QuoteClerk AI vs manual quoting: the same list, two ways

One list, quoted by hand

A contractor sends a 40 line list for a kitchen and bathroom refit: pipe, fittings, valves, a few radiators, some cable and accessories, half of it written in trade shorthand. At the desk, someone:

1. reads the list and works out what each line means;
2. opens the quote screen in the merchant system and adds the customer;
3. searches for each product, often twice, because the words on the list are not the words in the catalogue;
4. checks the quantity and unit, such as lengths against metres;
5. picks the price: the account price, a special, a break for the quantity;
6. checks stock, notes anything to order in, and saves the quote as a PDF;
7. emails it back.

None of it is hard. All of it takes attention, and it happens between phone calls and counter customers. A 40 line list is a real block of someone's time, and the quote goes out when the desk is quiet, not when the contractor needs it.

The same list, with a draft

The list arrives as it always does: a spreadsheet, a PDF, the text of an email or a photo. It is read line by line, each line is matched to your own catalogue, and each matched line is priced from your own price data for that customer. What reaches your team is a draft with every line laid out next to the original wording, and the doubtful lines flagged with the reason: no clear match, a unit that does not fit, two prices tied, a price missing or out of date, stock missing or older than 24 hours. Uncertain lines are flagged for a person to check, which matters most on photographed and handwritten lists.

Your team works through the flags, checks the rest, approves the draft, and sends the A4 PDF with your logo. A UTF-8 CSV of the approved lines comes with it.

Step by step

What stays with your team

• Which product, when it is not clear. Ranked candidates are shown, never chosen on their own.
• Substitutes. Swapping the product on the customer's line for a different one needs a short technical review note.
• Commercial calls. A sharper price for a job you want to win is your decision, not a rule.
• Approval. A draft cannot be approved while any line is blocked, and every line needs a person's review.
• The customer. Your team sends the quote, from your own email, under your name.

Where mistakes go

By hand, a mistake is invisible until the customer finds it: the wrong level of price, an expired special, a length quoted as a metre. With a draft, the same doubts surface as flags before approval, with the reason written next to the line. That does not make a person unnecessary. It makes the person's time go on the lines that need it.

How the difference is measured

I do not publish a time saving figure, because it depends on your lists. The paid pilot measures it. You get a time scorecard with the minutes per list before and after, across extraction, matching, review, correction and handover, plus a catalogue issues report, the field mapping of your export columns, the approval rules in use, and a written go or no go note. Before the pilot starts, the success criteria are agreed in writing. If the pilot misses them, you pay nothing for it.

Pricing

Starter is £390 a month for up to 20 requests, and Team is £790 a month for 3 users and up to 200 requests, both with no setup fee. The full rollout is £1,290 a month for up to 500 requests and 10,000 line items, or £3,900 a month for up to 1,500 requests and 30,000 line items, plus £5,950 onboarding including the £1,490 paid pilot. How the paid pilot works ↗

Frequently asked questions