Best takeoff software in Australia: the takeoff you can check

So what makes takeoff software the best?

The best takeoff software is the one that does the takeoff for you, then lets you check every quantity it came up with. For a commercial builder in Australia, we reckon that's Estimator+. It's ours, so you'd expect us to say it, but the reasoning holds up.

Here's the thing about automated takeoff. A quantity you can't trace back to the drawing is a quantity you end up measuring again, and then what was the point? So there are two halves to this. Something has to do the measuring, and you have to be able to see exactly what it measured.

Estimator+ is built around both halves. The AI does the takeoff, and every number it produces sits on a shape you can look at on the sheet.

How does the takeoff actually get done?

The AI is Claude or ChatGPT, running in its desktop app on your own plan, connected to the tender folder on your computer. Ask for the takeoff and AI Takeoff V5 reads the whole tender before it measures a line. That's the spec, schedules, notes, contract and every addendum. Our AI takeoff software post covers that reading in detail.

Then it measures, working off the current revision of every sheet. A typical floor gets measured once and carried to each level it covers, with the exceptions each level prints. Scanned and rotated sheets with no text layer get measured too, along with clean PDFs.

Each line is filed under the standard trade list, one line per thing for the whole job, named with the drawing's code and product. So you get "CT1 - 13mm plasterboard ceiling; ceiling area", and not a pile of room-by-room fragments to stitch back together.

Why trust a quantity the AI never typed?

This is the bit that matters most. The AI never types a quantity. Every quantity is a shape (an outline, a line or a dot) or a pin on a published sheet, and Estimator+ works out the number from that shape and the sheet's printed scale.

Why's that a big deal? Because a typed number can be wrong in ways nobody can see. A shape can't hide like that. If a quantity's off, the shape behind it is off, and an outline running through the wrong wall is something your estimator can see just by looking at the plan.

So you check a shape on the drawing it was measured on, the same way you'd check a junior's markup, rather than taking a figure on faith.

Where does the scale come from?

From the sheet. The scale is proven on the sheet itself, either from two points and a printed length, or from the printed ratio where the sheet prints one. And if your estimator has calibrated a sheet themselves, that calibration is never replaced.

That's worth a moment, because scale is where a takeoff quietly goes wrong. A set that's been reduced, re-plotted or scanned can carry a ratio that no longer matches the page. A printed dimension string proves the scale whatever happened to the paper along the way.

Example: checking the vinyl on a fitout

Here's an invented example. A single-storey medical centre fitout, ground floor plan A-101 Revision C. The finishes legend codes the sheet vinyl FF2, so the line reads "FF2 - Sheet vinyl; floor area". The scale's proven from two points on the printed 7,200 mm dimension between grids B and C.

Switch on Review mode and every outline on A-101 carries the line's name. There are three FF2 outlines.

You don't have to take 86.4 on trust. The corridor's printed 2,400 wide and runs 36,000 between grids 1 and 6, so 2.4 × 36.0 = 86.4 m², which is what the outline gives. The other two you check by eye against the room walls they trace.

Then there's the store, 6.3 m². The plan hatches it as FF2, but the finishes schedule on A-601 lists it as FF3, sealed concrete. The AI assumed the schedule, left the store off FF2, said so, and asked you about it at the end of the run. Tell it the plan's right and the store's outline goes on the line, which then reads 186.5 + 6.3 = 192.8 m².

So what stays with your estimator?

Plenty, and that's how it should be. Every assumption is shown: each assumed height, extent and unit. Where two drawings disagree, like that store, it's flagged. The calls the documents leave open come back as questions, together, once the run's done, instead of the AI stopping halfway to ask.

Your estimator checks the takeoff and owns the price, the picks and the strategy. The manual tools are right there too, for area, length, count and wall height, so anything you'd rather measure yourself, you can.

Now the honest catch. Estimator+ reads drawings as PDFs, not CAD or BIM models. If your tenders are priced off a model, it won't take that off. And the takeoff needs the Claude or ChatGPT desktop app with your tender folder open. (Curious what ChatGPT can manage on its own? There's a post on whether ChatGPT can do a takeoff.)

What happens to the lines after the takeoff?

They go straight to work. The lines land in the priced tender, where every subcontractor's price sits side by side, next to the lines it covers. A rate follows the live quantity, so if that store joins FF2, the flooring rates move with it. A lump sum stays fixed to the quantity it was priced on.

You can export BOQs with marked-up plans, in PDF and Excel, for subbies to price from. The flooring subbies see the same outlines you checked, so they're pricing your 192.8 m² and can see where it is. The Excel exports carry live formulas, and they can price inside the sheet.

That's the join people usually cover with a measuring tool and a spreadsheet. Our Bluebeam and Excel post follows one wall type across it.

Want to see it on your own drawings?

Any vendor's sample job is picked to look good, ours included. So bring a tender you've already priced, ideally one with an addendum and a finishes schedule that argues with the plans.

We'll run the takeoff on it, at your office or on Google Meet, and you can check every shape against your own numbers. Book a demo and bring the set you know best.