Choosing How to Work Together
Seven engagement models, and how to buy the right one.
You do not buy a van for one delivery. For one delivery you book a courier. For a weekly run you lease something. If moving things is your whole business, you buy the van and hire a driver. Nobody finds this confusing, because everyone matches the model to the use without thinking about it. The cost, the commitment and the control all scale with how much moving you actually need.
Buying behavioural science works the same way and gets muddled anyway. Organisations buy a workshop when they needed a verdict. They commission an audit when they are quietly avoiding a decision. They sign a retainer when one sharp week would have done the job. The work is good and the fit is wrong, and a wrong fit wastes money just as surely as bad work does, only more politely.
Seven models, and what each cannot do
The useful thing to know about each model is not only what it is for, but what it cannot do: because the limits are where the mismatches happen.
- Advisory
- Senior judgement on live decisions, fast. Cannot build anything.
- Audit
- An honest read of where behaviour breaks today. Cannot fix it in the same breath.
- Sprint
- One defined behaviour, moved in weeks. Cannot shift culture or trust, which run on slower clocks.
- Embedded partner
- Behavioural input through design, testing and rollout. Cannot work without real access.
- Fractional lead
- Behaviour woven into strategy, with a seat at decisions. Cannot add value from the mailing list.
- Capability build
- A team taught to own the work, with support and an exit. Cannot be a quick fix.
- Toolkit development
- Codifying what already works so it travels. Cannot conjure a rhythm that does not exist.
Choose between them with three dials, not by habit. How big is the risk if this goes wrong? How mature is the team that will carry it? How fast is the decision coming? Turn those three and the right model usually names itself. Reach for the model you always reach for and you will sometimes be right by luck, which is a hard thing to repeat.
Access determines value
Every model carries an access requirement, and access is the thing organisations most often withhold while still expecting the full result. This is where the biggest silent failures happen, because the engagement looks correct on paper and quietly starves in practice.
A fractional lead without a seat at real decisions is an expensive newsletter.
Let the model change
Every engagement is chosen at the moment of least knowledge, right at the start, before anyone has learned what the work actually needs. Six weeks in, everybody knows more, and quite often the original model is no longer the right one. A healthy working relationship lets the model follow the learning rather than treating the first choice as a cage.
The smaller recommendation
Sometimes the honest advice is to buy less. A sharp advisory week instead of the six-month retainer. A single sprint instead of the embedded partnership. This recommendation costs the practitioner revenue in the short term, which is precisely why it is worth so much. It buys something no marketing can: the client's belief in every recommendation you make afterwards.
Get the full guide
The complete guide, with worked examples, the running order for a team session, and all the detail the excerpt leaves out, is available as a free PDF.
Free. No spam. One email with your download link.