Start with the arithmetic
Strip away the philosophy and this is a math problem: what does one estimating seat cost over five years? Here's the field, using vendor list pricing (verified July 2026 — check current pricing before you buy):
| Option | Year 1 | Years 2–5 | 5-year total / seat |
|---|---|---|---|
| Groundwork Takeoff perpetual, no maintenance | $2,500 | $0 | $2,500 |
| Groundwork perpetual + optional maintenance | $2,500 | $295/yr | $3,680 |
| Groundwork Takeoff subscription | $950 | $950/yr | $4,750 |
| Bluebeam Revu Core → Max | $330–590 | $330–590/yr | $1,650–2,950 |
| PlanSwift | $2,000 | $2,000/yr | $10,000 |
| Kreo Pro | $2,100 | $2,100/yr | $10,500 |
| STACK | $2,988–3,588 | same/yr | $14,940–17,940 |
| Togal.AI | ~$3,588 | ~$3,588/yr | ~$17,940 |
Two honest observations before we go further. First, Bluebeam is the cheapest line on that table — but Revu is a PDF and markup platform, not a takeoff-and-estimating tool; its AI sits only in the $590 Max tier, and you'll still need somewhere for the estimate to live. Second, our own subscription is on the table too, and at five years it costs nearly double our perpetual. That's not an accident of pricing; it's the entire argument of this article. Subscriptions are the expensive way to buy software you intend to use for years — which is exactly how estimating software gets used.
The break-even is simple: at $950/yr versus $2,500 once, ownership pays for itself during year three. Every takeoff after that is running on a tool that's already paid for. Against the $2,000–3,588/yr competitors, the perpetual pays for itself before the end of year two.
The history lesson vendors hope you skip
"But subscriptions guarantee updates and support!" True. The uncomfortable part is what happened to people who chose ownership when it was offered:
- Bluebeam, 2023. Perpetual sales ended in September 2023. Revu 20, the last owned version, hits End of Support this July and End of Life on December 31, 2026 — at which point it loses Studio and license transfers. An "owned" Revu seat is becoming a stranded one on a published schedule.
- PlanSwift, 2025. Users who paid for "lifetime" licenses were told to start paying annually or lose access. An April 2025 Capterra review of the episode is titled "Bait & switch scammers." We wrote up the full story here.
- Drawboard. The Drawboard PDF product converted its paid apps to subscriptions, generating the same user backlash — a smaller echo of the same play.
What "perpetual" has to actually mean
For a perpetual license to be worth paying for, three things must be true, in writing:
1. The grant is irrevocable. "Perpetual" describes duration; "irrevocable" removes the vendor's exit. You want both words in the grant clause. A perpetual-but-revocable license is a subscription the vendor bills whenever it feels like it.
2. Non-payment of maintenance stops updates, not the software. Maintenance should buy new versions and support. Skip it and your installed version must keep working — fully, not in some "reader mode." If maintenance lapse deactivates the product, you never owned it.
3. The software survives the vendor. If every launch requires a license server's blessing, your ownership expires whenever that server does — via shutdown, acquisition, or a strategy pivot. Offline-capable software that runs without permission is structural protection no contract clause can substitute for.
This is why Groundwork Takeoff's perpetual is $2,500 with optional $295/yr maintenance, the no-revocation language goes in the order form, and the app itself is offline-first — it's a native Mac and Windows application that does not need our servers' permission to open your plans. (Subscriptions exist too, at $95/mo or $950/yr with crew discounts at 3+ and 5+ seats, for teams that value flexibility over five-year cost. Both are public.)
When subscription genuinely is the right call
Fairness cuts both ways. Rent instead of buy if: you're a seasonal or per-project operation that scales seats up and down; you're trying a tool before committing (that's what trials and monthly billing are for); or you expect to switch platforms within two or three years anyway. Cash-flow reality matters too — $950 hits a small shop's February differently than $2,500 does. The problem was never that subscriptions exist. It's that for most of this industry, the option to own was quietly removed so the comparison could never be made.
Six questions to ask any vendor before you buy
- "Is the perpetual license irrevocable, and will you put that word in the order form?" A no or a dodge answers everything.
- "What exactly stops working if I never pay another dollar?" Get the list in writing. "Updates and support" is acceptable; anything touching core function is not.
- "Does the software run without contacting your servers?" Ask for the offline behavior specifically — how long, and what degrades.
- "What happens to my license if you're acquired or shut down?" Vendors hate this question, which is why it's a good one.
- "Can I export all my data — takeoffs, assemblies, estimates — to open formats?" Excel/CSV export is your leverage in every future negotiation.
- "What did you do to your last generation of perpetual customers?" Past behavior is the only pricing forecast that's ever been reliable. The receipts are usually public.
The bottom line
Over five years, one seat of takeoff software costs anywhere from $2,500 (owned) to nearly $18,000 (rented, at the top of the market). The industry's shift to subscriptions moved thousands of dollars per seat from your margin to theirs, while the revocation history shows what "trust us" was worth without contract language. Do the math for your shop, ask the six questions, and whatever you buy — get it in writing.
Own your tools again.
14-day free trial · every feature · no credit card. Then $950/yr — or $2,500 once, irrevocable, in writing.
Start free trial