Think Build Implement Repeat
London, UK +44 7367 067226
WhatsApp FOLLOW f in X
  1. Home
  2. Blog
  3. We Price Every Job by Gut Feel and Win the Unprofitable Ones. Can Machine Learning Help?
Problems We Solve

We Price Every Job by Gut Feel and Win the Unprofitable Ones. Can Machine Learning Help?

When quotes are priced by instinct, you win the jobs you underpriced. We use your past job data to predict true cost and flag quotes likely to lose money.

Updated 3 min readBy SpiderHunts Technologies

Free estimateNo obligation

Get a free estimate

Tell us what you need. A senior engineer reads every enquiry.

Takes under a minute. We never share your details.

  • Free consultation
  • No commitment
  • NDA on request

Prefer to talk? Book a free 30-minute call →

Quick answer — TL;DR

If each quote is priced from experience and the jobs you win are disproportionately the ones you underpriced, you are learning the wrong lesson from every win. A model trained on your past jobs, comparing what was quoted with what the job actually cost, can predict the likely true cost of a new job and flag quotes that are probably too low before they go out.

The jobs you win are the ones that hurt

Your estimators price jobs from experience: look at the drawings or the brief, think about similar work, add a margin, send the quote. Some jobs come in on budget. Others overrun badly. When you look at the year, a pattern appears that nobody likes: the jobs you win most easily tend to be the ones that lose money.

It makes sense once you see it. If your price is too low, the customer says yes quickly. If it is right or high, they shop around. Gut-feel pricing does not just produce errors; it selects the worst ones for you to carry out.

Why estimates drift

Estimators remember the jobs that stood out, not the typical ones. They rarely see the final cost of a job they priced, because by the time it closes they are on to the next quote. Material prices move, labour rates change, and certain types of job keep overrunning for reasons nobody has analysed, such as access difficulties, particular customer types or small jobs that carry fixed overheads.

Most businesses have the data to fix this. Job costing, timesheets and invoices hold what every job actually cost. It is just never joined back to the original quote.

What gut-feel pricing costs

PatternConsequence
Underpriced jobs won easilyMargin lost on the work you take
Overpriced jobs lostGood, profitable work goes to competitors
Inconsistent quotesTwo estimators price the same job differently
No feedback loopThe same errors repeat every year
Knowledge in one headPricing quality depends on who is in the office

The effect is often invisible in any single job. It shows up as a business that is busy all year and has less profit than it should at the end.

How we predict job cost from your own history

  1. We join quotes to actual outcomes by bringing together your quoting system or spreadsheets, job costing, timesheets and invoices, so every past job has a quoted figure and a real cost.
  2. We work with your estimators to list what drives cost in their view, such as job size, type, location, access, materials, customer and season, and turn those into model inputs.
  3. We train a model that predicts the likely true cost of a job from those inputs, with a range, and test it on past jobs it has not seen.
  4. We add a check at quoting time: the estimator enters the job details as usual, and the tool shows the predicted cost range next to their price and flags quotes that fall below it.
  5. Where you have enough data on won and lost quotes, we add a second view showing how price relates to win rate for similar jobs, so pricing decisions weigh margin against the chance of winning.
  6. We feed each finished job back into the data so the model keeps up with material prices, labour rates and new types of work.

The estimator still sets the price. The model is a second opinion that has seen every job you have done, not just the memorable ones.

What changes for the estimators

Before a quote goes out, the estimator can see what similar jobs actually cost and whether this price is in the danger zone. A flagged quote prompts a second look, not an automatic change. Over time, recurring overruns become visible by type, so you can fix the cause, such as a missing line item or an access allowance, rather than absorb it.

New estimators learn faster, because the tool shows them what the business's own history says rather than relying only on shadowing.

Managers get a view they did not have before: quotes that went out below the predicted cost, and what happened to them. That turns pricing from a private judgement into something the team can review together, without it becoming personal.

Is this your situation?

  • Jobs are priced from experience rather than a costed model.
  • Some types of job regularly overrun and nobody is sure why.
  • You win work easily but margins are lower than expected.
  • Quoted prices are never compared with final job costs.
  • Different estimators price the same work differently.

FAQ

Frequently asked questions

The questions readers ask us after this guide.

Still have a question?

Ask us directly — a senior engineer will get back to you.

Ask about your project

How many past jobs do we need?

Enough of each main job type to see a pattern. If you have a few hundred completed jobs with costs, that is often a workable start; with fewer, we focus on the most common types.

Our job costing is incomplete. Is that a problem?

It limits accuracy. Often the first step is tidying how costs are recorded, which is useful in its own right.

Will this set our prices automatically?

No. It predicts cost and flags risk. Pricing stays a commercial decision for your team.

What drives the cost?

How scattered the job data is across systems and how much tidying it needs, and where the tool needs to sit in your quoting process.

What do you need from us?

Past quotes, job costing or timesheet data, invoices, and time with your estimators.

Keep reading

More on Problems We Solve

Start here

Tell us what you are guessing today

Describe the decision, the data you keep and how it is made now. We will tell you honestly whether machine learning would help, what it would need, and if a simpler fix would do the job, we will say so.

  1. You tell us what you needTwo minutes on the form, or a message on WhatsApp.
  2. A senior engineer reviews itAnd comes back with questions, a realistic range and an honest view on fit.
  3. Free 30-minute scoping callWe talk through scope, options and a realistic estimate — with no obligation.
Free estimateNo obligation

Talk to someone who builds this

Send a short brief and we will come back with an honest view and a realistic range.

Takes under a minute. We never share your details.

  • Free consultation
  • No commitment
  • NDA on request

Prefer to talk? Book a free 30-minute call →