On this page
Key takeaways
- A sprint answers "what should this company prove first?" A strategy engagement answers "how should this company run with AI over three years?"
- A sprint costs thousands and takes weeks. A strategy engagement costs six figures and takes months. Both can be right; the order matters.
- Mid-market companies usually get more from proving one workflow than from a portfolio of twenty initiatives.
- Whatever the format, the output should name the number, the owner and the first proof.
- The questions at the end of this post work on either provider.
Last updated: October 2026
An AI readiness sprint is a fixed-scope, two-to-four-week assessment that returns the use cases worth funding first, a business case for each and a 90-day plan. A full AI strategy engagement is a months-long consultancy project that returns an operating model, a target architecture, a portfolio of initiatives and a multi-year roadmap. A €20M–€500M company usually needs the first before it can judge whether it needs the second, because the strategy question is easier to answer once one workflow has been proved. The cost of getting this wrong shows in the adoption data: McKinsey's 2025 survey found 88% of organisations using AI and 7% with it fully scaled.
The two shapes defined
The sprint. A small team spends two to four weeks with the leadership team and the people who run the candidate workflows. It maps the operation, checks the data and systems, and returns a ranked shortlist with a business case and a 90-day plan. The scope and the fee are fixed before it starts. Twistag's version is the AI Readiness Sprint, from €5,000.
The strategy engagement. A consultancy team spends three to six months, sometimes longer, on an enterprise-wide view: ambition and principles, operating model, data and platform architecture, a portfolio of initiatives, governance, a change plan and a roadmap. It is designed for large organisations with many business units and a transformation office to receive it.
Duration, cost, inputs, outputs
| AI readiness sprint | AI strategy engagement | |
|---|---|---|
| Duration | 2–4 weeks | 3–6 months or more |
| Cost | Fixed fee from €5,000 | Typically six figures, time-and-materials or phased |
| Team | Two to three senior people who also build | A partner, managers and analysts |
| Inputs | Interviews with workflow owners, data and system checks, leadership priorities | Interviews across the organisation, benchmarks, market analysis, workshops |
| Outputs | Ranked shortlist, business case for the top use cases, 90-day plan | Operating model, target architecture, initiative portfolio, multi-year roadmap |
| Decision it enables | What to prove first and what to fix before it | How to organise, govern and invest over years |
| What happens next | A proof on one workflow, measured against a number | A programme, often delivered by a separate team |
| Fits best | Mid-market companies, PE portfolio companies, single business units | Large enterprises with a transformation office |
When each is right
The sprint is right when the company has budget, has more AI ideas than it can fund, and has nobody who can say with confidence which one to start with. It is also right for a private equity operating partner who needs the same question answered across several portfolio companies in the same quarter. The sprint format exists because the first step is where most companies stall: ISG's 2025 survey of 400 senior AI decision makers found only 31% of enterprises with a top-three funded use case in full production, and a long strategy phase delays the first proof by months.
The strategy engagement is right when the company is large enough that the operating model is the problem: several business units with their own data, a platform decision that will bind the group for years, a board that needs a documented plan, and a team ready to run the programme afterwards. For a company with 200 to 2,000 people and one or two lines of business, most of that machinery is weight rather than help.
The common mistake is buying the strategy engagement first because it feels thorough. Thoroughness is the wrong measure when nothing has been proved: MIT's 2025 report found that 95% of corporate generative AI pilots stall before scaled adoption, and a longer document does not change the odds for the first one.
What happens after each
After a sprint, the company proves one workflow in six to eight weeks on real data, measured against the number agreed in the sprint. If the number moves, the second workflow builds on the same foundation. If it does not, the company keeps the code, the findings and a clear view of what would have to change, for a few weeks of cost. The sprint's 90-day plan is the strategy, revised each quarter from evidence.
After a strategy engagement, the company stands up a programme: a transformation lead, governance, a platform build, and a portfolio of initiatives sequenced over years. The first proof typically arrives in the second or third quarter, delivered by a different team from the one that wrote the strategy. The strategy is revised annually.
What leadership should ask either provider
- Which number will the first use case move, and when will it be baselined?
- Who on the team has built and run a system like the one being recommended?
- Which data and system blockers were found, and who owns fixing each?
- What is fixed before the work starts: scope, fee, duration?
- Is the recommendation tied to a platform or a vendor, and how was it chosen?
- What does the company keep if the engagement stops after the first phase?
- How soon could a proof start after the assessment ends?
Frequently asked questions
Can a sprint replace an AI strategy?
For most mid-market companies, yes, in the form that matters: a ranked list of where AI creates value here, what has to be true, and what to prove first, revised each quarter. A written multi-year strategy becomes useful once two or three workflows run and the company knows what its AI operation looks like.
Can the two be combined?
Yes, and the sequence is the sprint first. A strategy engagement that starts with one proved workflow and real data on what it took is grounded in the company; one that starts from interviews alone describes the industry.
Why is the sprint priced as a fixed fee?
So the risk is known before work starts. The scope is fixed by the number of teams involved, the fee is agreed up front, and the company decides what to do next with the shortlist in hand, with no commitment to build with the same provider.
