FOXNUT AI

Definition

How to choose a consultant, and when not to hire one

Choose a consultant by making each one show judgment on your real case. First check that you can state your problem in one sentence with a number.

By the Foxnut team · Updated

How do you choose a consultant?

To choose a consultant, make the candidates show judgment on your case before you commit, and weigh behaviours over credentials. Four behaviours carry most of the signal: whether they can explain how they would work in plain terms, whether the person you meet is the person who does the work, whether they can say what a defined piece costs before building it, and whether they have ever told a prospect no.

We are Foxnut AI, a senior-only consulting practice in Bengaluru and Paris, so yes, a consultant wrote this guide, and we come back to that below. There’s a link to talk to us at the end of this article. Let’s get your checks in order first.

Decks, logos and certificates are cheap to produce. Those four behaviours are not. The rest of this page makes the checks concrete, and it starts with the prior question most guides skip: whether you should be hiring anyone at all yet.

When should a business hire a consultant?

A business should hire a consultant when it can describe the problem in one sentence with a number attached. That is the clearest readiness signal practitioners describe. “We lose track of leads” is a complaint. “We miss about a third of after-hours calls and never call back” is a project.

A consultant hired against a complaint will spend the engagement defining it, at consulting rates, and a vague brief is how discovery phases breed further discovery phases.

If you cannot write that sentence yet, the cheapest next step is not a vendor call. It is measuring the problem until you can. A free self-run readiness check exists for exactly this test, and passing it is what makes every check below usable, because each one measures the candidate against your number.

What questions should you ask when hiring a consultant?

Ask the questions in the table below. You can run them in one or two calls, and none of them needs technical depth. All of them need the seller to be specific about your case, not just fluent about their method.

CheckWhat to askWhat a good answer sounds like
The how”Walk me through how you would approach our specific number”A mechanism in plain terms, with what they would need from you and what could make it fail
Who delivers”Who exactly does the work, and will I meet them before signing?”Named people, in the room now - not a partner who hands off after the signature
Scoping”What does the first defined piece cost, and what is out of scope?”A price and a boundary before work starts, however small the piece
Refusals”What work do you turn away?”Specific categories, with reasons - a firm that refuses nothing is selling capacity, not judgment
The exit”What do we own and run ourselves when you leave?”Artefacts your team operates without them, named in the contract
Evidence”Show me something I can verify myself”Checkable specifics - shipped work, public records, named mistakes - not testimonials

How much does it cost to hire a consultant?

There is no honest general answer, because it depends entirely on the shape of the engagement. A seller who quotes confidently before understanding your case has answered a different check, badly. What you can legitimately demand is the scoping behaviour in the table: published shapes and ranges, and a price for the first defined piece before it begins.

What are the red flags when you hire a consultant?

The red flags are behaviours, not resumes, and every one worth acting on is the inverse of a check above.

The seller cannot explain the how beneath the second question. The senior person who pitched will not name who delivers, which is the bait-and-switch pattern buyers complain about more than any other.

Discovery is proposed with no priced destination. Nothing has ever been refused. The deliverable, pressed for specifics, turns out to be a document about the work and not the work.

References deserve a flag of their own. They are weak evidence even when offered, because no seller supplies a bad one. That is why the evidence check above asks for something verifiable instead, and why a candid “here is what went wrong on our own projects” is worth more than three arranged phone calls.

Should you trust a guide written by a consultant?

Only if you turn it on its author, because every guide to choosing a consultant is written by a consultant, including this one. Buyers ask for vetting advice “no vendors please” for exactly this reason, and a guide that hides it is already failing its own evidence check.

So let’s turn this one on us. Foxnut AI publishes its own working rules - named delivery, response commitments, and what it refuses, and its shipped record with named mistakes on the record of what Foxnut AI has shipped.

We would fail two checks today, and we say so plainly. We have no publishable named client case study yet. And a two-person senior-only bench means execution volume is a check we fail by design, so if you need many hands, choose a firm shaped for that. The remaining checks, run against the consulting practice that publishes this library, are yours to score.

Which check decides which consultant to hire?

The check that decides it is whether your team will work with the person, and it is not the one guides usually lead with. Our founders have sat in the buyer’s chair. At Viviva Colors, the art-supplies brand Ameya co-built, the website conversion rate had capped out around 2.5 percent against a Shopify industry average near 3, so they went out to buy conversion-rate optimisation help.

The first filter was the obvious one, proof of work: only candidates who had demonstrably done this before. It did not narrow the field enough.

The second filter decided everything. They invited each remaining candidate to a one-hour freewheeling brainstorm with the entire team on the real problem. Within minutes it was clear which consultants would take the team along and get the job done, and which would talk down from a position of authority, deliver the gospel truths of conversion optimisation, and leave it at that.

They hired the candidate without the best credentials: the one most respectful of the team, quickest to take in its suggestions, and honest enough to name his own gaps in this particular market. The team welcomed him, looked forward to the weekly meetings, and went out of its way to implement his ideas, because it respected him as a person as well as an expert. Conversion nearly doubled, past the Shopify benchmark.

The rule the founders now state from that purchase: a consultant has to get under the hood of your organisation, extract truth from the people who do the work, and be welcomed while doing it. So between a highly capable consultant your team will not get along with and a good one it will, choose the second every single time.

A brilliant project poorly implemented loses to an average project well implemented, twice on Sunday. That is the biggest rookie mistake buyers make with this list.

Run this list on us too. The people who sign a Foxnut AI proposal do the work, there is one tier, and we designed, made and launched our own products first, so the person your team meets is the person it will work with. The first step is a free expert audit: a senior practitioner reviews your setup, with no RFP and no pitch deck.

You now have every check in the table above, and the two we would fail today besides. Run all of them on us: book the call and ask.

Foxnut AI takes on work like this from Bengaluru and Paris. If you want to see how we do it before you talk to anyone, here is the shape of each engagement, and its range.