inTEST’s Q4 beat is less a semi recovery call than a backlog-and-mix reset for 2026

inTEST reported the kind of quarter that changes the burden of proof, though not enough to declare the cycle healed. On the street-comparison basis, Q4 FY2025 revenue was $32.8 million versus the $31.2 million estimate, a +5.1% surprise, while EPS was $0.16 versus the $0.16 estimate, a 0.0% surprise. That combination matters because it says the upside was not a broad earnings breakout, but a revenue and gross-margin recovery that still had to run through a cost base that is large for the current revenue scale. The print is constructive because Q4 revenue stepped up sharply from the Q3 trough and gross margin improved, but the right interpretation is disciplined: this is a company emerging from a semiconductor-led downturn through diversified demand, backlog conversion, and cost actions, not a company calling a semiconductor equipment upcycle.

That distinction is central because the reported trajectory has been uneven for several quarters. Revenue moved from $36.6 million in Q4 FY2024 down to $26.6 million in Q1 FY2025, $28.1 million in Q2 FY2025, and $26.2 million in Q3 FY2025 before rebounding to $32.8 million in Q4 FY2025. Gross margin followed a less linear but still important path, from 39.7% in Q4 FY2024 to 41.5% in Q1 FY2025, 42.6% in Q2 FY2025, 41.9% in Q3 FY2025, and 45.4% in Q4 FY2025. The latest quarter therefore looks like a genuine operating recovery from Q3 FY2025, but it also sits below the $36.6 million revenue level of Q4 FY2024, and the company’s own commentary makes clear that the year-over-year comparison was still pressured by semiconductor weakness. The right market reaction should reward the sequential inflection while still testing whether Q4 was a durable run rate.

The sequential improvement looks more credible because management provided enough bridge detail to separate demand recovery from timing help. CFO Duncan Gilmour said Q4 revenue increased by $6.6 million, or 25%, from $26.2 million in Q3 to $32.8 million, reflecting a gradual improvement in capital spending, momentum in new product sales, and about $2 million of revenue that slipped out of Q3. That last clause is the necessary caution: some of the Q4 rebound was catch-up, not pure acceleration. But the composition of the increase was not narrow. Industrial accounted for $3.3 million of the increase, Defense and Aerospace $3.2 million, Life Sciences $2.1 million, and Auto EV about $1 million, partially offset by a $2.9 million decline in semi. That is the quarter’s most important business mix message. The rebound did not come from semiconductor; it came despite semiconductor.

That mix point also explains why the full-year narrative is better than the semiconductor equipment label alone would imply. Management said that, compared to Q4 FY2024, revenue declined by $3.8 million, reflecting lower Auto EV, Semi, and Safety and Security revenue totaling $11.7 million. It also said roughly three quarters of the nearly $17 million decline between 2024 revenue and 2025 revenue was directly attributable to semiconductor market weakness. Those statements place the year’s weakness squarely in the company’s more cyclical exposure, while the order commentary shows that non-semi end markets are now carrying more of the forward load. Nick Grant said life sciences orders were up 137% year over year on a full-year basis, auto EV orders were up 89%, and industrial was up 53%. Those are not revenue guarantees, but they do support management’s case that the business has more than one route to growth in 2026.

The backlog is the reason that 2026 guidance deserves more attention than the Q4 beat itself. Management ended the year with backlog of $53.9 million, up 9% sequentially and 36% year over year, and Grant’s wording around duration is important because it cuts both ways: “Approximately 60% of our backlog is expected to ship beyond 2026, providing forward visibility into the year.” The commitment embedded in that sentence is not simply that backlog is high; it is that a substantial portion is already dated beyond the current fiscal year. That supports visibility, but it also means not all backlog is available to de-risk near-term revenue. For a small-cap equipment supplier, that is a better problem than a depleted book, yet it prevents a simplistic backlog-to-revenue conversion argument. The 2026 thesis rests on enough backlog converting this year, supplemented by orders, while the longer-dated piece anchors a broader multiyear demand base.

The guidance framework is therefore more balanced than a headline reading might suggest. Management projects FY 2026 revenue of $125 million to $130 million, and Gilmour said that “at the midpoint, this represents growth of approximately 12% over 2025, or $113.8 million.” The important language is that the outlook is supported by diversified demand in Industrial, Aerospace and Defense, Auto EV, and Life Sciences, but does not contemplate a meaningful rebound in Semi sales. That is the crux of the investment case after this call. If semiconductor demand merely stops getting worse, the company can still grow on management’s plan; if semiconductor turns, there may be optionality; if it stays weak, the guide is not explicitly built on a rebound. For Q1 2026, management projected revenue of $31 million to $33 million, which frames Q4 as a step-up quarter rather than a straight-line acceleration into the next period.

The capacity story explains the margin guide, because Q4’s 45.4% gross margin included favorable mix that management does not expect to repeat fully in Q1. Gilmour’s Q1 guide was unusually useful because it identified both the number and the reason: “For Q1 2026, we project revenue of $31 million to $33 million, gross margin of approximately 44%, this is a step down from the 45.4% we delivered in Q4, primarily reflecting expected Q1 product and customer mix versus Q4's particularly favorable Alphamation contribution, operating expenses of $13.3 million to $13.7 million, Q1 operating expenses reflect the typical first-quarter annual compensation resets, and amortization of $800,000.” The phrasing matters because it does not blame under-absorption or weak demand; it points to mix and compensation resets. For the full year, management guided gross margin of approximately 45%, tying that to higher volume, continued manufacturing efficiency, and new higher-margin products. The implication is that Q4 margin was not a new quarterly floor, but the full-year margin goal is still above the Q1 guide.

