Work Understanding - Application Overview
Objective. Teach the customer value of Work Understanding: why work takes as long as it does and which change to validate first - and make concrete the sales 'friction finding with evidence, not causality' story and the CSM routine of reviewing detection confidence, investigating waiting and measuring an improvement before/after.
Work Understanding explains the time inside the work: where it waits, which steps vary, where friction and rework sit, and which single change is worth validating first. This module teaches reps to present a friction finding with its evidence while being disciplined about causality, and teaches CSMs to review detection confidence and taxonomy, investigate waiting, and run a genuine before/after improvement experiment with an owner.
The customer problem
Customers know a process is slow or expensive but not why. They see the SLA miss and the backlog, but the time disappears into a black box of handoffs, waiting, swivel-chair between systems, and rework. So improvement becomes guesswork - they reorganise, add headcount, or buy automation for the wrong step, and the cycle time barely moves.
The business impact is wasted improvement spend and persistent cycle time. Consultants are hired to map processes by hand - a snapshot that is stale on delivery. Automation is pointed at a visible step while the real delay is waiting between steps. In RCM, this is the difference between shaving minutes off data entry and attacking the days a claim sits waiting for information - only one of which moves days-in-AR. The customer needs to know which change to validate first, and have evidence for it, before spending.
How Time Champ addresses it
Work Understanding discovers how work actually flows and where time is lost. The demo areas include Work explorer, detection, workflow discovery, variants, conformance, step time, systems, root cause, friction, rework, automation, AI impact, an improvement workspace, taxonomy, standards and privacy.
The value sequence is: discover the real workflow from observed work (not a whiteboard), expose its variants and conformance (how often reality deviates from the intended path), quantify step time and waiting, and localise friction and rework to specific steps and systems. The improvement workspace then turns a finding into a tracked experiment with a before/after measure. Detection confidence and taxonomy are first-class because the honest unit is 'observed contribution with a confidence', not 'proven cause'.
Confirm with product how work is detected and what confidence/coverage applies for a customer; the demo shows the method on seeded data.
The value and how to measure it
The value is targeted, evidence-led improvement: spending effort and automation budget on the step that actually holds the time, with a measured before/after rather than a hopeful narrative. It replaces hand-drawn process maps and intuition with observed flow.
Measure with the grounding's candidate measures: validated work coverage (how much of the real process is observed with adequate confidence - low coverage means findings are partial); reviewed findings (friction/rework findings a human has actually examined, not just generated); improvement experiments (findings converted into tracked before/after changes with an owner); and measured time released (the verified reduction after a change, not the modelled opportunity). The discipline is that opportunity is a hypothesis and released time is the result - never report the first as the second.
Sales task made concrete
Your task is to show a friction finding and its evidence, and separate observed contribution from proven causality. Concretely:
- Open the Work explorer (https://work-understanding.demo.timechamp.io/#/explorer) and Work detection (https://work-understanding.demo.timechamp.io/#/sessions) from the overview (https://work-understanding.demo.timechamp.io/#/home).
- Pick one process and show its discovered flow and variants - make the point that this is observed, not a whiteboard drawing.
- Surface a friction or waiting finding and show the evidence behind it (step time, where it sits, how often).
- Then the integrity move that builds trust: 'This is an observed contribution to delay with a confidence level - it is a strong candidate to validate, not a proven root cause. Here is the experiment we would run to confirm it.'
Discovery to pair: 'When you last tried to speed up this process, how did you decide what to change?' 'How do you know today where the time actually goes - waiting, rework, or handoffs?' 'If we showed you the one step holding the most time, could you own a change and measure it?'
CSM task made concrete
Your task is to review detection confidence and taxonomy, investigate waiting, choose an improvement owner and measure before/after. A routine that produces a real result:
- Detection confidence and taxonomy: with the customer, confirm that detected work is mapped to the right process taxonomy and that confidence/coverage is adequate for the process in scope. Low-confidence findings are leads, not decisions.
- Investigate waiting, not just active step time - in most RCM processes the days sit in waiting (for information, for a payer, for an approval), and that is where the gain is.
- Choose one finding and one owner: improvement without a named owner and a single experiment does not land.
- Measure before/after in the improvement workspace: baseline, change, re-measure. Report measured time released, not the modelled opportunity.
- Guard against the overlap trap below when the customer wants to add up several findings.
Objections and honest limits
'You found four frictions worth X hours each, so that is 4X we can save.' No - the core interpretation check is that overlapping findings cannot simply be summed. Two frictions may touch the same time, and fixing one may absorb the other. Savings claims must come from measured before/after on validated, de-duplicated findings.
'We already did a process-mapping project.' That was a manual snapshot, stale on arrival and based on how people say they work. Work Understanding discovers how work actually flows, including the variants the map never captured, and keeps measuring.
'Just tell us what to automate.' It will surface automation candidates, but automating a step that is not the real constraint wastes money. The value is validating the right change first.
Honest limit: findings are observed contributions with confidence, not proven causality, and coverage may be partial. Detection method and coverage for a customer must be confirmed with product.
Interpretation check
Overlapping friction findings cannot simply be summed into a savings claim. Separate findings may touch the same time, and a finding is an observed contribution with a confidence level, not proven causality. Savings come from measured before/after on validated, de-duplicated findings.
In the field
KnackGlobal RCM: instead of shaving seconds off data entry, Work Understanding localises the waiting that drives days-in-AR (claims sitting for information or payer response), so the customer validates the one change that actually moves the cycle - evidence that outlasts a manual process-mapping engagement.
Practical task
Using the demo, pick one process: present its discovered flow and one friction/waiting finding with evidence, state the finding as an observed contribution with confidence (not a cause), and write the before/after experiment you would run - baseline metric, the single change, the owner, and the re-measure. Note how you would avoid double-counting if a second finding overlaps.
Self-check
The demo shows three friction findings of roughly 2 hours each. A prospect asks if that is 6 hours saved. What do you say?
No. The findings may overlap on the same time and each is an observed contribution with a confidence, not a proven cause. Savings must come from measured before/after on validated, de-duplicated findings - not from adding opportunities.
Why investigate waiting rather than active step time in an RCM process?
The cycle-time days usually sit in waiting (for information, a payer, an approval), not in active handling. Speeding up an active step that is not the constraint barely moves days-in-AR.
What does 'detection confidence' change about how a CSM uses a finding?
Low-confidence findings are candidates to validate, not decisions. The CSM confirms taxonomy mapping and coverage first, then runs an experiment, rather than acting on a generated finding directly.
How is Work Understanding different from a consultant's process map?
It discovers the actual flow from observed work, including real variants and conformance deviations, and keeps measuring - versus a manual, self-reported snapshot that is stale on delivery.
Verified vs demo. The workflow discovery, variants/conformance, friction/rework/step-time views and improvement workspace are demo-shown on seeded data; the actual detection method, confidence and coverage for a customer's processes must be confirmed with product.