Consulting Engagement Retrospective Template
Two documents and three sheets that separate scope changes from estimating error and check every task-type lesson before your next proposal repeats it.
Free download · No account needed
Retrospective Note · Solvane Consulting · Colston Mutual Insurance claims process redesign · Reviewed 15 Jul 2026
124 days against 106 estimated, and one type carries almost half the overrun
| Task type | Est | Actual | Factor |
|---|---|---|---|
| Discovery & Interviews | 16 | 17 | 1.06 |
| Process Mapping & Analysis | 38 | 43 | 1.13 |
| Workshop Facilitation | 24 | 34 | 1.42 |
| Deliverable Drafting | 28 | 30 | 1.07 |
Workshop facilitation isn't a new problem. An unrelated engagement five months earlier logged the same pattern at a 1.36 factor, with an instruction to add 35% whenever more than three client stakeholder groups attend. This proposal met that condition. Nobody checked before it was priced.
Correcting for the known lesson isolates $7,140 as avoidable, 47% of this engagement's entire $15,300 overrun.
Search consulting engagement retrospective and the advice has converged: run it within a week of closeout, and compare what was promised against what happened. Capture lessons under what worked, what didn't, what to change, and route the output into a template or an estimating model, because a lessons document nobody reopens is journaling. The best of the category goes further, joining the SOW's deliverable table against actual time data to compute a variance record, deliverable by deliverable.
A deliverable-level variance doesn't transfer. "The migration plan ran 40% over" tells the next proposal nothing without a migration plan in it. A task type does, since every engagement runs interviews, analysis, workshops and drafting regardless of client. That number is only honest, though, if scope changes never contaminate it: FAR 52.243-1(b) requires an equitable price adjustment for any change affecting cost, precisely because unpriced scope and a bad estimate are different problems. This pack's Estimate versus Actual sheet counts original-scope hours only; every addition goes in its own sheet, tagged billed or absorbed.
Solvane Consulting's ten-week claims process redesign for Colston Mutual Insurance closed 18 days over its 106-day estimate, $15,300 at the day rate, with $8,500 of it concentrated in workshop facilitation alone. That number meant something specific once checked against the workplan's own task register: an unrelated engagement five months earlier logged the identical pattern, and the instruction sat unchecked. Isolating what the known lesson would have caught prices the avoidable share at $7,140, 47 percent of the whole overrun.
What's in the pack
Retrospective Note
What the engagement was supposed to deliver against what it took, with every claim tied to a number from the sheets rather than a general impression.
Practice Improvement Actions
A short, specific list. Each action names the artifact it edits, the owner, and a date, not a summary of the note above it.
Estimate versus Actual by Task
Every task grouped by type, with scope changes already excluded, so the factor per type reflects estimating accuracy alone.
Scope Change Analysis
Everything the client added or changed, tagged billed or absorbed, with the margin impact of each kept separate from a properly priced change order.
Lesson Register
Every pricing lesson the practice has logged, checked against the current engagement, with a dated record of whether it was actually applied.
How it works
- 1
Open in River, or take it blank
Edit with AI installs the pack as a private Space and starts the retrospective. Download gives you the Word documents and CSV sheets free, no account.
- 2
Send the workplan, actuals, and any scope changes
The original task-level estimates from the workplan, the recorded time, and anything the client added, priced or not.
- 3
Get the honest factor per task type
Scope changes are pulled out first, so an absorbed add-on never inflates the number meant to measure estimating accuracy.
- 4
Check it against what you already knew
Every existing Lesson Register entry gets checked against this engagement's task types, and an unapplied lesson gets priced, not just noted.
Frequently asked questions
Is this engagement retrospective template free?
Yes. Download gives you the whole pack with no account and no email: two Word documents and three CSV sheets. Edit with AI creates a free account, installs the same pack as a private Space, and works through your actual workplan, timesheet and scope changes.
What format are the downloaded files?
Word (.docx) for the Retrospective Note and Practice Improvement Actions, and CSV (.csv) for the Estimate versus Actual, Scope Change Analysis and Lesson Register sheets, zipped together. Everything opens natively in Word, Pages, Google Docs, Excel, Numbers and Sheets.
How is this different from a generic lessons-learned document?
A lessons-learned document is prose: what worked, what didn't. This pack computes a factor per task type with scope changes already excluded, then checks that factor against every lesson the practice already logged. A repeat mistake gets priced and flagged as a repeat, not written up as if it were new.
We don't have a Lesson Register from an earlier engagement. Can we still use this?
Yes. The register starts empty and this retrospective's findings become its first rows. The Estimate versus Actual sheet and the scope-change separation both work on a single engagement with no history behind them at all.
How do you tell a real lesson from ordinary estimating variance?
By the size of the factor and whether it recurs. A type finishing between roughly 1.05 and 1.15 is normal variance and gets logged as no action, not written up as a finding. A factor well above that, especially one matching an existing register entry, is worth pricing specifically.
How is this different from the consulting workplan and resource tracker?
That pack forecasts a live engagement while it's still running. This one starts after closeout and asks a different question: not what will this engagement cost, but what should the next one be priced at, given what this one actually took by task type.
What happens to an absorbed scope change once it's found?
It becomes a practice fix, not a write-off. The Scope Change Analysis sheet names what was added and what it cost, and the escalation path or a standing intake rule is where the fix goes: no added session gets scheduled again without a signed change order first.
Turn this engagement into next proposal's pricing
Take the Word documents and CSV sheets blank, or send River the workplan, the actuals and what changed, and get the factor per task type back.
Edit with AI