Cutting cabinet quote turnaround from three days to one hour

Business context

A custom kitchen cabinet workshop builds fitted kitchens, wardrobes and home offices for homeowners and a handful of builders. Sixteen people work in the shop and the office, and every quote passes through the owner, who has priced cabinets for twenty years.

We built an online quote calculator so the workshop could quote the same way every time and send a price in the same week a customer asked. The project sits alongside our other web app development work; for a different example of replacing a manual process, see the booking portal we built for a physiotherapy clinic.

The problem

Quoting lived in one spreadsheet. It held the owner's pricing rules, material costs, finish options and a set of formulas that had grown for years. Only the owner fully understood it.

When an enquiry arrived, a staff member copied the customer's sizes into the spreadsheet, guessed at a few options, and left it for the owner to check. The owner often found the guess wrong, recalculated by hand, and sent the quote two or three days later. Some customers went elsewhere in that time.

Prices drifted. Two staff quoting similar kitchens could land thousands apart, and nobody could explain why. The owner corrected more than half of the quotes that reached him, which was the real cost.

What we built

We turned the spreadsheet into a web app that follows the owner's rules. A staff member enters the room dimensions, cabinet front, material, finish and hardware, and the calculator returns a price line by line. It shows the totals the owner already used, not a new pricing model.

Each quote ends in a branded PDF the workshop can email from the tool. The document is generated from the same record as the on-screen quote, so the number a customer sees is the number the office sees. Quotes carry a reference and a date, which made follow-up possible for the first time.

Behind the price is a set of rules stored in the database rather than in formulas. The owner edits material costs, labor rates and option prices from an admin page, and the change applies to every new quote. Nothing needs a code change. We stored the rules as data, with a history of who changed what and when, because the workshop had never had that record before.

Where the rules came from

We sat with the owner and wrote every pricing rule into a document before we wrote any code. That step did most of the work. It surfaced three rules that contradicted each other, including a labor rate that had been applied in two different ways depending on who was quoting.

Rollout

• Week 1: Interviewing the owner and two staff, then writing the pricing rules down.
• Week 2: Building the calculator and checking it against 60 historical quotes.
• Week 3 to 4: Adding the PDF quote, the admin page for prices, and quote history.
• Week 5: Running the tool beside the spreadsheet, with staff comparing both.
• Week 6: Turning off the spreadsheet and training the office team on updates.

Before and after

Step	Before	After
Quote preparation	Owner reviews every one	Staff produce the first draft
Turnaround	Two to three days	Under an hour for standard kitchens
Pricing consistency	Varied by who quoted	One set of rules for everyone
Price updates	Owner edits spreadsheet formulas	Office updates prices on an admin page

Results

Median turnaround fell from three days to about an hour for standard kitchens. The workshop sent 42 quotes a week instead of 11, without hiring. The share of quotes the owner had to correct dropped from 55% to 9%, and most of the remaining corrections were for unusual layouts the calculator flags for a manual price.

The owner's time moved rather than shrank. He still reviews complex jobs, but he stopped retyping the routine ones. He now spends that time in the showroom and on two large builder accounts he had no time for before.

Lessons learned

• Writing the pricing rules down exposed three that contradicted each other. We expected to find gaps, not active conflicts. Two staff had been pricing the same work differently for years, and both believed they were following the system.
• The hardest part was not the calculator. It was getting the owner to trust a number the tool produced without him. We earned that by matching 60 historical quotes exactly before launch, not by explaining the code.
• We built customer self-service first and staff saw little use for it. Most customers still wanted to talk through a kitchen, so we kept the calculator for staff and made the enquiry form the public entry point. That fit the business better than the original design.