Four priorities that separate substance from hype

The workshop's conclusions came down to four priorities shared by teams that actually put AI to work rather than just talk about it:

  1. A named owner for AI inside the team: Not “someone will look at it someday”, but a person with the mandate to decide what gets adopted, what gets dropped and what gets built in-house. Without that role, AI experiments drift.
  2. Cost and benefit for every use: Not every AI application is worth it. The pattern is the same everywhere: one or two high-impact uses pay for a long list of low-impact tests. It takes discipline to tell them apart without getting attached to the new tool.
  3. Solutions that work and deliver now: The priority is work that goes live this quarter. A demo that impresses internally but changes nothing in a client account is not a result.
  4. Team training and a testing culture: AI is not a switch you flip. It is continuous work. Teams that treat it as a culture change, not a tool purchase, build an edge that grows over time.

Where AI is actually moving the numbers

In the discussions, the uses that genuinely improved performance were few and specific:

  • Creative speed: producing, testing and retiring ad creatives at a pace manual work cannot match.
  • Account structure analysis: spotting patterns in large accounts that an analyst would miss or need far too long to find.
  • Content for conversations and intent: from content ready for AI Overviews to copy for ChatGPT Ads.
  • Reporting automation: freeing senior account managers from weekly reports so they spend time on the work those reports are supposed to measure.

Just as important was what did not make the list: AI running campaigns on its own without supervision. Every team in the room had tried it, and all had seen uneven results at best.

What we are taking back to clients

For Omnicliq, Bratislava confirmed a direction we were already moving in. These are our priorities for 2026, the same ones we bring to clients through AI consulting:

  • Clear AI owners in every team, not centrally but per discipline (PPC, SEO, creative, analytics).
  • Every proposed AI application scored on a simple cost and benefit grid before any development time is committed.
  • Small, measurable improvements, fast, such as automated dashboards, creatives and keyword discovery, instead of one big, vague project.
  • Investment in team training as ongoing work, because knowledge about specific AI tools ages so fast that learning has to be continuous.

Two days of real value, in a collaborative setting that pushes the whole industry towards clearer strategy and more discipline.