The short version
- Work out the fully loaded hourly cost of your estimator, then multiply by the hours a bid takes. That is your floor, not your total.
- Divide by your hit rate to get what each won job cost in bidding. At one in six, six bids paid for one win.
- The largest cost is usually not the estimator at all, it is the bid you did not submit because there was no capacity.
- The real comparison is not price against price, it is which bids get produced at all.
The cost of one bid
Start with the estimator. Take the salary, add employer taxes, pension, holiday cover, insurance, software licences, the seat, and the share of overhead a desk carries. In most markets that lands somewhere between 1.5 and 1.8 times the base salary.
Divide by productive hours, not paid hours. After holiday, sick leave, training and the meetings that are not estimating, an estimator on paper for 2,080 hours is available for something closer to 1,500.
| Figure | |
|---|---|
| Estimator base salary | $85,000 |
| Loaded at 1.65 | $140,250 |
| Productive hours per year | 1,500 |
| Fully loaded hourly cost | $93.50 |
| Hours on a mid-size commercial bid | 38 |
| Direct cost of one bid | $3,553 |
| Hit rate | 1 in 6 |
| Bidding cost per job won | $21,318 |
That last line is the one worth sitting with. It is not an argument that bidding is wasteful, it is the cost of sale for a construction business and it should be managed like one.
Hit rate does more work than price
Improving a hit rate from one in six to one in five drops the bidding cost per win by about seventeen percent, and it does it without touching a single rate. The two levers on hit rate are bidding on better-suited work and submitting a more credible number, and the second is a function of how much time the estimate got.
Which is where the capacity problem bites. When four bids land in the same fortnight, something gives. Either one is not submitted, or all four get less attention than they needed. The cost of the first is obvious and unrecorded. The cost of the second is invisible until a job goes badly.
The most expensive bid in most years is the one that was never submitted.
The honest comparison
We are an outsourced estimating service, so treat this section with the scepticism it deserves. Here is what we think is actually true.
| In-house estimator | Outsourced | |
|---|---|---|
| Cost shape | Fixed, paid whether bidding or not | Variable, per package |
| Knows your business | Deeply, and this is the real advantage | Learns it over several jobs |
| Knows your subcontractors | Yes, and it matters | No, unless you share the data |
| Peak capacity | Fixed at one person | Scales with the week |
| Cost per bid at low volume | High, the salary is spread thin | Low, you pay for what you use |
| Cost per bid at high volume | Low | Higher than a busy salaried estimator |
| Availability at 48 hours notice | Depends what else is on | Contractual |
| Institutional memory | Leaves when they leave | Held in your files, not ours |
If you bid consistently and heavily, a full-time estimator is cheaper per bid and better informed. That is simply true. The case for outsourcing is not that it beats a good in-house estimator. It is that it covers the peaks a single estimator cannot, and the trades outside what they know well.
When outsourcing genuinely makes sense
- Bid volume that swings, where a fixed salary is idle half the year and overwhelmed the other half
- A trade outside your normal work, where your estimator would be learning on a live bid
- A second opinion on a large or unusual job, before you commit to a number
- Cover while an estimator is away, on a job that cannot wait for them
- Growth, where you want to bid more before you can justify the second hire
And when it does not: routine work in your core trade, at steady volume, where your own estimator is faster and knows which subcontractor will actually turn up. We say that to prospective clients regularly, because the alternative is a client who never comes back.
Working out your own number
- Fully loaded cost of your estimating staff, including software and overhead share.
- Divided by realistic productive hours, not contracted hours.
- Times the average hours per bid, which you will have to measure for a month to know.
- Divided by your hit rate over the last two years, not last quarter.
- Then look at how many bids you declined for capacity, and price those separately.
Step five is usually the one that changes the decision.
Written by the estimating team at QuantX Estimation. Since 2016 we have measured and priced construction work for contractors in the United States and United Kingdom, at a fixed fee agreed before we start.
