A shortlist in 72 hours, contributing inside two weeks, and a 30-day guarantee if the fit is wrong.
72-hour shortlist · 30-day replacement guarantee · Month-to-month
When a project overruns, the post-mortem usually blames delivery. Engineering was slow, estimates were optimistic, scope grew. Occasionally that is what happened. Far more often the work was specified in a way that guaranteed the outcome before anybody wrote code.
The mechanism is consistent. Requirements arrive already shaped as a solution — "we need a dashboard", "we need an approval workflow" — rather than as the problem behind them. Nobody asks what decision the dashboard supports or what goes wrong today without the workflow. The thing gets built exactly as described, works exactly as described, and does not fix the problem, because the problem was never written down.
The second mechanism is the happy path. A specification describes what happens when everything works. It does not say what happens when the payment is declined halfway, when the upstream system is unavailable, when a user submits the form twice, or when a record already exists. Those cases turn up in week six as questions, and each one is a decision somebody now makes under time pressure.
A good analyst finds those before the estimate, not after it. That is most of the value of the role, and it is almost never what the job description asks for.
We screen for the habits that stop a specification falling apart in week six of the build.
Whether they ask what problem a requested feature solves, or document the request and move on. The core skill.
Timeouts, duplicates, partial failures and records that already exist. What a specification omits is what stalls the build.
Criteria a developer can build to and a tester can verify, rather than a restatement of the story title.
Resolving contradictory requirements rather than recording both and letting the team discover the conflict.
How quickly they get fluent in an unfamiliar business. We ask about a domain they knew nothing about and how they got up to speed.
A live conversation in English, every time. This role is almost entirely conversation, with people who do not enjoy being interviewed.
The cost of weak analysis is not a bad specification. It is a project that looks fine for two months and then slows down for reasons nobody can name.
Here is the usual shape. Build starts against a set of user stories with acceptance criteria like "user can search for a customer". Reasonable-sounding. Then a developer asks whether search should match partial names, whether it should include archived customers, what order results come back in, what happens with four thousand matches, and whether somebody in one region should see customers from another. Five questions, each requiring a business decision, each blocking the ticket. They go to whoever is available rather than whoever should decide, and the answers get made up.
Multiply that across sixty stories and the project is running on decisions nobody senior ever made. Some will be wrong. They surface at user acceptance testing, which is the most expensive point to discover them, and they arrive as a list of defects rather than as what they are — requirements that were never specified.
Integration work is where this bites hardest. An interface specified as "send the order to the fulfilment system" leaves out everything that matters: what happens if it times out after the order was accepted, whether retrying creates a duplicate, what reconciles the two systems when they disagree. Those are business decisions with financial consequences, and if nobody asks, a developer will choose one at build time and nobody will know which.
Business Analyst (2–4 years). Documents requirements against a defined scope. Runs workshops, writes stories, keeps the backlog coherent. Needs support when stakeholders disagree.
Senior BA (4–7 years). Gets to the problem behind the request, specifies exception paths as a matter of course, and resolves conflicting stakeholder requirements rather than recording both. Owns the acceptance criteria the build is judged against.
Lead BA / Principal (7+ years). Shapes the programme, maps processes across departments, handles the politics of a change that makes somebody's job smaller, and is willing to tell a sponsor their requested solution will not solve their problem.
Discovery before a build. Establishing what is actually needed before committing budget. The highest-return use of the role and the most frequently skipped.
Embedded in a delivery team. Ongoing refinement, acceptance criteria and stakeholder access so developers are not blocked waiting for a decision.
Process mapping. Documenting how work is actually done rather than how the manual says. Usually reveals more than the software requirement did.
Integration and migration specification. Field-level mapping, exception handling and reconciliation rules. Detailed work that prevents the most expensive class of defect.
Enterprise implementations. Where configuration follows process. Common alongside SAP, Dynamics 365 and Salesforce work.
Week one. They talk to the people who do the work, not only the people who commissioned it. Expect them to come back with at least one thing the sponsor did not know — a workaround everybody relies on, a step performed out of order, a spreadsheet that is quietly the real system of record.
Weeks two to four. A written view of the current process with the gaps marked, and requirements that specify what happens when things go wrong. Expect a list of open decisions with names against them and dates by which each is needed. That list is the single most useful artefact a BA produces.
The warning sign is an analyst who only writes down what they are told. Stakeholders describe solutions, and they contradict each other. A BA who records both positions without reconciling them has moved the problem into the backlog rather than solved it, and the developer inherits it.
Hiring a BA to write documents. If the measure of the role is documentation produced, you will get documentation. The measure should be decisions made and questions closed before build. Say that in the brief and interview for it.
Skipping analysis to start sooner. The reliable false economy. Two weeks of discovery routinely removes more than two weeks of rework, and the rework is discovered late where it costs most. Teams under deadline pressure cut this first, which is precisely when it is worth most.
Confusing the role with product management. A BA specifies what is needed and why. A product manager decides what is worth doing and in what order. Related, not interchangeable, and hiring one expecting the other leaves a gap nobody notices until prioritisation stops happening.
Accepting requirements from one stakeholder. The person who commissions work rarely does it. Requirements gathered only from a manager describe the process as understood at management level, which is reliably different from what happens, and the difference is where the exceptions live.
Rates depend on seniority, domain, and engagement length. Tell us the role and we will give you a firm number.
Tell us the role and we will send a shortlist within 72 hours.
Tell Us the RoleTell us the role. Three to five profiles within 72 hours.