How many hours will this scope actually take?
Paste past engagements with what was known at proposal time and get back a fee estimate and how much to trust it. Twenty-two delivered engagements against three things known at proposal time. The model explains 94% of the variation and not one of its three predictors is significant, which is a diagnosis rather than a failure.
Work Intermediate Regression free
After you install, this is the model to open.
What Should This Engagement Cost to Deliver?
- In your spreadsheet, click the Sortia icon in the strip of icons down the right-hand edge. No strip? Click the arrow at the bottom-right to open it. You can also use Extensions, then Sortia, then Open Sortia.
- Click Start from a template and put that name in the search box.
- Pick the card with that name and click Load this template. It arrives on a new tab with real numbers already in it.
The answer
- Hours per scope item
- 25.9 95% interval 22.8 to 29.0
- A fifteen-item scope
- 428 hours, on a 39.5-hour baseline
- At a $185 blended rate
- $79,200 priced before anyone opens a timesheet
- The model
- R2 = 0.94 scope items alone carry 93.8% of the variation
Twenty-two delivered engagements with the three things you know before you sign one. Click Run. R2 comes back 0.9404 and Significance F is about 3 in a hundred billion, so the model as a whole explains 94% of the variation in delivery hours and is unarguably better than nothing. Then read the coefficient table and nothing is significant: scope items at p 0.055, stakeholders interviewed at p 0.920, repeat client at p 0.365.
A model that explains almost everything while none of its parts explains anything is not a contradiction, it is a specific and diagnosable condition, and the correlation matrix names it. Scope items and stakeholders interviewed correlate 0.988. They are the same variable measured two ways, because a bigger engagement has more of both, so the fit cannot tell which one the hours belong to and splits the credit into two coefficients with error bars wide enough to swallow either.
Now do the second run: set the predictor range to the scope items column alone and rerun. R2 is 0.9376, so dropping two predictors cost you three thousandths of your explanatory power, and the coefficient comes back at 25.91 hours a scope item with a standard error of 1.50, a t of 17.33 and a 95% interval from 22.79 to 29.03. That is a usable number.
The pricing block at the side of the sheet fits that same one-predictor line on the sheet rather than quoting it, so it keeps working when you paste your own history in: it reads 25.91 hours an item over a 39.5-hour baseline, which prices a fifteen-item scope at 428 hours and, at a $185 blended rate, at about $79,200. The repeat-client flag deserves a separate word: it never clears the bar in any version of this model, which is not proof that repeat work is no cheaper, only that on twenty-two engagements the effect is smaller than the noise.
When two predictors correlate above about 0.9, keep the one you can measure earliest and most reliably and stop. What the model cannot tell you is how the hours will be spread across the engagement, only the total. To use your own history, put delivered hours in the first column of the sheet, keep proposal-time facts in the next contiguous columns, and widen both ranges to your rows.
The model
It arrives on a tab called Template: What Should This Engagement Cost, carrying these columns:
- Delivery hours
- Scope items
- Stakeholders interviewed (count)
- Repeat client (1 = yes)
- Engagement
with the model computed beside the data:
| Hours per scope item, fitted here | 25.91 |
| Baseline hours before the first item | 39.53 |
| Hours the fit predicts | 428.1 |
| Fee that implies ($) | 79,203.9 |
Once it is in your sheet
- The model arrives with real numbers in it and runs as it stands, so you can press the button first and understand it second.
- Change the numbers to yours. The sheet marks which cells are inputs and which hold formulas, and most labels carry a note explaining the row.
- Press the run button at the bottom of the panel. It is labeled for the tool you are in, and the result lands on its own tab, with a written reading of it beside the figures.
Never used Google Sheets? Start here goes the whole way, in seven steps, and assumes nothing.
Next question
- What if the partner with the fee book leaves?What Does Losing One Person Cost?
- A consulting engagement risk register ranking five risks by exposure and by mitigation value, since the two rankings are not the same list. Free template.What Could Go Wrong on This Engagement?
- Is your busiest person really the problem?Is Anyone Booked on Two Clients at Once?
- What does one conflict of interest cost you?Who Works on Which Client?
- Which client is quietly about to leave?Which Engagements Renew?
- Which date can you defend when the client pushes?Can I Promise That Date to the Client?
Every model like this one, and the method behind them: Statistics in Google Sheets.