Time and motion study, rev D
A keystroke-level comparison of fourteen routine BIM tasks performed in native Revit versus IQ Tools, built from Autodesk's own documented procedures and validated against a live commercial coordination model.
Less physical workload across fourteen tasks: 1,250 clicks and 679 keystrokes reduced to 73 and 47.
Recovered per active project model per year, moderate scenario. 20 hours conservative, 153 intensive, before any portfolio multiplier.
Faster to commit 50 edits. 9.87 s of frozen Revit UI versus 0.27 s, measured live by the Revit API.
Headline figures are the conservative case throughout. Revision C added six workflows and the reduction moved only a few tenths of a point: the measured surface nearly doubled while the rate held. Revision D then re-priced Workflows 02 and 06 against Revit's own Properties palette — the faster native route for a single uniform value across a multi-selection — which dropped the headline to 93.8%.
Not a vendor stopwatch demo.
Native task times were reconstructed with the Keystroke-Level Model (Card, Moran & Newell), the standard HCI method for predicting expert task time from documented procedures. Every native step comes from Autodesk Help, ASCENT courseware or Wiley's Mastering Autodesk Revit. Human time and machine time are reported separately, so a reader can attack either half.
Each workflow is priced against whichever native route is genuinely fastest — proficient-keyboard schedule entry, or the Properties-palette multi-select Apply. On Workflow 6 the palette route now wins; across all fourteen workflows the advantage still exceeds 93%.
On 4 and 5 September 2026 the predictions were tested inside Revit 2025 against a detached copy of a real central file: 136 doors, 195 sheets, 193 warnings. Every write test ran in a rolled-back transaction group.
The revision-C measurements open no transaction and write nothing, so they can run against a live production central. That is the natural basis for your own verification.
Fourteen tasks, side by side.
| Workflow | Native clicks | IQ clicks | Native keys | IQ keys | Native time | IQ time | Reduction |
|---|---|---|---|---|---|---|---|
| 1 · Ad-hoc parameter audit | 22 | 3 | 0 | 0 | 45.4 s | 3.95 s | 91.3% |
| 2 · Bulk parameter edit (50, Properties palette) | 4 | 4 | 8 | 9 | 12.0 s | 11.4 s | 5.0% |
| 3 · Sequential renumbering (50) | 108 | 6 | 350 | 4 | 477.8 s | 10.8 s | 97.7% |
| 4 · Type harmonisation (20) | 82 | 6 | 160 | 10 | 253.4 s | 14.1 s | 94.5% |
| 5 · Visibility diagnosis | 12 | 2 | 2 | 0 | 35.1 s | 4.45 s | 87.3% |
| 6 · Sheet set update (100 sheets, Properties palette) | 6 | 11 | 24 | 24 | 21.8 s | 24.1 s | −9% |
| 7 · CAD links & layer audit | 103 | 7 | 40 | 0 | 287.6 s | 13.8 s | 95.2% |
| 8 · Model group divergence audit | 90 | 2 | 0 | 0 | 457.5 s | 4.6 s | 99.0% |
| 9 · Find element behind a row | 5 | 2 | 7 | 0 | 14.3 s | 3.10 s | 78.3% |
| 10 · Normalise values (136 cells) | ~141 | 9 | ~66 | 0 | 250.2 s | 14.3 s | 94.3% |
| 11 · Grouping with subtotals | 12 | 1 | 0 | 0 | 23.7 s | 2.85 s | 88.0% |
| 12 · Warnings triage (12 of 193) | ~66 | 15 | 10 | 0 | 151.2 s | 21.4 s | 85.9% |
| 13 · Export to Excel | ~13 | 3 | 12 | 0 | 30.7 s | 4.97 s | 83.8% |
| 14 · Sheet × revision matrix | ~586 | 2 | 0 | 0 | 19.4 min | 3.95 s | 99.7% |
| Totals | ~1,250 | 73 | ~679 | 47 | 53.7 min | 2.3 min | 95.7% |
Proficient-keyboard native baseline. Against the mouse-driven baseline the first eight tasks alone cost 767 clicks and 50.6 minutes; the lower figure is reported throughout.
Where the numbers are weakest.
A study you cannot argue with is a study you should not trust.
- Task frequencies are estimates. No published time-on-task study for Revit exists. Per-occurrence times are sourced; how often you do each task is modelled, stated as three scenarios.
- Transaction times vary with model size. 1.5 s is a medium-model normalisation. Larger models are slower, which strengthens the case for IQ Tools further.
- Four of seven live tests model the native side. Those rows compare a documented procedure against a measured query, and are labelled as such.
- Fourteen workflows is not the whole product. They cover roughly half its features. Single-cell editing, search and several sheet and CAD tools are not measured at all.
- One model, one Revit version, single runs. No large-model test, no cold-start timing, Revit 2025 only.
Shaped by what broke.
IQ Tools was made by someone who runs model data for a living, and it went through beta inside a working architectural practice, on many live project models, through their actual workflows and deadlines. What ships is what survived that: the grid the beta team used every day, sharpened by what broke and by what they asked for.
Beta teams were productive in the grid after one 30-minute session, with no scripting knowledge on the team.
One practitioner's judgement is the strength at two seats and the question at two hundred. The continuity commitment on the support page answers it.