Manhattan Associates has spent years moving customers of its warehouse, order-management and point-of-sale software from perpetual licenses onto cloud subscriptions, and the conversion is far enough along that the company's fastest-growing number never touches the income statement. Contracted revenue not yet recognized — what the company calls remaining performance obligation — reached $2.47bn at the end of June, up 23% from a year earlier. Reported revenue grew 9.3%, to $297.8m.

That gap is arithmetic, not spin. A license sale lands largely at signature; the same customer on a subscription pays over a multi-year term, so the meter compounds while the profit-and-loss statement lags. What turns the mechanism into a story is that the market accepted the argument all at once. Manhattan's second-quarter report on 28 July beat consensus on both lines — adjusted earnings of $1.39 a share against $1.32 expected, revenue against $287.7m — and the shares gained roughly a third across the two sessions spanning it, with press coverage attributing about 27 points of that to 29 July alone. At $225.83 the stock sits at a 52-week high on about 41 times consensus 2026 earnings of $5.50, against roughly 25 times at its 27 February close of $135.43. A full year of de-rating was undone in a month, for consensus earnings growth of 10.3% this year and 11.5% next.

What the meter says

The bookings side earns its keep. Cloud revenue reached $127m, up 26%, in a third consecutive record-bookings quarter, and management raised full-year guidance to revenue of $1.160–1.166bn, cloud revenue of $505.5m and backlog of $2.62–2.68bn — 18–20% growth, which the company expects to land toward the high end. Conversions of existing on-premise customers made up more than 40% of new cloud bookings and new customers more than 25%, so the pipeline is not purely a migration of the installed base. Excluding the license runoff, revenue grew 13% rather than 9%.

The socket is contested. Manhattan sells warehouse management against Blue Yonder, SAP, Oracle and Körber, and ABI Research's ranking puts the top five vendors together at only 25–30% of that market — a fragmented field in which a stated win rate above 70% is a claim about deals seen, not share held.

What the accounts say

The same quarter that produced record backlog produced falling profit. Generally accepted accounting principles operating income dropped 10.2% to $66.2m; gross margin compressed to 53.2% from 56.8%. A restructuring charge of $8m, worth $0.11 a share, explains part of it, and management said the savings are being reinvested with no margin benefit this year. On an adjusted basis operating margin was 34.9%, and operating cash flow rose 22% to $91m. The balance sheet carries $186m of cash and no debt, with $275m of a $500m buyback authorization already spent this year. A currency tailwind of roughly one point in the first half is expected to reverse to a one-point headwind in the second.

The re-rating was not Manhattan's alone

Eight specialized enterprise software names moved together after 29 July, and the ones with no transaction meter moved hardest: Intapp rose 39.7%, AppFolio 25.4%, Blackbaud 16.6% on revenue growth of 3.0%. Bentley Systems, which sells infrastructure-engineering software and grew 12.8%, rose 2.1%. Software broadly had fallen about 27% from October 2025 to mid-July on the belief that AI would erode software economics, then bounced roughly 15% from 22 July as earnings beat and Nvidia's Jensen Huang publicly dismissed the disruption thesis. Blackbaud, at 9.3 times forward earnings, rallied while promising very little: "organic total revenue growth of 4% to 6% annually," chief executive Michael Gianoni told investors on 29 July, alongside a target of double-digit annual earnings-per-share growth through 2030. Descartes Systems, whose customs-filing network bills per transaction, rose 2.8% and reports on 10 September.

The verdict

Manhattan earns the bookings half of its move: the backlog is contracted, the cloud line is compounding at 26%, and guidance went up rather than down through a soft retail spending cycle. What the business does not yet earn is the multiple. Roughly 41 times forward earnings is a growth-software price paid for around 10% consensus earnings growth, against 26.9 times for Bentley and 9.3 times for Blackbaud, and it was reached in a month in which the whole group re-rated for reasons that had nothing to do with warehouses. Backlog is a promise about future revenue; it is not evidence that reported margins stop compressing while services and reinvestment carry the mix.

Manhattan has not yet set a date for its third-quarter report. When it comes, the backlog figure will arrive in the same release as the earnings it is meant to justify — and for the first time in a year, it is the one holding up the price.