Applause 2026 State of Digital Quality Report: AI Use in Functional Testing Surges as Defects Rise

Applause's 2026 digital-quality survey found that more than 92% of respondents now use AI in software testing, up from 60% a year earlier. However, higher AI adoption has not yet translated into fewer quality issues, with 29% of respondents reporting an increase in issues, indicating that testing and quality-control challenges remain despite rapid deployment.

Analysis

The investable implication is not a broad AI-software demand signal; it is evidence that code-generation is expanding the defect surface faster than automated QA tooling is reducing it. That should favor vendors monetizing production observability, incident response and developer workflow controls—DDOG, DT, ESTC and GTLB—because failed releases create recurring telemetry ingestion, remediation and governance spend rather than one-off testing-tool purchases. The strongest revenue sensitivity should emerge over 6-18 months as enterprises move AI-assisted development from pilots into customer-facing systems, where reliability failures carry direct revenue and regulatory costs.

Near term, this is insufficient to change estimates for any public name: the source is a vendor-sponsored survey, lacks disclosed respondent mix, and provides no evidence of incremental budget allocation. The contrarian read is that AI coding may compress demand for seat-based testing software even while it raises demand for independent validation and production monitoring; vendors unable to prove defect reduction could face pricing pressure. Watch GitLab's net retention and Ultimate adoption, Datadog's log/APM consumption growth, and Dynatrace's net-new ARR over the next 1-3 earnings cycles; a deceleration in those measures would falsify the reliability-spend thesis.

Key Decisions for Investors

• No immediate directional trade on the survey alone; treat it as a watch signal rather than a catalyst, given the absence of public-company budget or contract data.
• Build a 6-12 month relative-value watchlist: long DDOG or DT versus short GTLB only if upcoming results show observability consumption accelerating while GitLab seat growth or net retention weakens. The thesis is AI-driven operational complexity shifting spend from development seats to production reliability; exit if GTLB Ultimate adoption and retention reaccelerate materially.
• For existing DDOG exposure, use the next earnings print to test the thesis: add only if management attributes incremental APM/logs or security-module consumption to AI application deployment and raises forward consumption expectations. A sub-guidance consumption outlook or sustained gross-margin pressure would invalidate the incremental-spend case.
• Monitor private-market read-throughs from Applause, Tricentis and SmartBear funding, pricing, or enterprise contract announcements. Evidence that outsourced/independent QA is winning budget would strengthen the case for public observability names, but would not by itself justify a direct listed-equity position.