FISTA Solutions does not load Google Analytics until you accept. Rejecting keeps optional analytics off. Read the Cookie Policy.

All field notes

Hiring ┬╖ 5 minute read

How to Hire Business Analysts: Signals, Tests and Scope

Business analysts find out what people actually do, translate it into something a team can build, and identify what should change rather than be automated as-is. Test for process discovery skill and willingness to challenge requirements, not for documentation templates or tool familiarity.

By FISTA Solutions┬╖ AI-Native Engineering Team┬╖
How to Hire Business Analysts: Signals, Tests and Scope article cover

Business analysts earn their cost by finding out what people actually do, which is reliably different from what the process document says. Hiring well means testing for discovery skill rather than documentation technique. This guide covers it, drawing on FISTA Solutions' forward deployed engineers work.

What is the core skill?

Process discovery. Documented processes describe intentions; the workarounds describe reality, and systems have to work with reality.

Evidence sourceTells you
Process documentationWhat was intended
Observation of real workWhat actually happens
Spreadsheets beside the systemWhere the system fails
Support ticketsWhere users get stuck
Exception handling volumeWhere the design does not fit

What should you test in an interview?

Ask how they discovered that a documented process was not being followed, and what they did about it.

Strong candidates describe observation and conversation rather than workshops, and describe changing the design rather than enforcing the document. Weaker ones describe gathering requirements in meetings, which captures what people say they do.

Should analysts challenge requirements?

Yes. A requirement that encodes an inefficient process produces an expensive system that perpetuates it.

Ask about a requirement they pushed back on. The job includes asking why a step exists and whether the underlying need could be met differently. Analysts who never push back are transcribers.

Why are workarounds so valuable?

Because they mark where the current system fails people. Every spreadsheet maintained beside an official system is evidence about a requirement nobody wrote down.

Ask what workarounds they found and what they concluded. That question separates analysts who went and looked from those who ran workshops.

How do you evaluate written clarity?

Ask for a sample of their work. Analysis that engineers cannot act on is wasted, however thorough it was.

Look for specifics, explicit assumptions, named edge cases, and a clear distinction between what is required and what is preferred.

What about scope discipline?

Ask how they handled scope growth. Good analysts make the cost of additions visible rather than absorbing them quietly, which is how projects overrun without anyone deciding to overrun.

How does the role differ across contexts?

ContextEmphasis
Enterprise software implementationProcess versus configuration decisions
Product developmentUser needs, prioritisation
Data and reportingMetric definitions, sources
Automation programmesException rates, decision boundaries

State which one you have. Candidates strong in one context can be indifferent in another.

How does AI change analysis work?

It raises the stakes on knowing where the exceptions are. Automating a process with AI means deciding which cases the system handles and which route to a person, and that boundary can only be drawn from real data about how often unusual cases occur.

Analysts who describe only the standard path produce automation that fails on the long tail. See what is an escalation policy.

Contract, staff augmentation, or permanent hire?

Augmentation suits programmes with defined endpoints тАФ an implementation, a process redesign, an automation assessment. Permanent hiring suits organisations where process knowledge accumulates and is worth retaining.

What are the common hiring mistakes?

Screening on documentation templates and tool familiarity. Hiring transcribers rather than analysts. Giving the role no standing to challenge requirements. And measuring on documents produced.

How do you onboard them well?

Give them access to the people doing the work, not just their managers. Managers describe the intended process; the people doing it describe the real one.

What does good look like after 90 days?

A documented as-is process that differs from the official version, a list of workarounds with what each reveals, and at least one requirement reframed or removed.

When do you not need this role?

When the team building the system talks directly to the people using it and has capacity to do the discovery itself. That works well at small scale and breaks as distance grows.

What should be measured?

Rework caused by requirements discovered late, adoption of what gets built, and exception rates against what was predicted.

What should you do first?

Ask three people who do the work to show you how they actually do it. The gap between that and your process documentation is the analyst's first month.

How do you handle stakeholders who disagree?

Most analysis work involves people whose interests differ: the department that wants the process simplified, the one that added the control being simplified away, and the manager who does not want a change at all. Ask how a candidate handled that.

Strong answers involve documenting the disagreement and escalating the decision rather than quietly choosing a side. Analysts who resolve conflicts by picking the loudest stakeholder produce designs that a second department rejects at user acceptance testing, which is the expensive place to discover it.

What about data and volumes?

Ask what numbers they gathered before recommending a change: how often each case occurs, how long each step takes, how many exceptions arise. Analysis without volumes produces recommendations that are plausible and unweighted, and the expensive redesign frequently addresses a case that happens twice a year.

How FISTA Solutions helps

FISTA Solutions embeds analysis alongside engineering through forward deployed engineers and staff augmentation: discovery based on observing real work rather than documented intentions, workarounds treated as requirements evidence, requirements challenged where they encode avoidable process, exception rates measured before automation boundaries are set, and the human-versus-automated boundary defined explicitly through AI agents and AI enablement. The record is 150+ projects for 50+ companies across 12+ countries.

To add analysis capacity, message FISTA on WhatsApp, or read hire scrum masters.

Share-ready article cover

Download the generated social format.

Download cover

Clear answers

Questions raised by this field note.

Straightforward guidance for evaluating scope, fit, and the next step.

01What is the core skill?

Process discovery. Finding what people actually do, including the workarounds nobody documents, is the foundation of useful analysis. Documented processes describe intentions; the workarounds describe reality, and systems have to work with reality.

02What should be tested in an interview?

Ask how they discovered that a documented process was not being followed, and what they did. Strong candidates describe observation and conversation rather than workshops, and describe changing the design rather than enforcing the document.

03Should analysts challenge requirements?

Yes. A requirement that encodes an inefficient process produces an expensive system that perpetuates it. The job includes asking why a step exists and whether the underlying need could be met differently, which is analysis rather than overreach.

04Why are workarounds so valuable?

Because they mark where the current system fails people. Every spreadsheet maintained beside an official system, and every step performed outside the tool, is evidence about a real requirement nobody wrote down.

05What should be measured?

Rework caused by requirements discovered late, and adoption of what gets built. Documents produced measures activity, and thorough documentation of the wrong thing is worse than none.

Start with the hard problem

Need the outcome owned, not merely analyzed?

Tell us where delivery is constrained. WeтАЩll map the fastest credible path from intent to verified production.

Start a project