Costs are escalating. A high-visibility project is at risk. The board is watching, and someone needs a credible pair of hands, fast. Under that kind of pressure, it's easy to hire the first consultancy that sounds confident. Confidence isn't the thing worth screening for.
A handful of direct questions, asked before any contract is signed, tend to separate the consultancies that can actually help from the ones that only sound like they can.
How fast, specifically
Vague answers, "quickly," "as soon as possible" aren't answers. A real answer is a specific range, held consistently across every conversation. Ours is three to six weeks, every time, because it's a commitment we can hold ourselves to rather than a placeholder. It's worth noticing, too, whether the answer changes depending on who's asking, a consultancy that quotes a tighter timeline to a nervous sponsor than to a sceptical procurement lead is telling you something about how it handles pressure more generally.
Who actually shows up
Many consultancies staff recovery work with junior consultants supervised remotely by someone senior. It's worth asking directly for the specific experience of the person who will actually be embedded, not the firm's average. Every practitioner we embed has fifteen-plus years of experience, because a programme in crisis isn't the place to learn on the job. A useful follow-up question here: ask to speak to the actual person before signing, not an account lead who will hand the work off afterwards. A consultancy confident in who it's deploying will have no issue with this. One that hesitates usually hasn't decided yet, which means the answer to "who's coming" is still genuinely unknown.
Whether the advice is actually independent
If a consultancy sells a platform, a methodology, or a staffing model, its recommendation is rarely fully independent, watch for hesitation when this gets asked directly. Being technology-agnostic means every recommendation serves the client, never a vendor relationship the consultancy is incentivised to protect.
This matters more than it might initially appear, because the bias rarely shows up as an obviously wrong recommendation, it shows up as a consistent, subtle tilt towards whatever the consultancy happens to sell, dressed up in the language of best practice.
What "success" is actually measured against
"High success" isn't a number. It's worth pushing for the figure and the baseline it's measured against. Ours is 94%, measured against programmes that were genuinely at risk when we started, not a cherry-picked baseline. A success rate measured against every engagement a firm has ever taken on, including the straightforward ones, tells you very little about how that firm performs under genuine pressure. Ask specifically what counts as "at risk" in their number, and what counts as "success" on time, on the revised plan, or against the original commitment.
Whether disagreement is welcome
This is the question most often skipped, and the one that matters most under pressure. A consultancy that only ever agrees with its client isn't advising, it's order-taking, and order-takers rarely raise the uncomfortable thing early enough to act on it. Being an advisor rather than an order-taker means saying so when a decision looks like it's heading toward the same failure mode that caused the crisis in the first place, even when it isn't what the room wants to hear.
One practical way to test this before signing anything: ask the consultancy to describe a time they told a client something the client didn't want to hear, and what happened next. The answer tends to be more revealing than any reference call.
What's left behind
Recovery that collapses the moment the consultancy exits was never recovery, it was a temporary patch. It's worth asking directly what capability gets left behind, and who owns it once the engagement ends. A consultancy genuinely focused on handover will have a specific answer to this, named people, specific documentation, a defined point at which responsibility formally transfers. A vague answer here usually means the engagement was designed around the consultancy's continued presence, not its eventual absence.
Benchmarking the answers against reality
It helps to know what "good" looks like at scale before scoring anyone's answers. NISTA's latest Annual Report on the UK's largest, most scrutinised portfolio of government projects, 213 projects worth close to £1 trillion, still shows a majority sitting at Amber delivery confidence, not Green, per the
NISTA Annual Report. That's the benchmark for closely governed and still difficult. Any consultancy claiming recovery is simple, fast, and guaranteed regardless of context is describing something that doesn't match reality at any scale.
Where to start
Agile Minds offers a free, two-week assessment before any of this becomes a commercial conversation.
If your programme needs a straight answer rather than reassurance,
we should talk.
