
A project that begins on a guess resolves its unknowns during the build, at build rates, with the schedule already committed. The scoping week resolves them beforehand, at a fixed small cost, while changing your mind is still free.
If the decision you are actually weighing is whether to do any of this externally at all, the comparison that matters is against a hire, with the assumptions shown.
Frequently asked questions
Is the $1,500 refundable if I do not proceed?
No, and that is the point. It buys a week of real work. What you get instead of a refund is ownership of the document, including the findings, with no restriction on where you take it.
What if the scoping week finds the project is bigger than expected?
Then you learned that for $1,500 rather than during month two of a fixed fee build. Sometimes the output is a recommendation to cut scope, phase the work, or not do it.
Do you need production access?
Read access to the relevant systems, yes. A staging copy works for most of it. We will send the specific access list before the week starts so nothing waits on IT.
Can we run the scoping week and then build it ourselves?
Yes. The document is specific enough to be built from, and it is yours. We would rather you do that than hire us for work you can do internally.
Get the next one