Value propositions and business cases
Feature → advantage → benefit
- A feature is what the product has or does ("automated triage of chest X-rays").
- An advantage is what that does better than the status quo ("urgent findings flagged in minutes instead of queuing").
- A benefit is what that means for this buyer's stated goals ("this closes the delayed-findings risk behind your two incident reviews — the thing your clinical director said keeps him up at night").
Features are generic; benefits are personal and only exist because you did discovery. Never state a feature without carrying it to a benefit tied to something the buyer said — which is why demoing before discovery is malpractice.
Quantifying value
Value in B2B comes in four currencies, and a good business case mixes them:
- Time saved → converted to money via loaded staff cost, or to capacity ("frees 0.4 of an admin FTE" — full-time equivalent).
- Cost avoided/reduced — outsourcing spend, locum cover, penalties, rework.
- Risk reduced — clinical incidents, compliance breaches, reputation. Anchor it to what the buyer spends on mitigation, or the cost of one occurrence.
- Revenue/throughput enabled — more activity through the same resource; in the NHS, capacity and target performance play this role.
A simple ROI model
ROI (return on investment) = (annual gain − annual cost) ÷ annual cost. Payback period = how long until cumulative gains cover the cost.
Nothing fancier is needed than:
- baseline (what the problem costs today, in their numbers) →
- conservative, sourced improvement assumption →
- gain − cost →
- payback.
Present ranges and invite the buyer to attack the assumptions — a business case survives finance only if the buyer co-owns the inputs.
Worked example: automating GP patient registrations
(One NHS-facing startup sells automation in exactly this space; every number below is an illustrative assumption you would validate in discovery, not a claim about any real product.)
Before: a GP practice of ~12,000 patients handles ~1,300 new-patient registrations a year. Each arrives as a form that admin staff check, chase and key into the clinical system — assume 15 minutes each including rework.
| Line | Calculation | Value |
|---|---|---|
| Admin time on registrations | 1,300 × 15 min | ≈ 325 hours/yr |
| Loaded cost of admin time | 325 h × ~£15/h | ≈ £4,875/yr |
| Error/rework, duplicate-record cleanup | assume +20% | ≈ £975/yr |
| Baseline cost of the problem | ≈ £5,850/yr | |
| Software cost (illustrative) | £2,000/yr | |
| Automation rate (conservative) | 80% fully automated | saves ≈ £4,680/yr |
| Net gain | £4,680 − £2,000 | ≈ £2,680/yr |
| ROI | 2,680 ÷ 2,000 | ≈ 134% |
| Payback | assuming the year's fee is paid up front, gross savings repay it in ~5 months | ≈ 5 months |
After, told as a story: registrations complete same-day instead of over a week; the practice manager redeploys freed hours to call handling — what patients actually complain about; and across a four- or five-practice Primary Care Network the numbers multiply by four or five, which is how you sell the group deal.
What really carries the case isn't the modest cash line but the staff-time and patient-experience story in a sector with chronic admin shortages — which is why you quantify time and risk, not only pounds. Walking through a model like this is a complete answer to "how would you build a business case?"