Your quotes depend on one person and go out too late

In machining and sheet metal, an eight to ten day response time loses jobs that your price would have won.

Thirty minutes, no commitment

Describe your flow. I'll tell you what can be automated and where I would start.

An enquiry arrives with a PDF drawing or a STEP file. It waits until the right person has time. That person opens the drawing, estimates the setup times, looks up the material price, digs out the quote they did for a similar part two years ago, copies, adjusts, and produces a price.

That judgement cannot be replaced: it is your trade, and no algorithm will notice that a part needs a special fixture. What can be replaced is everything around it — finding the material price, formatting, chasing and tracking.

What it looks like in your shop

Estimating lives in a workbook only its author understands

Tabs per part family, coefficients tuned over the years, formulas nested twelve levels deep, and hard-coded cells overriding a formula for some special case long forgotten. That file is a company asset and it is neither documented nor seriously backed up.

The material price is the one from the last quote

You ought to ask the stockholder for a fresh price, but that takes two days. So the old price gets reused with a safety percentage on top. Through periods of price movement, that approximation turns a won job into a lost one, and nobody notices before year end.

Nobody knows what happens to a quote once sent

No status, no systematic chasing, no win rate by customer or by part family. So you don't know whether you are losing on price, on lead time or on the quality of the response — and you adjust your prices blind.

The same parts get estimated from scratch again

A customer re-orders a part you have already made, with a small variation. Unable to find the previous estimate, you start over. Worse: you can't find the time actually spent last time either, so you repeat the original estimating error.

The quote goes out without ever being compared to actuals

There is no feedback loop. Nobody systematically compares estimated hours against booked hours. The part families you have been losing money on for three years carry on being quoted the same way.

What it actually costs

Estimating has the peculiarity that its errors never show up straight away. Here are the four figures to pull.

Average response time: 5 to 10 working daysTake the enquiry date and the send date on your last 30 quotes. An OEM consulting three suppliers often awards before your answer arrives: you are not losing on price, you are losing by absence.
Estimating time per enquiry: 45 minutes to 3 hoursAt ten enquiries a week that is a full-time post — occupied by your most experienced person, the one who should be on the difficult jobs.
Win rate: usually unknownIf it is unknown, that is already a finding. An SME that discovers its win rate per customer generally changes its commercial policy within the month.
Estimated versus actual gap: measure it on your last 20 jobsCompare estimated hours with booked hours. The systematic gap per part family is the single most profitable piece of data in all of this.

Beware the false problem: many managing directors want « software that quotes automatically from the 3D model ». On varied parts in small batches, those tools produce results a tradesperson has to rework anyway. The real gain is elsewhere: in response time and in the feedback loop.

How it gets fixed

The aim is to take your estimator from three hours to forty minutes per quote, and to make their knowledge readable by someone else.

1. We get the estimating logic out of the spreadsheet
Hourly rates per work centre, standard times per operation, setup coefficients, quantity break rules. Written down once, properly, with a history. This step is done with your estimator, at their pace: it is their knowledge being written up, not a generic model being imposed on them.
2. The incoming enquiry is prepared automatically
The enquiry arrives, drawings are filed, the customer is recognised, and the history of similar parts already quoted and already made is displayed alongside. Your estimator starts with the context in front of them instead of reassembling it.
3. Material prices are refreshed, not copied
Depending on your suppliers, either by pulling their price lists or by one periodic entry feeding every estimate. A quote never again goes out with last year's material price unless somebody decided that.
4. Quotes are tracked and actuals feed back into estimating
Status, chaser on due date, win rate by customer and by part family. And above all: when the job is made, booked hours come back alongside estimated hours. Within six months you are estimating on your own numbers rather than on recollection.

The estimator keeps control of the final price. At no point does a quote go out without a human approving it.

What this does not fix

What this does not fix:

It does not estimate for you. No automatic drawing recognition will replace your judgement on a difficult part, and I do not sell that promise.
It does not catch customer design errors or impossible tolerances. Your estimator still reads the drawings.
If your standard times are wrong, estimating stays wrong — faster and more consistently. That is why feeding actuals back is part of the project and not an option.
It does not suit high-volume repetitive work, where the subject is entirely different, nor one-off projects where the quote is a project in itself.
It will not win you a job you are losing on price. It will at least let you know it is the price, and not the response time.

The questions I get asked

What does it cost?

The two-day audit is €1,500. An estimating project generally lands between €10,000 and €15,000 fixed price, because writing up the estimating logic takes time with your estimator. It is also the one with the best return, because it touches margin.

How long before it runs?

Four weeks, with one honest caveat: your estimator's availability is the limiting factor. Budget two to three half-days of their time, and if those are not booked in advance, the project slips.

Do we need to change ERP?

No. Estimating plugs into what exists and the approved quote goes back into your usual tool for the rest of the sales process.

Does it read 3D files?

Simple, useful information can be extracted from a STEP file — bounding box, volume, declared material — to pre-fill and save time. I do not claim to derive a full machining routing from it: those tools exist, they are expensive, and on varied parts in small batches they need so much rework that they don't keep their promise.

Will our estimator lose their job?

No, and it matters to tell them early: this project does not succeed without them. They spend less time hunting for information and more deciding. In practice, the main resistance comes from the fear that their knowledge becomes transferable — which has to be discussed frankly, because it is partly the point, and because a company whose estimating rests on one person is fragile.

What if you're no longer available?

The code, documentation and estimating rules are with you, and readable. Your pricing logic is a strategic asset: it should depend neither on a supplier nor on a subscription.

Who owns the code?

You do. The assignment is written into the quotation, with the estimating rules exportable in an open format.

In your trade

This job is not equally urgent everywhere. These pages say where it sits in the running order, trade by trade.

Let's talk for thirty minutes

You describe your flow, I tell you what can be automated, what cannot, and where I would start. If I'm not the right person, I'll say so during the call.