The operating leverage discussion is the part of the print that most directly connects revenue recovery to earnings power. Gilmour said the company generated $6.6 million in incremental revenue while absorbing only $1.4 million in incremental operating expenses, reducing operating expenses as a percentage of revenue to 41.5%. He also said Q4 FY2025 operating expenses increased $1.2 million year over year, from $12.5 million in Q4 FY2024 to $13.6 million in Q4 FY2025. That shows both sides of the model. Sequentially, inTEST demonstrated leverage as revenue recovered from Q3 FY2025. Year over year, the expense base is still higher despite lower revenue than Q4 FY2024. Adjusted EBITDA was $3.2 million in Q4 FY2025, representing an adjusted EBITDA margin of 9.7%, up from the Q3 FY2025 trough of $400,000 at a 1.5% margin. But for full-year 2025, adjusted EBITDA was $4.0 million, representing an adjusted EBITDA margin of 3.5%, compared to $10.8 million and an 8.3% margin in full year 2024. The quarter proved the model can respond when revenue returns; the year proved the fixed-cost sensitivity remains material.

That sensitivity is also why EPS alone understates the strategic importance of Q4. The quarterly history shows diluted EPS of -$0.19 in Q1 FY2025, -$0.04 in Q2 FY2025, -$0.08 in Q3 FY2025, and $0.10 in Q4 FY2025 on the company’s quarterly history basis, while the street-comparison basis showed $0.16 versus $0.16. The company’s own account of the full year was still loss-making, with a net loss of $2.5 million and net loss of $0.21 per share. Liquidity was not framed as a stress point, with full access to a $30 million delayed draw term loan facility and a $10 million revolver, and capital expenditures expected at 1% to 2% of revenue. That matters because the 2026 plan does not appear to require a capital intensity reset. The model is being asked to grow through backlog, mix, new products, and manufacturing efficiency, not through a step-change in investment.

The call tone supports the interpretation that management is more confident, but not uniformly promotional. The tone history shows overall sentiment rising to 0.42 in Q4 FY2025 from 0.13 in Q3 FY2025, and guidance_tone improving to 0.45 from -0.08. Q&A sentiment also improved to 0.30 from 0.18, while uncertainty edged down to 70.6 from 72.2. That is consistent with a team that had better numbers, better backlog, and a clearer guidance framework. At the same time, tone_confidence slipped to 0.36 from 0.39, prepared_sentiment fell to 0.02 from 0.13, and qa_evasiveness rose to 70.9 from -34.6. The delivery therefore had two layers: the prepared guidance was anchored in specifics, but the Q&A pattern still carried caution. For investors, that is not necessarily negative. It fits a company guiding recovery while explicitly declining to underwrite a meaningful semiconductor rebound.

That tone pattern also lines up with the peer context, where inTEST remains a different kind of equipment story than the larger Japanese test and assembly names. The peer table shows several companies with much stronger latest-quarter growth and higher gross margin, including ATEYY with revenue YoY of +43.8% and gross margin of 67.4%, DSCSY with revenue YoY of +12.3% and gross margin of 70.8%, and 6871.T with revenue YoY of +48.3% and gross margin of 47.3%. Against that backdrop, inTEST’s Q4 FY2025 revenue YoY of -10.3% and gross margin of 45.4% are not sector-leading numbers. The comparative point is not that inTEST is outgrowing the test and assembly complex; it is that its recovery is more idiosyncratic, tied to diversified industrial, life sciences, auto EV, and defense demand rather than the broad semiconductor momentum that appears in parts of the peer set. That makes the story less clean, but also less dependent on a near-term semi snapback than investors might assume from the ticker’s sector classification.

The supply-chain read-through is limited by the disclosure set, which is itself useful discipline. The data identify no named customers of inTEST and no named suppliers to inTEST, so this print should not be used to infer a direct order change for any specific counterparty. The read-through is instead end-market rather than company-specific: industrial, defense and aerospace, life sciences, and auto EV capital projects are contributing to orders and revenue, while semiconductor remains the area of weakness. For suppliers and customers not named in the data, the quarter says only that inTEST’s pockets of demand are improving outside semi, not that any identifiable customer program has accelerated.

The bottom line is that Q4 FY2025 was a credible reset, not a full victory lap. The beat was real on revenue, EPS met the street, gross margin recovered to 45.4%, and backlog of $53.9 million gives management a firmer base for a FY 2026 revenue outlook of $125 million to $130 million. The more investable part of the story is the quality of that outlook: management is asking investors to believe in diversified demand and execution, not in a semiconductor rebound it has not embedded. The risks are equally clear. Q1 gross margin is guided to approximately 44%, Q1 revenue is guided to $31 million to $33 million, operating expenses remain substantial, and backlog duration means a large portion does not ship inside 2026. Still, after a year in which full-year 2025 revenue was $113.8 million and net loss was $2.5 million, the company has put forward a plausible path back to growth and better profitability. The proper conclusion is constructive but bounded: inTEST has earned more credit for execution and visibility, while the next proof point is whether Q4’s leverage can survive normalizing mix and a still-muted semiconductor market.