Earnings Reaction · Q3 2026October 27, 2026 · reported after the close · call Wed Oct 28, 8:30 AM ET
TeradyneTER · NASDAQ
Revenue above the top of the guide, EPS 12% above the Street, record memory and integrated-systems test, Robotics profitable — and an in-line Q4 guide that the tape sold for 6%. Third semiconductor name tonight to lose a beat on a flat slope.
Last
$304.02
−6.1% after hours (close $323.68) · 52-wk $144.07–$487.91
Fair value (prob.-wtd)
$325
+7% vs AH · base $330
Q3 actual
$1,306M / $2.29
Street $1,268M / $2.05 · guide $1.20–1.30B / $1.85–2.15
Q4 guide
$1.25–1.35B / $1.95–2.25
Street $1,267M / $2.02 — in line at the mid
Gross margin
59.1%
non-GAAP · op margin 31.6% (33.7% in Q2)
Semi Test
$1,072M
Product Test $138M · Robotics $96M, profitable
Valuation
26.5× / 23.1×
FY26E $9.10 → ~$9.4 · FY27E $11.46 pre-print
Balance sheet
$461M cash
no debt · inventory +$171M q/q · capex $86M
Earnings reaction — beat the high end, guided in line, sold −6%. Q3 revenue $1,306M (+70% y/y, −2% q/q) vs. $1,268M consensus and the $1.20–1.30B guide; non-GAAP EPS $2.29 vs. $2.05 (guide $1.85–2.15); GAAP EPS $2.17. Semiconductor Test $1,072M with record Memory and Integrated Systems Test "driven by all things AI"; Robotics $96M, its best Q3 and profitable. Q4 guide $1,250–1,350M and $1.95–2.25 — midpoint $1.30B / $2.10, against $1,267M / $2.02 Street: in line on revenue, 4% above on EPS, and flat to Q3 at the midpoint. Non-GAAP op margin slipped to 31.6% from 33.7% in Q2 as opex rose 3.5% q/q on flat revenue; inventory built $171M (+42% q/q) and prepayments $165M — supply being staged for 2027 demand, or a build the market reads as the top. Cash $461M, revolver repaid, $99M of buybacks. Why −6%: the Q3 guide had already set a sequential decline from Q2's $1,329M; the beat put Q3 back near flat, so the Q4 guide reads as two flat quarters off the Q2 peak at 79% implied vol in a stock 37% off its high. Same pattern as NXPI (−8%) and STX (−7%) tonight: deceleration gets sold whatever the absolute level. Call is Wednesday 8:30 AM ET — the 2027 framing is the whole trade.
What the quarter said
| Line | Q3 26 | Q2 26 | Q3 25 | Read |
|---|
| Revenue ($M) | 1,306 | 1,329 | 769 | Beat the $1.30B guide top; −2% q/q |
| Semi Test ($M) | 1,072 | — | — | Record memory + ISG test; SoC compute the other leg |
| Non-GAAP GM | 59.1% | 59.8% | 58.5% | E&O provision $10.9M vs $3.6M |
| Non-GAAP op margin | 31.6% | 33.7% | 20.4% | Opex +3.5% q/q on flat revenue |
| Non-GAAP EPS | $2.29 | $2.47 | $0.85 | +169% y/y |
| Operating cash flow ($M) | 314 | — | 49 | Inventory +$171M, prepayments +$165M |
| Robotics ($M) | 96 | — | — | Profitable; best Q3 ever |
The balance sheet tells the forward story: inventory went from $380M at year-end to $575M, accounts payable from $269M to $517M, and prepayments from $428M to $614M. That is a company staging material for a larger 2027, not one winding down — but it is also $340M of working capital absorbed in nine months, which is why operating cash flow ($314M) is half of net income.
The one metric that matters
The 2027 memory-test outlook on Wednesday's call. Q4 is guided flat; what matters is whether HBM/DRAM test demand (the record this quarter) keeps compounding into 2027 or has pulled forward, and whether SoC compute test for AI accelerators steps up to replace it. Secondary: the gross-margin path back toward 60% once E&O normalizes, and whether the inventory build converts to Q1 revenue or sits.
Where it touches the threads
- Memory cycle. TER is the test-capacity read on MU/SNDK/STX capex: record memory test revenue says the memory makers are still adding capacity. STX's 2× capex tonight is the same signal from the HDD side.
- AI capex vs. backlog. "Demand extended across the AI value chain" plus a $340M working-capital build is backlog evidence; a flat Q4 guide is the deceleration the "peak capex" camp reads. Both are true tonight.
- Robotics turning profitable removes a bear argument that has run for three years; it is still 7% of revenue.
Valuation and scenarios
At $304 after hours the stock is ~$47.5B of market cap (156M shares) with no debt, 26.5× FY26 (consensus $9.10, tracking ~$9.4 after tonight) and 23.1× FY27 ($11.46). Test equipment has historically traded 20–30× through cycles; the stock is mid-range on an estimate that assumes 2027 growth of ~25%. Fair values are independent house estimates on FY27 earnings power.
Bull25% probability
$420+38%
- Memory + compute test both grow in 2027; GM back to 60%
- Q1 guided up sequentially; inventory converts
- FY27 EPS ~$13 × 32×
Base50% probability
$330+9%
- Q4 delivered; 2027 up mid-teens on AI compute test, memory flat
- Op margin 32–34%; buyback resumes
- FY27 EPS ~$11.5 × 29×
Bear25% probability
$220-28%
- Memory test digests in 1H27; SoC does not fill the gap
- Inventory becomes E&O; margin to high-50s
- FY27 EPS ~$9 × 24× — retest of the spring range
Probability-weighted fair value
$325
vs. after-hours $304.02
+7%
In-line guide, in-line value: the stock is where the base case says it should be. The call decides which tail.
What to watch
- 2027 framing — memory vs. compute test mix and any numbers
- Q1 seasonality language — flat or down from Q4
- Inventory — backlog-driven staging or finished goods
- Gross margin path and the E&O provision
- Robotics margin durability
- Buyback pace with $461M cash and no debt
The desk read: three semiconductor beats sold tonight (NXPI, STX, TER) on the same rule — the slope matters more than the level. Teradyne's is the mildest case: the guide is flat, not down, and the working-capital build argues 2027 is bigger. The after-hours move is pre-call; Wednesday morning's 2027 commentary is the print. Levels: $300 then $280 below; $324 (close) and $340 above.
Sources
- Teradyne Q3 2026 results, Business Wire via IR — https://investors.teradyne.com/news-events/press-releases/detail/455/teradyne-reports-third-quarter-2026-results
- 8-K to follow on EDGAR, CIK 97210 — https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000097210&type=8-K
- Pre-print consensus: Briefing/Schwab workbook. Prices and after-hours quotes: Schwab market data, Oct 27, 2026 ~4:15 PM CT.
For educational/entertainment purposes. Sikuli Research publishes independent opinions; nothing here is investment advice, a research report under FINRA rules, or a solicitation to buy or sell any security. Fair-value scenarios are the author's assumptions and may change without notice. Prices are as of the time stated and should be confirmed against a live feed. The author may hold positions, long or short, in securities mentioned.