AI Media Spend Surges While GenAI Adoption Stumbles on Data Quality Issues

AI media budgets are rising while in-house generative AI stalls: Inside Mediaocean's 2026 H1 Outlook

Marketers are putting money where the attention is. Mediaocean's 2026 H1 Advertising Outlook shows 54% plan to increase AI media spend, outpacing search (47%) for the first time in the survey's history. At the same time, 42% report data quality issues that block broader generative AI implementation.

Demand is shifting to conversational AI platforms and video-first environments, even as internal tooling and workflows lag. If you run media, creative, or analytics, this split is your roadmap: lean into where attention is going, and fix the plumbing that holds AI back.

About the survey

Mediaocean surveyed 320 marketing professionals in November 2025 via SurveyMonkey, including brands, agencies, media companies, measurement firms, and tech platforms. It's the ninth wave of a series running since late 2021.

Where budgets are moving in H1 2026

• AI media: 54% plan to increase spend, surpassing search (47%).
• Connected TV (CTV) and digital display/video: 63% plan to increase each.
• Social platforms: 61% plan to increase.
• Influencer/creator: 53% plan to increase, edging past search.
• Retail media: 40% plan to increase; most others will hold steady.
• Traditional media declines: print (-49%), local TV (-36%), national TV (-34%); radio and OOH also down.

CTV's share has already doubled from 14% (2023) to 28% (2025). The current wave suggests that trend continues into 2026.

Orchestration and AI are now tied as top capabilities

AI and cross-platform orchestration each earned 39% as "most critical capabilities" under current macro conditions. Translation: teams need unified planning, creative, and activation across CTV, social, display, and retail media-without swivel-chair work.

• Performance-driven paid media: 51% (still the north star).
• Brand advertising: 48%.
• Measurement and attribution: 48%.
• Privacy concerns: up 33% since May 2025, now at 24% of respondents.
• Automation: 30%-the practical partner to AI at scale.
• First-party data mastery: down to 19% as teams prioritize near-term performance.

Consumer trend radar: AI still leads, CTV surges

• Generative AI: 70% (still the top trend).
• Connected TV/streaming: 63%, up 24% from May 2025.
• TikTok and social video: 43%, a strong climb.
• Consumer privacy: 28%; political/advocacy: 18%.
• Gaming: 13%; metaverse: 12%.
• Sustainability/carbon impact: newly tracked at 17%.
• E-commerce everywhere: down to 27% as attention shifts to AI and video-first discovery.

Creative remains the bottleneck: 72% of marketers reuse or lightly tweak assets instead of building platform-native work. That leaves performance on the table in CTV and short-form video.

How teams are using generative AI (and where it stalls)

• Now common: data analysis (43%), market research (43%).
• Expanding: creative development (33%), campaign optimization (31%), creative personalization (23%), campaign orchestration (19%).
• Stable but selective: copywriting (29%), image generation (23%).
• Low adoption: website development at 6% and other technical builds remain niche.

The gap: while marketers plan to advertise on conversational AI interfaces (e.g., ChatGPT, Google AI Overviews), many can't fully deploy generative AI internally because of data quality, access, and governance issues.

Why the gap exists: data quality, signal loss, and messy workflows

• Data quality: 42% say poor data blocks broader AI use-think fragmented IDs, inconsistent taxonomies, and limited consented signals.
• Signal loss: privacy changes and platform policies reduce match rates and event fidelity.
• Workflow sprawl: point tools create duplicate work and slow feedback loops between media, creative, and analytics.
• Measurement friction: incomplete MMM/MTA coverage and weak incrementality testing slow decision-making.

What to do next: a practical 90-day plan

1) Fix the plumbing (data and privacy)

• Standardize event naming, UTM rules, and taxonomy across paid channels.
• Audit identity resolution: ensure consented IDs, clean audience deduplication, and clear join keys.
• Strengthen consent flows and retention policies; document data lineage and access rules.
• Pilot a clean room with one priority partner to enable privacy-safe reach/frequency and incrementality.

2) Orchestrate creative and media together

• Create a channel matrix for CTV, YouTube, TikTok, Reels, and display with spec + hook + CTA rules.
• Build 3-5 platform-native variants per core concept (first 3 seconds, captions, aspect ratios, CTAs).
• Connect creative metadata (hook, length, topic) to performance data so AI can recommend next iterations.

3) Make AI useful inside campaigns

• Use genAI for insight generation on query trends, audience clustering, and creative themes.
• Automate weekly optimizations: budget pacing, bid/goal checks, and creative fatigue alerts.
• Introduce guardrails: approved tone, claims libraries, compliance checks before ad launch.

4) Update measurement for less signal

• Run lightweight geo or PSA holdouts to validate platform lift claims.
• Blend MMM for budget allocation and MTA for short-cycle optimization; reconcile differences monthly.
• Track "view-through + engaged-view" actions on CTV/social to capture upper/mid-funnel impact.

Channel notes to guide budget decisions

CTV and streaming (63% increasing)

• Plan frequency caps across publishers; unify creative rotations by audience and episode context.
• Use QR, short vanity URLs, or sequential retargeting to map attention to action.

Social and short-form video (61% increasing)

• Prioritize distinct hooks per platform; avoid generic edits.
• Rotate creator-led and product-led cuts; keep versions under strict length ceilings per placement.

Influencer/creator (53% increasing)

• Negotiate paid usage rights upfront; plan whitelisting and spark ads as part of the SOW.
• Measure creator lift with matched-market tests or brand lift studies.

Retail media (40% increasing or steady)

• Sync promotions and inventory windows; use audience overlays tied to loyalty and category signals.
• Tie closed-loop sales to creative/message variants to fund continued spend.

Governance that keeps you moving

• Approve a short list of AI use cases (what's in/out), plus review steps for regulated claims.
• Create a shared glossary for taxonomy and naming so analytics doesn't break when teams swap tools.
• Document how data flows from consent to activation to reporting; review quarterly.

The big takeaway

Budgets are following attention-AI media and video-first channels are getting the lift. The constraint isn't imagination; it's data and workflow quality. Teams that fix data standards, orchestrate creative with media, and automate the boring parts will get the upside sooner-on both the buy side and inside their orgs.

Helpful resources

• IAB Tech Lab for identity, privacy, and measurement standards.
• AI certification for marketing specialists to level up practical genAI workflows.