Demos done right

If a previous role ever had you demonstrating software, presenting technical work, or training people on a system, this section gives you the vocabulary to narrate that experience as deliberate sales craft.

If you've never given a demo in your life, learn the doctrine anyway: interviewers grade candidates on knowing what a good demo is, and stating it crisply puts you ahead of experienced reps who still give tours.

A demo is proof, not a tour

The worst demo is a feature walkthrough ("and over here you'll find…"). A good demo is claims and evidence: discovery produced a shortlist of pains; the demo proves, pain by pain, that your product resolves them.

Structure it as scenes — "you told me GP chasing eats your admin team's week; watch what happens to a routine report now" — each opening with their words and closing with the payoff. Money shot in the first five minutes; never save the best for a finale.

Voice from the field

The signature instruction of the Great Demo! method is "Do the Last Thing First": open with the end result the buyer actually cares about — the way a cookbook shows a photograph of the finished dish before the recipe — and only then earn the right to show how it's made.
— Peter Cohan, Great Demo! (2003; 3rd ed. 2023)

Tailoring is the whole game

Same product, different demo for a service manager (throughput, workflow) versus a clinical director (safety, evidence) versus IT (integration, security).

Tailoring is only possible if discovery happened — which is why experienced presales people refuse to demo cold, and why "can we do 30 minutes of discovery first so I show you the right 20%?" is a strength move, not an obstacle.

Field note

The tour remains the default failure mode even at companies with mature presales teams, because the polished standard walkthrough everyone knows by heart is always the path of least resistance when calendars are tight and discovery was thin. The tell in your own delivery is the phrase "and you can also…" — the moment you hear it, you've stopped proving and started touring. The hard part isn't knowing the doctrine; it's declining the shortcut on a busy Tuesday.

The "so what" test

After everything you show or say, imagine the buyer asking "so what?" If you can't complete "…which means [thing they told you they care about]", cut it. Ruthlessly applied, this halves most demos and doubles their impact.

POCs and pilots

A POC (proof of concept) is a limited evaluation where the buyer tests the product on their data or environment before committing; a pilot is a small live deployment. Powerful, and dangerous. The traps:

• No success criteria. A POC without pre-agreed, written, measurable pass/fail criteria can't be won — the goalposts drift forever. Agree criteria and the consequence of success ("if we hit these, we proceed to contract") before it starts. If any prior role had you defining acceptance criteria, test plans or evaluation gates — QA, project delivery, lab protocols, research methods — that is the same discipline; say so in interviews.
• Scope creep. "While you're here, could it also…" Every addition delays the close. Fix: written scope plus a polite trade — "happy to look at that as phase two, once we've proven the core."
• Free consulting. Unbounded POCs let buyers extract your expertise and hand your homework to a cheaper rival. Time-box, charge where the market allows (a paid pilot signals a serious buyer), and run POCs only for qualified deals with a power sponsor — never as a substitute for a decision.

Multi-threading, power, and champions

Multi-threading and stakeholder mapping

Multi-threading means holding relationships with multiple people in the buying organisation, rather than one contact (single-threading). Single-threaded deals die when your contact leaves, gets overruled, or goes quiet — and B2B decisions typically involve 6–10 people.

Stakeholder mapping is drawing that buying group: each person's role in the decision, personal win, and disposition toward you. The recurring roles:

• Economic buyer — owns the budget and can say yes when everyone else says maybe. In the NHS this may be a director, a finance committee, or a procurement board; finding out is part of discovery.
• Champion — wants you to win and has influence and will sell internally when you're not in the room. All three; a friendly contact without clout is a coach — useful, not sufficient.
• Users and evaluators — live with the product daily; can rarely say yes, can absolutely say no.
• Technical/procurement gatekeepers — security, information governance, integration, contracts. Engage early: in health tech, IG and clinical-safety sign-off can add months if found late.
• Blockers — people the status quo serves. Map them; sometimes convert them; never be surprised by them.

Selling to power

Selling to power means insisting, politely and early, on access to the economic buyer rather than settling for whoever will meet you. The lever is trade, not demand: bring something power wants ("services we work with usually want their finance lead in the room for the business case — shall we invite yours?").

If a deal reaches proposal stage and you've never met power, your forecast is fiction.

Champion building and enablement

Champions are made, not found: give a motivated stakeholder personal wins (their pet problem solved, credit internally), then equip them.

Enablement means arming them with materials built for internal reuse: a one-page business case in the organisation's own numbers, a short deck they can present without you, pre-written answers to the objections they'll face, references. Most of the selling in a complex deal happens in meetings you're not invited to; enablement is how you attend anyway.

The test of a real champion: they'll spend their own political capital — an introduction upward, an agenda slot. If they won't, keep them as a coach and keep threading.