Bentley Trades a Maximo Integrator for a Seismic One — and the Tape Approves
ARR accelerated to 13% and retention hit a record 110%, but the quarter's real work was a portfolio swap, a per-asset pricing thesis, and a quiet demotion of the AI-in-design story to the far side of 2027.
BSY · Earnings Call · 2026-11-05
A 44% rally into a fifth-of-peak multiple
Bentley walked into this print with the tape already voting. The stock is up 43.7% in ninety trading days, an unbroken up-17-week run, and it follows a 46-week stretch that took 48% off the shares. The long arc is a de-rating that has been brutal and is only now being contested: price-to-revenue sat at 5.6x at the last quarter-end, roughly half the year-ago level and a fraction of the 21.0x touched in 2021Q2. The 2021 peak close was $71.17; the current drawdown from that high is still 36.9%. So this is not a breakout story so much as a compounding franchise that got thrown in the discount bin and is being fished back out.
The operating evidence supports the fishing. Executive Chair Greg Bentley opened by noting the company's key metric “surpassed our annual outlook range” — Gregory Bentley, Executive Chair · 2026-11-05 after accelerating in each quarter of 2026 to 13% constant-currency ARR growth — the second consecutive quarter of acceleration. Net revenue retention hit a record 110% against a 99% account retention rate, and sequential constant-currency growth of 3.1% was entirely organic. Total revenue of $422 million grew 12.2% year over year, and adjusted operating income less stock comp printed a 32.1% margin versus 28.1% a year earlier. On the filed numbers, revenue has grown 126% over seven years, and free cash flow less stock comp has reached $489 million on a trailing-twelve-month basis.
The swap nobody led with
The headline was ARR, but the substance was a portfolio exchange executed in the same week. Bentley divested Cohesive, its captive global Maximo systems integrator, merging it into UK-based Naviam in exchange for a significant minority stake. It also acquired Eliis, a seismic-interpretation specialist, folding it into Seequent. One services business out; one software business in.
The Cohesive post-mortem is unusually candid for a prepared script:
The enlarged Cohesive would go on to encounter sustained headwinds as Maximo customers responded slowly to IBM's enforced migration from sunsetted on-premise Maximo to managed services... digital twin integration with Maximo remains nascent.
The financial consequences are real and unfriendly to optics. Cohesive was roughly half of services revenue and about 3.5% of total revenue year-to-date; CFO Werner Andre guided that “beginning in the fourth quarter, we, therefore, expect services revenues to decline year-over-year through 2027” — Werner Andre, Chief Financial Officer · 2026-11-05, with a ~$12 million hit to Q4 revenue and $55–60 million of higher cash taxes on the transaction. Subscription mix rises from 92% to roughly 95% — higher quality, worse optics.
Eliis is the offset, contributing approximately 100 basis points of ARR growth and, per management, roughly margin-in-line and accretive to the revenue multiple. Net effect on the reported ARR line is deliberately close to a wash, while the underlying mix is cleaner. That is a portfolio trade dressed as a divestiture.
The keystone thesis: charge per asset, not per engineer
The most consequential thing said on the call was not a number. It was Greg Bentley's framing of Asset Analytics as the bridge from a four-decade per-user model to a per-asset one: “our revenues per asset should be comparable to our revenues per user over time” — Alexei Gogolev, Analyst · 2026-11-05. That is the entire TAM argument in a sentence — Bentley has historically been capped by the global census of engineers; infrastructure assets are not so capped.
The current manifestations are concrete rather than conceptual: strategic expansions in telecommunication tower inspection, including deals with tower equipment manufacturers, a natural extension into distribution poles, and further DOT road-monitoring wins. Roughly half of the improvement in sequential growth over last year came from large expansions in existing accounts, with Asset Analytics doing the heavy lifting. The ambition, as Greg put it, is that instant-on digital twins finally let owner-operators use the modeling and simulation they already bought for CapEx during the OpEx life cycle.
The catch is timing and texture. This business is lumpy, sold in major procurements, and last year's fourth quarter was particularly strong — a comp the CFO explicitly flagged. It is also priced per asset per year, which means it will never look like the metronomic subscription line.
What stopped being said
The keyword map tells a story the transcript soft-pedals. In the prior quarter, electric grid was a top-two theme and permit reform a top-ten one; this quarter both were the largest conceptual decliners in the company's own keyword set. Grid is not weak — Power Line Systems growth was again called out as standing out — but it migrated from headline to supporting cast, absorbed into the Asset Analytics story as distribution-pole inspection. Permitting reform, which dominated the prior call's narrative, surfaced only when an analyst asked, and Greg's answer was notably hedged: bipartisan, committee-approved, but likely not in the lame-duck session.
The second retreat is subtler and more important. AI in design has been the romantic part of this equity story for four quarters. On this call, Nicholas Cumins drew the line plainly: “we don't expect monetization of AI in design to be a material growth driver in 2027. It will be the year where we start.” — Nicholas Cumins, Chief Executive Officer · 2026-11-05 That is consistent with what he said three months ago — “we're not planning to monetize this year. We're planning to start monetizing next year” — Nicholas Cumins, Chief Executive Officer · 2026-08-06 — but the addition of "not material in 2027" is a deliberate lowering of the bar, and it lands while the company is asking investors to pay for applications at machine speed on faith. The AI in design payoff is now a 2028 conversation.
One genuinely new revenue line does exist today: seismic interpretation via Eliis, which management argues is no longer an oil-and-gas-only signal as acquisition costs fall — geothermal, carbon capture, deeper mineral detection, and civil tunneling. It is the same subsurface logic that has powered Seequent, extended laterally.
The contrast worth holding
There is a striking divergence between Bentley's narrative and the tape it sits inside. Cumins cited "the scale of private investment in AI data centers" as a portfolio-wide tailwind, and Greg turned the post-election scrutiny of data-center siting into an opportunity — “we can imagine site plans for data centers now and needing to include infrastructure digital twins, and we actually see that as an opportunity.” — Gregory Bentley, Executive Chair · 2026-11-05 Meanwhile the global tape's single worst thirty-day cluster is AI data centers, down roughly 31% across 55 decliners in that window. Bentley is up 44% over the same stretch. The market is not paying for the data-center adjacency; it is paying for recurring revenue, a 1.7x leverage ratio that fell from 1.9x, $168 million of buybacks year-to-date, and the fact that net debt has been cut by roughly $650 million since the 2022 trough.
That is the honest read of this quarter. Bentley did not accelerate because of AI hype; it accelerated on mining, grid and lumpy asset inspection, while simultaneously trading a problem child for a strategic stake and a seismic asset. The stock's re-rating is a bet that 5.6x sales was wrong for a 95%-subscription business compounding ARR at 13%. The risk is that the market wanted an AI-in-design story and got a very good industrial software story instead.