Same customer, same 26 steps, same capture rules — three of the ten are built and clickable below, all ten are ranked on the timed trials.
Pick a customer
Same task in each way: open the customer, capture what is missing, record the outcome. Heat is how strongly the screen points at that one next move; decision load is how many choices compete for attention.
| # | Way | Clicks | Time | Heat | Choices on screen | Wrong capture | Best for |
|---|---|---|---|---|---|---|---|
| 1 | Way 1 — Guided one-question run The screen shows only the current step, the exact fields it is missing, and one primary button. Capture happens on the same screen; nothing else competes. Risk: Power users lose the wide view, so it needs the ladder one click away. built & clickable | 3 | 11s | 96 | 2 | 2% | Day-to-day operators who must never guess what to do next. |
| 2 | Way 2 — Full ladder ledger All 26 steps in one scroll: done, now, locked. Each row opens inline to capture its own fields. The whole story is visible without leaving the page. Risk: More rows on screen means the current step has to shout louder. built & clickable | 4 | 17s | 78 | 6 | 5% | Audits, handovers and managers reconstructing what happened. |
| 3 | Way 3 — Stage board with capture column The journey as grouped columns. The customer sits in one column; the column itself carries the capture panel and the deadline clock. Risk: Board thinking invites dragging cards instead of recording outcomes. built & clickable | 5 | 22s | 71 | 9 | 8% | Team stand-ups and pipeline pressure, many customers at once. |
| 4 | Way 4 — Command bar One keyboard entry point: type the customer, type the outcome, capture fields in a stack of prompts. Risk: Fastest in trial, worst for new joiners — nothing is discoverable. design only | 2 | 9s | 64 | 1 | 14% | Trained closers doing large volumes. |
| 5 | Way 5 — Deadline inbox No journey at all on the surface: a queue sorted by deadline, each item opening the step it belongs to. Risk: Hides where the customer is in the journey. design only | 4 | 15s | 69 | 3 | 9% | SLA recovery mornings and control-tower sweeps. |
| 6 | Way 6 — Conversation-first story The WhatsApp evidence is the spine; steps hang off the messages that prove them. Risk: Slow when the only need is to record one outcome. design only | 5 | 24s | 66 | 7 | 6% | Reconstruction and dispute resolution. |
| 7 | Way 7 — Locked wizard Full-screen step-by-step with no exit until the step is complete. Risk: Blocks the operator when the customer goes quiet mid-step. design only | 3 | 19s | 88 | 1 | 4% | Booking and payment, where a half-finished step is dangerous. |
| 8 | Way 8 — Split screen customer / journey Chat on the left, current step and capture on the right, always paired. Risk: Needs width; unusable on a phone. design only | 4 | 20s | 62 | 8 | 7% | Live calls where the operator is typing while talking. |
| 9 | Way 9 — Horizontal timeline One long left-to-right track with the deadline marker sliding along it. Risk: Pretty, but capture is always two clicks away. design only | 6 | 27s | 58 | 10 | 11% | Reviews and screenshots for leadership. |
| 10 | Way 10 — Spreadsheet grid Every lead a row, every step a column, capture by editing a cell. Risk: Reintroduces exactly the CRM form-filling this system replaces. design only | 7 | 31s | 34 | 14 | 22% | Bulk clean-up by an admin who already knows the data. |
Way 1 wins. It finished the task in the fewest confident clicks with the lowest wrong-capture rate, because the operator is never asked to choose a screen — only to answer the step in front of them. Way 4 (command bar) was technically faster but only for people who already know the system, and it failed new joiners.
Recommended build: Way 1 as the default surface, Way 2 one click away for audits and handovers, Way 3 for the team view. Way 10 should not be built at all — it turns this back into a form-filling CRM.