Descartes Earnings: Why an EPS Miss Rarely Sinks This Stock
The Descartes Systems Group (TSX: DSG, Nasdaq: DSGX) reports fiscal second quarter earnings on September 10, 2026, after the market close, and the scorecard it carries in looks alarming. Yahoo Finance’s tracker shows the Waterloo logistics software company missing EPS consensus in seven of its last eight reports, at an average surprise of negative 11.6%. The stock rose the next session after five of those eight reports, including after every one of the last four misses, by 7.0%, 14.4%, 4.6% and 5.3%.
Both of those statements are true, and reconciling them is the useful work before this print. Below is what the street expects, why that miss record sets two different measuring sticks against each other, and what the stock has actually done after each of its last eight reports. One unit note first: Descartes reports in US dollars, so every company figure here is USD, while the TSX share price is Canadian. Street data is Yahoo Finance as of the September 4, 2026 close, and company figures come from Descartes’ own filings.
What the Street expects
Consensus for fiscal Q2 2027, the quarter ended July 31, 2026, is $0.69 USD in earnings per share, compiled from five analysts with a range of $0.56 to $0.79 (Yahoo Finance). Hold that number loosely until the next section, because the basis it is built on is the whole story with this company.
The estimate has drifted up rather than down. Ninety days ago the Q2 figure stood at $0.6675. It has been $0.69 at every checkpoint since, 60 days ago, 30 days ago and 7 days ago. The bar rose slightly after the last report and has not moved since.
The top line is where the real analyst effort sits. Revenue consensus is $199.5 million USD from ten analysts, in a tight range of $196.8 million to $202.7 million, against the $179.8 million Descartes reported for the year-ago quarter. That is roughly 10.9% implied growth. Ten analysts model the revenue line against five on the EPS line, and on the full year the revenue coverage widens to twelve: $812.2 million USD for fiscal 2027, up 11.4% on the $729.0 million Descartes reported for fiscal 2026, and $903.5 million for fiscal 2028.
The beat and miss record, and what it is actually measuring
Descartes has been scored as missing EPS consensus in seven of its last eight reports, with an average surprise of negative 11.6% (Yahoo Finance).
| Report date | EPS estimate | EPS actual | Surprise |
|---|---|---|---|
| 2024-09-04 | $0.45 | $0.40 | -11.9% miss |
| 2024-12-03 | $0.42 | $0.43 | +3.4% beat |
| 2025-03-05 | $0.55 | $0.44 | -19.6% miss |
| 2025-06-04 | $0.60 | $0.44 | -26.1% miss |
| 2025-09-03 | $0.52 | $0.48 | -7.0% miss |
| 2025-12-03 | $0.61 | $0.51 | -16.2% miss |
| 2026-03-11 | $0.56 | $0.53 | -5.8% miss |
| 2026-06-03 | $0.63 | $0.56 | -10.0% miss |
Consensus and prices: Yahoo Finance. Data as of September 4, 2026.
A company that undershoots analyst profit estimates seven times running is normally a company with an execution problem. Descartes does not look like one in its own numbers. In the most recent quarter, Q1 fiscal 2027, it reported revenue of $193.6 million USD, up 15% year over year, net income of $48.5 million at 25% of revenues, and GAAP diluted EPS of $0.55, up 34% from $0.41.
Here is the mismatch. Descartes publishes no adjusted EPS. Its headline non-GAAP profitability metric is adjusted EBITDA, and there is no company-issued “adjusted” earnings per share number anywhere in its release or shareholder report. For the June 2026 report, Yahoo carried a consensus of $0.63 and scored an actual of $0.56, while Descartes itself reported GAAP diluted EPS of $0.55 and basic EPS of $0.56 for that quarter.
We read the persistent gap as two different measuring sticks rather than eight quarters of operational shortfall: an adjusted-style consensus line on one side, a company that reports on a GAAP basis and never publishes the adjusted equivalent on the other. That reading also carries a warning about arithmetic. Yahoo’s year-ago EPS figure for this quarter is $0.479, while Descartes reported GAAP diluted EPS of $0.43 for that same quarter, Q2 fiscal 2026. Comparing the $0.69 estimate against the company’s own reported figures produces a growth rate that does not mean what it appears to mean, so the honest comparison in this report is revenue and adjusted EBITDA, like against like.
Two of the eight quarters were genuinely punished, and they are the ones where the gap was widest: March 2025 at a 19.6% shortfall and June 2025 at 26.1%. The market has not ignored the EPS line entirely. It has treated small gaps as noise and large ones as information.
What the stock did after the last eight reports
| Report date | Next-day move | 30 days later | 90 days later |
|---|---|---|---|
| 2024-09-04 | -3.5% | +3.7% | +22.3% |
| 2024-12-03 | +3.4% | +0.3% | -2.1% |
| 2025-03-05 | -8.6% | -14.2% | -0.7% |
| 2025-06-04 | -12.1% | -11.3% | -14.1% |
| 2025-09-03 | +7.0% | -4.5% | -15.9% |
| 2025-12-03 | +14.4% | +1.0% | -17.7% |
| 2026-03-11 | +4.6% | -6.9% | +11.1% |
| 2026-06-03 | +5.3% | +0.3% | +8.6% |
Consensus and prices: Yahoo Finance. Data as of September 4, 2026.
The averages, as history and not as a forecast. The next-session move across all eight reports averaged positive 1.32%, and the stock was higher the following day in five of eight. After the seven scored misses, the average next-day move was positive 1.02%. The single scored beat, December 2024, produced a 3.42% gain, which is one observation and not a pattern.
