DV Q3: AI and CTV Launches Aim For Efficient Growth Amid Retail Weakness
DoubleVerify's Q3 CY2025 came in soft against expectations, but the story isn't just the miss. It's about how new AI and CTV products could reset the company's growth and margin profile if adoption sticks. For product leaders, this quarter reads like a case study in shipping automation that cuts cost, expands coverage, and opens new revenue lanes.
By the numbers
- Revenue: $188.6M (11.2% YoY), vs $190.2M expected
- Adjusted EPS: $0.22 vs $0.27 expected
- Adjusted Operating Income: $50.66M (26.9% margin), above $49.37M expected
- Operating Margin: 11.2%, down from 15.2% last year
- Q4 Revenue Guide (midpoint): $209M vs $210.8M expected
- Q4 EBITDA Guide (midpoint): $79M vs $77.86M expected
- Market Cap: $1.51B
- Share price: $9.44, down from $10.96 pre-earnings
What stands out for product teams
- AI is doing double duty: it's the core of new products and a lever for operating efficiency.
- Social and CTV are gaining traction, with early signs of demand from large brands.
- Zero churn among the top 100 customers shows resilience and room for upsell.
- Retail headwinds are real and broad-based, so near-term growth depends on new product ramp.
AI verification suite: scope and impact
DV launched an AI Verification suite featuring Agent ID Measurement and SlopStopper. These tools use generative models to classify activity and spot synthetic content at scale. Management expects classification volume to double and efficiency to improve meaningfully as automation takes over routine review.
For product leaders, that's the blueprint: ship AI that compresses cycle time, scales coverage without linear headcount, and opens new pricing surfaces. The near-term benefit shows up in margin and throughput; the long-term benefit shows up in stickiness and cross-sell.
Social and CTV momentum
Social activation is one of the fastest growers, with Authentic AdVantage and Meta pre-screen tools winning early adoption. CTV measurement volumes rose 30% year over year, a signal that connected TV continues to pull budget and requires verification parity with display and video.
Management sized Authentic AdVantage plus Meta prebid solutions at a potential $120M to $160M annual revenue opportunity over time. Execution hinges on integration quality, partner alignment, and proof that these tools lift media efficiency (not just compliance).
Customer dynamics: stability with room to expand
Zero churn among the top 100 customers is a strong signal. Upsell activity was led by AI-powered solutions, suggesting existing relationships are the fastest path to monetizing innovation. This matters when a major vertical (retail) is soft-expansion can offset slower net-new adds.
Guidance, margins, and the automation thesis
Revenue guidance for Q4 sits slightly below expectations, but EBITDA guidance is above. The gap tells you where management is placing its bet: automation will help margins even if top-line growth is uneven.
The operating margin fell to 11.2% from 15.2% last year. Leadership expects AI-driven process gains to support margin expansion into 2026 as classification volume increases without equivalent cost growth.
Risk: retail softness and macro
Retail advertising was weak across the board, not just a few accounts. The base case assumes no major macro change, but any further pullback could weigh on results. Upside depends on the pace of adoption for AI verification, social activation, and CTV measurement.
What to watch next
- Adoption curve and revenue contribution from Authentic AdVantage and Meta pre-screen tools
- CTV measurement growth vs. industry benchmarks (30% YoY this quarter)
- Automation throughput: classification volume per FTE and unit costs
- International expansion tied to large platform integrations
- Retail ad spend stabilization and budget cycles into peak seasons
Product playbook: how to apply this
- Ship measurable efficiency: track and market time-to-classification, cost per review, and coverage uplift.
- Price on outcomes where possible: tie value to verified impressions, reduced waste, or fraud prevention.
- Integrate deeply with demand channels: prebid, pre-screen, and measurement partnerships drive adoption.
- Design for upsell inside existing accounts: bundle AI features as modular add-ons with clear ROI.
- Build synthetic content defenses now: classification quality, model drift checks, and explainability.
- Instrument margins: treat automation as a product with KPIs-latency, accuracy, cost per decision.
If you work on ad measurement or trust-and-safety, aligning to industry standards can speed partner approvals and reduce sales friction. For reference, see the IAB and the Media Rating Council (MRC) for guidance frameworks.
Skills and team enablement
Teams building AI-driven classification and automation benefit from focused upskilling-prompt engineering, evaluation pipelines, and product analytics for AI features. For curated, job-focused resources, see AI courses by job and automation-focused materials.
Bottom line: DV missed on revenue and EPS, but the product thesis is clear. If AI verification, social activation, and CTV keep scaling-and automation keeps trimming unit costs-the company can grow more efficiently even through choppy retail spend.
Your membership also unlocks:
