6.7% to 11.2% net profit in 120 days
A tenant improvement GC was using each new job to finish the last one, and could see two weeks ahead at best.
What it looked like from the owner's chair
Revenue had swung between $3M and $8M with $9M projected, and cash was reactive the whole way. He had debt from a bad year and was working on a two week visibility window, so a job trending tight announced itself when it hit, not 60 days out. In his own words he was using income from new projects to finish and close projects already underway, and he couldn't see how to get off it.
What was actually happening
Four things stacked. Estimates weren't capturing every cost, so markup became a contingency for scope that was never priced. Overhead went into bids at 10% while the business had actually run 16.4% the previous quarter, because tenant improvement work is feast and famine and the assumption never flexed with it. There was no job costing, so the fact that project manager and superintendent time was being charged at roughly half what it cost wouldn't reach a report. Collections were slow on top of all of it. Every job was underpriced before a crew mobilized, and the shortfall had to be funded by whatever job came next.
This is a known failure and it has a page of its own, with what it costs and how to size it in your business.
The work
We rebuilt the estimating template to price the work honestly: 10% of all labor for superintendent nonproductive time, 10% of superintendent time for project manager support, 10% for assistant project manager support, and a financing line for funding material and subcontractors 30 days ahead of getting paid for them. Then weekly cash flow forecasting, which is what gave him enough runway to see work slowing and trim overhead ahead of it rather than after. That took 4 points out of the overhead rate, about $200,000 a year at his volume.
The result
Net profit went from 6.7% to 11.2% inside 120 days, a 4.5 point swing worth roughly $225,000 a year. Two merchant cash advances were retired, one at $4,200 a week and one at $14,200 a month, about $389,000 a year of debt service that stopped leaving the business, plus another $38,000 of debt paid down. Most of that gain is the overhead cut. The rebuilt estimating template hadn't fully worked through at the 120 day mark because jobs bid under the old pricing were still moving through backlog, so there's more coming.
Other contractors with the same problem
All 14 are on one page, filterable by which step did the work.
What owners ask about this one
What was actually wrong?
Four things stacked. Estimates weren't capturing every cost, so markup became a contingency for scope that was never priced. Overhead went into bids at 10% while the business had actually run 16.4% the previous quarter, because tenant improvement work is feast and famine and the assumption never flexed with it. There was no job costing, so the fact that project manager and superintendent time was being charged at roughly half what it cost wouldn't reach a report. Collections were slow on top of all of it. Every job was underpriced before a crew mobilized, and the shortfall had to be funded by whatever job came next.
What did you change?
We rebuilt the estimating template to price the work honestly: 10% of all labor for superintendent nonproductive time, 10% of superintendent time for project manager support, 10% for assistant project manager support, and a financing line for funding material and subcontractors 30 days ahead of getting paid for them. Then weekly cash flow forecasting, which is what gave him enough runway to see work slowing and trim overhead ahead of it rather than after. That took 4 points out of the overhead rate, about $200,000 a year at his volume.
How long did it take?
120 days. That is the time to the result on this page, not to the last piece of the install.
Which part of the system did it?
step 03, overhead calculation, step 04, estimating system and step 07, monthly cadence. Installed in dependency order, which is the same order every client gets, because a step that reads from a number the one before it never established produces output that looks finished and isn't.
Wrote CONTROL, the 8 step system behind these figures. More about the author.