The more important pattern is what happened after the initial move. Thirty days later, the average was negative 3.95%, with only four of eight positive. Ninety days later, the average was negative 1.06%, with three of eight positive. The two largest next-day pops in the set illustrate it: the December 2025 report was followed by a 14.4% jump and then a 17.7% decline over the following 90 days, and the September 2025 report gained 7.0% the next session and was 15.9% lower 90 days on. The earnings-day reaction has repeatedly faded, and in the March 2025 and June 2025 cases the decline simply continued.
The growth track record behind the estimates
Source: Descartes annual shareholder reports (Q4FY26 and Q4FY24).
Across the five fiscal years to January 31, 2026, revenue went from $424.7 million USD to $729.0 million, a compound annual rate of about 14.5%. Net income went from $86.3 million to $163.8 million, GAAP diluted EPS from $1.00 to $1.87, and adjusted EBITDA from $247.5 million in fiscal 2024 to $329.5 million in fiscal 2026. The company reconciles adjusted EBITDA in its segment note, and the fiscal 2024 annual report carries no such reconciliation, so the earlier two years are not shown on that measure.
The most recent quarter fits that line. Alongside the 15% revenue growth, Descartes reported adjusted EBITDA of $89.8 million USD, up 20% year over year and 46% of revenues, on gross margin of 78% against 76% a year earlier, per its fiscal 2027 first quarter results release. Services revenue was $180.5 million, 93% of the total. Cash from operations was $75.1 million against $53.6 million a year earlier, and cash stood at $377.0 million at April 30, up $20.5 million in the quarter.
Two things that cash funds. The first is acquisitions. Descartes bought Utordo Ltd. (OrderMine), a UK provider of AI-powered forecasting and demand planning software for ecommerce businesses, on March 11, 2026 for about $2.3 million net of cash acquired plus up to $1.0 million contingent, and Idelic Inc., a provider of AI-powered driver safety and performance management solutions, on April 22, 2026 for about $25.3 million net plus up to $12.0 million contingent. Acquisition cash outflow in the quarter was $29.7 million, and the company counts five acquisitions since the beginning of fiscal 2026. That acquisition pattern, and the logistics network the software sits on, is why Descartes appears on our ranking of the best AI stocks in Canada, where the fuller business background sits.
The second is the buyback. Descartes repurchased and cancelled 305,000 shares for $20.8 million in the quarter, and a further 196,800 shares between May 1 and June 2, 2026. It pays no dividend, and none appears in the filings.
What the company itself calibrates
Descartes does not issue conventional guidance. It publishes a baseline calibration, and knowing the difference stops a reader from misreading the release.
At May 1, 2026, the company estimated baseline revenues for fiscal Q2 2027 of approximately $169.0 million USD and baseline operating expenses of approximately $102.5 million, giving a baseline calibration of approximately $66.5 million, or roughly 39% of baseline revenues (Q1 fiscal 2027 report to shareholders, page 25). Baseline revenues are the company’s defined term for visible, recurring and contracted revenues, and the MD&A states explicitly that the calibration is not a projection of net income or of adjusted EBITDA (definitions, page 24). It is a floor built from business already under contract, not a forecast of the quarter.
That framing explains the distance between the two numbers on this page. Analyst revenue consensus of $199.5 million sits roughly $30 million above the $169.0 million contracted floor, and everything in that gap has to come from new business, acquired revenue and expansion within existing customers. The company also states a standing expectation that in a typical year it may lose approximately 5% to 7% of aggregate annualized recurring services revenues in the ordinary course, before counting new customers (page 24).
What to watch in this report
Four things will be new information rather than a rerun of the above.
Revenue against both marks, the $199.5 million consensus and the company’s own $169.0 million baseline. Adjusted EBITDA, and whether the margin holds near the 46% of revenues posted in Q1. Any detail on how the five acquisitions since the beginning of fiscal 2026 are contributing, with Idelic the largest of the recent pair and its first full quarter of ownership falling in this period. And the baseline calibration Descartes issues for the following quarter, which it updates with every report and which is the closest thing this company gives to a forward number.
When and how it reports
Descartes reports after the market close on September 10, 2026. All eight of its last eight reports came after the close, so an evening release is the established pattern. It makes for an earnings day with bookends: Sobeys parent Empire reports before the open the same morning, covered in our Empire Q1 earnings preview, and Descartes closes the day after the bell.
What it means for holders
At the September 4 close of $109.13 CAD, Descartes fell 2.11% on a day the TSX slipped 0.33%, and it is down about 4% over five sessions. The 52-week range is $85.26 to $150.74 CAD, which puts the stock roughly 28% below its high and roughly 28% above its low.
The constructive reading: revenue consensus implies about 10.9% growth on a line ten analysts model, the company compounded revenue at about 14.5% annually over five fiscal years, adjusted EBITDA was 46% of revenues last quarter, and the headline miss record is scored on a profit measure the company does not itself publish.
The cautious reading, from the same data: the stock has spent the past year well below its high, the next-session gain after each of the last four reports had eroded by the 30-day mark in all four cases, and the two quarters where the EPS gap was widest, March and June 2025, brought next-day declines of 8.6% and 12.1%. A basis mismatch explains why small scored misses have been shrugged off. It does not promise that a large one would be.
Disclaimer: The content on bestcanadianstocks.ca is for informational and entertainment purposes only and does not constitute financial advice. Past performance is not indicative of future results. Always consult a qualified financial advisor before making investment decisions. Consensus and prices from Yahoo Finance as of the September 4, 2026 close; company figures (USD) from Descartes’ Q1 FY2027 release and shareholder report.
