Signed the Engagement Letter? The First Three Weeks Tell You Whether You Hired Expertise or Confidence

The decision is made. A retainer has cleared, a scope document exists in some form, and someone you met twice is now responsible for a piece of your business you cannot easily supervise. The uncomfortable truth about hiring outside expertise is that the sales conversation is designed to be indistinguishable from competence, and in the hands of a capable salesperson it usually is. What separates good from merely plausible shows up later, in the first few weeks of actual work, and it shows up in details small enough to miss if you are not watching for them.

The first scoping call should make the job smaller, not bigger

A specialist who has done your problem forty times arrives at the kickoff with a narrower view of it than you have. They cut things out. They tell you the piece you were most worried about is routine, and that the piece you mentioned in passing is where the cost and the risk actually sit. The plausible version does the opposite: it absorbs everything you raise, agrees that all of it matters, and quietly expands the engagement to cover ground nobody has tested. Expansion in week one, before any diagnostic work has been done, is the single most reliable early signal worth tracking.

Ask directly what they expect to find, and write the answer down. Good practitioners will make falsifiable predictions, because they have seen the pattern often enough to bet on it, and they are comfortable being wrong in a specific way. Vaguer answers are not proof of anything on their own, but they give you nothing to check against in week four, which is the point. You are not testing their confidence. You are building a record you can compare against reality while the engagement is still young enough to redirect.

Reviews read differently once you are inside the engagement

Most people read reviews before hiring, when they have no basis for interpreting them, and never read them again. Go back now. The complaints that seemed like outliers, the ones about slow responses or a partner who sold the work and then vanished, read entirely differently when you have two weeks of your own experience to set beside them. A single angry review from a client who clearly wanted something the firm does not sell is a grudge. Three reviews across two years describing the same handoff failure, in different words, is a process.

The Federal Trade Commission oversees how businesses present endorsements and testimonials, which is worth knowing mainly because it tells you what the ratings on a profile page do and do not represent. A star average compresses a range of experiences into one number and tells you nothing about which end of the range your engagement resembles. What carries information is the texture: whether reviewers describe the same person, the same stage of the work, the same recurring gap. Patterns survive paraphrase. Grudges tend to be specific, singular, and loudly about one incident.

Find out who is actually doing the work, in writing

In professional services the person who wins the business and the person who performs it are frequently different, and that is not inherently a problem. Leverage is how firms stay affordable. The problem is unacknowledged leverage, where you believed you were buying twenty years of judgment and are receiving two years of it with a review step that may or may not happen. Ask for names and roles, ask what percentage of hours each person is expected to bill, and ask who signs off. Firms that run this cleanly answer immediately, because they have the staffing plan already.

Then check the invoices against the answer. The first bill is a document about staffing, not just money, and it will tell you whether the senior person who promised to stay involved logged four hours or forty minutes. If the mix has drifted from what was described, raise it at once, in a short email that asks a question rather than makes an accusation. Early correction is cheap and ordinary. The same conversation in month five, after a pattern has hardened into habit, costs goodwill you will want later.

Judge the first deliverable on its handling of what it does not know

Polish is easy to produce and tells you very little. Read the first real work product for its treatment of uncertainty: whether assumptions are stated, whether the limits of the analysis are marked, whether there is a line somewhere saying that a particular conclusion depends on data nobody has verified yet. Expertise is largely the ability to locate the edge of your own knowledge and say so in a document that a client will read. Work that is uniformly confident across every section is usually confident because nobody checked where it should not be.

The practical follow-up is to pick one claim and ask how they got there. Not adversarially, just as a client who wants to understand the reasoning well enough to defend it internally. Strong practitioners enjoy this and will show their work, including the parts they abandoned. The answer also teaches you something durable, which is the real return on the exercise: after two or three of these conversations you can evaluate the next firm you hire far more quickly, because you know what a real answer sounds like.

Judgment about professionals is built the same way judgment about anything is built, by comparing a prediction you wrote down against an outcome you observed, repeatedly, on engagements that are still running. The first weeks are where that comparison is cheapest, and where a small correction costs an email rather than a renegotiation.