What does AI estimating cost a Malaysian subcontractor?

Short answer: With UkurBQ there are three parts. A free BQ test on one past tender, one fixed setup fee of RM 9,800 for BQ drafting in one trade (RM 4,900 for our first three founding firms), and your own Claude or ChatGPT business plan, which you pay directly to Anthropic or OpenAI. There's no per-tender charge and no share of your contract value, and you know the setup fee before you commit to anything.

How is AI estimating usually charged?

If you've looked around, you'll have seen a few different models. Some estimating software charges a monthly licence per user. Some outsourced takeoff services charge per tender or per drawing set. Some AI tools charge per page or per upload. Each has its place, but for an M&E subcontractor pricing several tenders a month, per-tender and per-page charges add up on jobs you don't win.

UkurBQ is structured differently. You pay once to have the setup built, tested and handed over. After that, the running cost is your own AI subscription, which you'd control and could cancel like any other software.

What exactly do I pay for?

What changes the price?

The standard setup is RM 9,800 for BQ drafting in one trade. The tender tracker adds RM 2,500. The work isn't the same for every firm, though. A small electrical subcontractor pricing lighting and power from one consultant's BQ format is a different job from a firm doing electrical, ACMV and solar across several main contractors, each with their own layout.

What we confirm on the call before you commit:

• Trades covered. Electrical only, or electrical plus ACMV, plus solar.
• BQ formats. How many consultant or main contractor layouts you regularly price into.
• Measuring rules. How much of your method is already written down, and how much lives in your estimator's head.
• Rate sheet. Whether your rates are in one clean spreadsheet or spread across old BQs.
• Team size. How many people need training.

Each extra trade is quoted as its own fixed setup. We confirm the number in writing before you commit, and it doesn't grow afterwards because a tender turned out larger than expected.

What does the setup fee include?

1. Mapping how you price. Your measuring rules, allowances, BQ layouts and rate sheet, written down for the setup.
2. Building the setup. Inside your own Claude or ChatGPT business account, so you own it.
3. Proof on 3 past tenders. The first is run together with your estimator, line by line, adjusting wherever it differs from how you measure. The other two are run blind, with no changes, and compared with your own BQs.
4. Training. A hands-on session on a real tender, a written guide for your setup, a recorded walkthrough and WhatsApp support afterwards.
5. Addenda handling and a tender tracker for closing dates and supplier quotes.

Until the blind runs match your own BQs, your team keeps quoting the way it does now.

How much does a Claude or ChatGPT business plan cost?

Anthropic and OpenAI set their own prices, usually per user per month, and they change them from time to time. We don't quote those figures here because they would go out of date. Check the current price on the provider's own pricing page before you sign up.

A few practical points:

• You pay the provider directly, in your company's name. We don't resell or mark up the subscription.
• Business and team plans are what we set up on, not personal accounts, mainly for data handling reasons.
• You choose how many seats you need. Often it's only the people who run takeoffs.
• If you stop using it, you cancel with the provider like any other subscription.

How do I work out whether it's worth it?

We won't give you a payback figure, because it depends on your own numbers. You can work it out roughly with things you already know:

• How many tenders you price in a typical month.
• How many days of takeoff each one takes your estimator.
• How many tenders you turn down or rush because there's no time.
• What your estimator's time costs the firm per day.

Takeoff is spent whether you win or lose. If a meaningful share of your estimator's week goes into counting points and retyping quantities on jobs you don't get, that's the time a setup is meant to give back. The free test lets you see the draft on your own tender before you spend anything, which is a better basis for a decision than any figure we could publish.

Are there any hidden or ongoing UkurBQ fees?

The setup fee is agreed in advance and is fixed. 30 days of WhatsApp support are included after handover, then RM 180 per hour only if you need it. If you later want something substantial added, such as a new trade or a new main contractor's BQ format, we'd talk about it and agree it before any work starts. Nothing is billed without you agreeing to it first.

We're also taking on founding firms now. They get hands-on attention from me directly and pilot pricing in return for honest feedback.

FAQ

Is the BQ test really free?

Yes. One past tender, under NDA, with a draft BQ returned for you to compare. There's nothing to pay unless you decide to go ahead with a setup.

Do you charge per tender after setup?

No. Once it's set up and handed over, you run as many tenders as your AI plan allows without paying UkurBQ per job.

Can we use the AI plan we already have?

If it's a Claude or ChatGPT business or team plan, usually yes. If your team is on personal accounts, we'd recommend moving to a business plan first.

When do we pay the setup fee?

Half when you approve the proposal, and the other half only when the blind proof runs match your own BQ. If they don't match and the setup can't be fixed, you don't pay the second half.

Keep reading

• Can ChatGPT or Claude price an M&E tender? What works and what doesn't
• Is it safe to put tender drawings and rates into ChatGPT or Claude?
• Excel vs a tender tracker: when the spreadsheet stops working