Ask how long a quote takes and you'll get a number. Say it's nine days.
Walk through those nine days and most of it isn't work. It's waiting. Waiting on a site visit. Waiting on a supplier price. Waiting for the one person who knows the margin to get off another job and look. The estimating itself might be two or three hours.
Why faster estimating changes nothing
So people speed up the wrong part. They buy software to make the two hours of estimating faster, then wonder why the quote still takes nine days. The bottleneck was never the maths.
Take an hour off the estimate and the quote still sits in the same queue. The client still waits, and the job still goes to whoever replies first.
Map the waiting first
Before you change anything, take the last few quotes you sent and write down what actually happened on each day.
You'll normally find the same hold-ups every time:
- A site visit that can't be booked until someone is free.
- A supplier or subcontractor price that nobody chases until it's late.
- A finished quote sitting in someone's inbox waiting to be checked.
- The document itself, pulled together by hand from old quotes and templates.
None of those are estimating. All of them add days.
Where AI actually helps
AI's biggest lever in a quote is killing the wait. Drafting the moment a price lands. Flagging the job that's sat untouched for three days. Chasing the supplier price before it's late. Assembling the document so the only human step left is the judgement call.
That judgement call stays with your people. The margin, the risk, whether you want the job at all. AI just stops that person waiting on everything else to get there.
It's unglamorous, but it's where the days go. Map the waiting first and a nine-day quote comes back in days, with nobody estimating any faster. Then you're the one who replies first.