AdExchanger Just Called Out the Ad Intelligence Industry. The Tools Were Built for a Pre-AI World. The Decisions Are Not Faster.
AdExchanger argues the competitive intelligence category has not caught up with AI. The data is faster. The decisions are not better. Without a unified methodology across channels and markets, AI cannot compress the signal-to-decision path.
Access to data is no longer the problem. The bottleneck is the ability to translate that data into informed action.
AdExchanger published a sharp piece in its Content Studio this week. The ad intelligence category, the tools that track competitor spend, creative and performance across markets, has not kept up with the AI era. The data is faster. The decisions are not better.
The argument, from BIScience co-founder Assaf Toval, is that the industry now has near-real-time access to competitor ad activity across paid social, search, programmatic and CTV. What it does not have is a unified framework for turning that into a decision. Most ad intelligence platforms still report by channel, by region and by snapshot. Decisions get made on a campaign-by-campaign basis when they should be made on a portfolio basis.
The piece argues that to be useful in 2026, ad intelligence needs a consistent methodology across media and markets. Without it, AI cannot compress the signal-to-decision path. With it, AI becomes a real multiplier on analyst capacity.
Why it matters
Most Australian marketing teams under-resource competitive intelligence because the tooling is fragmented. Meta Ad Library, Google Ads Transparency Center, SEMRush, SimilarWeb and Pathmatics each show a slice. Stitching them together is a job. By the time the stitching is done, the campaign has shipped, the competitor has changed creative and the report is stale.
The compounding cost is that AI tools now expect a clean competitor data feed. If the input is fragmented and contradictory, the AI summary is worse than the human summary, not better. The dashboard says one thing in the Meta library, another in the Google one and a third in the SimilarWeb pull. The AI averages them and produces noise.
Australia's share of global digital ad spend. The local competitive set is small, which makes tracking competitor moves harder, not easier.
What to do about it
This is a 90-day workstream, not a tooling decision.
Pick one source of truth per channel. Meta Ad Library for Meta creative, Google Ads Transparency Center for Search and Display, SimilarWeb for site traffic. Stop pulling from three each.
Set a fortnightly competitive review cadence. Not monthly. Competitor creative cycles are now under two weeks.
Tag every competitor change to a hypothesis. New creative, new offer, new channel. If you cannot explain why a competitor made the change, you cannot react to it.
Feed the structured competitor data into one AI tool. Briefing competitor activity to Claude or ChatGPT works only if the data is consistently formatted. Standardise the input.
Action one decision per fortnight off the back of competitive intel. If the review does not produce a decision, the review is theatre.
The ad intelligence category is going to consolidate over the next 18 months. The teams that built their own internal layer first will be the ones the consolidated tools end up serving. Everyone else will be retrofitting their workflow to whatever the acquirer decides to keep.