Value hypothesis and pilot
Objective. Build a customer-owned baseline, a testable value hypothesis, a pilot scope and success criteria - and distinguish capacity value from cash in a one-page pilot charter.
Teaches you to convert interest into a disciplined pilot that produces a number the customer owns and will defend internally. You will write a one-page pilot charter with a baseline, a falsifiable hypothesis, a tight scope and explicit success criteria, and you will keep freed capacity rigorously separate from recovered cash so the business case survives finance.
See it in the demo
The customer problem
Pilots fail for predictable reasons: no baseline (so no before/after), a hypothesis so vague it cannot be falsified, a scope so broad it never produces a clean signal, and a value claim that mixes freed capacity with cash so finance rejects it. Business impact: a 'successful' pilot that cannot be turned into a funded rollout because the number is not defensible, or a pilot that drags because success was never defined. The buyer needs a number THEY own - built from their data, measured by an agreed method - because a vendor-asserted number does not survive their CFO. Your job is to design the pilot so that whatever it shows, it shows cleanly and credibly, and so that a modest, honest result is enough to justify expansion.
Baseline before hypothesis
A hypothesis without a baseline is a guess. First establish the customer-owned baseline for the one thing the pilot will move: current available-hours-to-output picture, current cost per clean unit, current time-to-detect a shortage, current denial/AR measure - whatever the discovered problem is. Often the honest baseline is 'they cannot see this today', which is itself the starting value (you are moving from no-visibility to a trusted number). Rules for a credible baseline: use the customer's own data and definitions, agree the measurement method up front (denominator, exclusions, time window, time zone, freshness), and write down what is NOT being measured. Where the data needed for a clean baseline is not available, say so and either narrow the pilot or flag a data dependency to a specialist - do not paper over it. The baseline is co-owned; the customer signs off on the method before the pilot starts, which is what makes the result theirs to defend.
What to say / do - the testable hypothesis and tight scope
Write the hypothesis as a falsifiable statement with a direction, a measure and a window: 'On <one team/process/payer>, within <N weeks>, Time Champ will let managers identify and act on <specific exception/finding> such that <agreed measure> moves in <direction> - measured by <method the customer agreed>.' Example (output): 'On the claims adjudication team, within 6 weeks, Work Output will surface complexity-adjusted output per paid hour and the top three stuck steps, such that the team agrees on at least two improvement actions with owners and we measure released capacity on the agreed clean-unit definition.' Scope discipline: one application (from S03), one team/process/payer, one primary measure, a fixed window, and named owners on both sides. Success criteria must be set BEFORE the pilot and be binary where possible ('managers ran the exception review weekly: yes/no'; 'we produced a trusted, agreed number: yes/no'; 'the number moved by the pre-agreed threshold: yes/no'). A pilot that proves the routine was adopted and the number is trusted is a success even if magnitude is modest - say that up front.
Capacity value vs cash - keep them separate
This is the non-negotiable discipline and the thing that makes or breaks the business case. FREED CAPACITY is hours or units released - real operational value, but it is not money until the customer decides what to do with it (redeploy, absorb growth, reduce overtime, avoid a hire). CASH is money actually collected or saved - denial recovery collected, overtime reduced, penalty avoided. The Value Leakage recovery view explicitly distinguishes recovered capacity from recovered cash (https://leakage.demo.timechamp.io/#/overview and the recovery area); use that framing. In the charter, state the capacity hypothesis and the cash hypothesis as SEPARATE lines with separate measures, and never add them into one headline. Also state two honest limits out loud: recoverable value, billed value, collected cash and capacity value are four different numbers (RCM Operations makes this explicit), and overlapping findings cannot simply be summed. A finance buyer will trust a smaller, cleanly-separated number far more than a big blended one.
Objections and honest limits
'Just tell me the ROI, skip the pilot.' - 'I could give you an industry number, but you could not defend it to your CFO; a six-week pilot gives you a number built from your data that you own - that is what funds the rollout.' 'What if the pilot shows a small result?' - Reframe success: if the routine was adopted and the number is now trusted and moving, that is a fundable win; set that expectation in the success criteria so a modest magnitude is not read as failure. 'Can you guarantee the capacity turns into savings?' - No, and say so: freed capacity becomes cash only when the customer acts on it; that decision is theirs, and we keep the two separate precisely so we never over-claim. Honest limits: the pilot proves movement on one scope, not platform-wide ROI; any data dependency that could not be met must be named; and no success criterion, measure or baseline figure may be invented - all come from the customer and the agreed method.
Interpretation check
Freed capacity and recovered cash are different numbers and must never be added into one headline. Recoverable value, billed value, collected cash and capacity value are four separate things; a defensible business case keeps them apart, and a smaller cleanly-separated number beats a big blended one.
In the field
KnackGlobal pilot charter: scope = one claims team, 6 weeks, Work Output; capacity line = complexity-adjusted clean units per paid hour released; cash line (separate) = denial-rework reduction tracked toward collected recovery, kept apart from capacity; success = managers run the weekly exception review AND the clean-unit number is trusted and moving - a defensible wedge against the ProHance incumbent that the sponsor can take to finance.
Practical task
Write a one-page pilot charter: customer-owned baseline (with agreed measurement method and what is excluded), a single falsifiable value hypothesis, the pilot scope (one application, one team/process/payer, fixed window, named owners), and binary success criteria set in advance. Present freed capacity and cash as separate lines - never a blended number.
Self-check
Why must the baseline and measurement method be agreed and signed off before the pilot starts?
So the resulting number is customer-owned and defensible to their finance team. A vendor-asserted number or a method agreed after the fact does not survive a CFO and cannot fund a rollout.
Write the skeleton of a testable pilot hypothesis.
On <one team/process/payer>, within <N weeks>, <application> will let managers identify and act on <specific finding> such that <agreed measure> moves <direction>, measured by <customer-agreed method>.
Is a pilot with a modest magnitude a failure? When is it a success?
Not necessarily a failure. It is a success if the routine was adopted and the number is now trusted and moving per the pre-agreed binary criteria - which is what you set up front so small magnitude is not misread.
A buyer asks you to guarantee freed capacity becomes savings. Correct answer?
No guarantee - freed capacity becomes cash only when the customer decides to act on it (redeploy, absorb growth, cut overtime). That is their decision; we keep capacity and cash separate to avoid over-claiming.
Why present capacity and cash as separate lines rather than one ROI figure?
Because they are different numbers and overlapping findings cannot be summed; a finance buyer trusts cleanly-separated, smaller numbers far more than a blended headline, and it protects credibility into the rollout.
Verified vs demo. The capacity-vs-cash separation and the four-number distinction are demo-shown (Value Leakage, RCM Operations); the baseline, hypothesis magnitude and success thresholds are customer-owned and must never be invented, and any unmet data dependency must be named.