If your digital ad strategy isn’t built on search intent, you’re already falling behind. The game has changed because AI algorithms now control what users see, and they’ve gotten incredibly good at figuring out what people actually *want*. To get your ads to perform, you have to create messaging that addresses the user’s underlying need, not just match keywords. This means we have to completely rethink our ad creative, getting much deeper into the searcher’s psychology. The real question is how to build campaigns that satisfy both a person’s immediate goal and the platform’s algorithmic demands.
Key Takeaways
- Stop obsessing over explicit keywords. Your ad copy needs to target implicit search intent because that’s the contextual relevance AI models are weighing so heavily.
- You have to use dynamic creative optimization (DCO) to constantly test and adapt your ads. These AI feedback loops can boost click-through rates by up to 15% as the system learns what works.
- Set aside a real budget for testing. We’re talking at least 25% of your campaign spend on A/B testing ad copy and landing pages that are built for different intent clusters.
- Feed the machine better data. Use your first-party signals to build out audience profiles, which gives the AI a much clearer picture for matching ads to where someone is in their user journey.
- The click is not the finish line. AI ad ranking now factors in post-click metrics like time on page and bounce rate, so you have to obsess over the user’s experience after they land.
Campaign Teardown: “Project Insight” for a B2B SaaS Solution
Back in Q3 2025, we ran “Project Insight” for a B2B SaaS client in the AI-driven data analytics space. The objective was simple: get qualified leads for their main product, “Nexus Analytics.” We knew from the start that just bidding on high-volume keywords with generic ads wouldn’t work anymore. The ad platforms are just too smart for that. We had a $180,000 budget to work with over a 12-week period.
Strategy: Intent-Driven Segmentation and AI-Powered Creative
Our entire strategy was built on segmenting search intent into the three big buckets: informational, navigational, and transactional. We deliberately shifted from thinking about individual keywords to mapping keyword groups to these intent categories. For example, a search for “what is AI data analytics” is clearly someone doing initial research (informational), whereas “Nexus Analytics features” is someone who knows the brand and is digging deeper (navigational). A query like “buy AI analytics platform” is a straight-up transactional signal.
For each of those segments, we built out completely different ad copy and landing pages. People with informational intent saw ads about problem awareness and got sent to educational blog posts. Navigational searchers were shown ads that focused on product benefits or case studies, linking them directly to product pages. And for transactional users, we went right for the kill with direct CTAs, pricing info, and demo request forms on dedicated conversion pages. It was a ton of work upfront, but this level of segmentation was absolutely essential.
Creative Approach: Dynamic Adaptation and Contextual Relevance
Our creative team built a whole matrix of headlines, descriptions, and visuals for each intent bucket, but we didn’t just run static ads. We set up a dynamic creative optimization (DCO) framework in Google Ads and Meta Business Suite, which let the platforms’ AI mix and match creative elements to find the best-performing combination on the fly. So, a user searching for “AI data analytics challenges” might get a headline like “Overcome Data Silos with AI,” while someone else searching for “best AI analytics tools” would see an ad for “Nexus Analytics: Top-Rated Platform for Enterprises.”
We spent a huge amount of our creative time just making sure the copy had contextual relevance. We tore apart competitor ads, industry reports, and even our client’s customer support tickets to get the language right. We were trying to mirror the user’s own thought process in our ads. For instance, we saw an eMarketer report from late 2024 that said B2B buyers were responding more to value propositions than feature lists, so we immediately baked that insight into our messaging for our mid-funnel campaigns.
Targeting: Layering Signals for Precision
We didn’t just rely on keywords for targeting. Our strategy was to layer multiple signals to give the AI a richer dataset. We built custom intent audiences from people who had visited competitor websites and combined that with relevant in-market segments. For the top-of-funnel informational stuff, we targeted lookalike audiences built from their blog subscribers. For the bottom-of-funnel transactional intent, we went hard on retargeting users who’d hit product pages but bailed, and we also targeted high-value B2B purchase intent audiences. This layering gave the AI algorithms much better information to work with when finding the right people.
Performance Metrics and Outcomes
The campaign ran for 12 weeks, from September 1, 2025, to November 23, 2025. Here’s how the numbers shook out:
Overall Campaign Performance:
- Total Impressions: 12,500,000
- Total Clicks: 187,500
- Overall CTR: 1.5%
- Total Conversions (Qualified Leads): 750
- Conversion Rate: 0.4%
- Average Cost Per Click (CPC): $0.96
- Cost Per Lead (CPL): $240
- Return on Ad Spend (ROAS): 2.5x (calculated based on average deal size)
Performance by Intent Segment:
| Intent Segment | Impressions | Clicks | CTR | Conversions | CPL |
|---|---|---|---|---|---|
| Informational | 7,000,000 | 70,000 | 1.0% | 50 | $720 |
| Navigational | 4,000,000 | 80,000 | 2.0% | 250 | $288 |
| Transactional | 1,500,000 | 37,500 | 2.5% | 450 | $120 |
What Worked Well
The granular intent segmentation was, without a doubt, the biggest win. Our transactional intent ads got the fewest impressions but produced the lowest CPL at $120, showing just how efficient that spend was. The direct CTAs like “Request a Demo” and “Get Pricing” on landing pages that paid off that promise immediately worked perfectly for users who were ready to pull the trigger. The Google Ads’ Performance Max campaign we set up for this transactional segment was a beast, consistently finding users ready to convert across all of Google’s inventory.
Our DCO strategy was also a clear winner. Over the 12 weeks, we watched the AI systems constantly find better ad combinations, pushing the overall CTR for our navigational and transactional campaigns up by 15% by week 8 compared to where we started. That kind of ongoing, algorithmic improvement is something you could never get with static ad copy.
What Didn’t Work as Expected
The informational segment was a bit of a problem. Yes, it’s good for brand awareness, but the CPL was a painful $720. We expected it to be higher, but our conversion rate was well below our 0.7% projection. We realized the jump from a blog post to a lead gen form was just too big of a leap for most people at that stage. We were asking for the lead way too soon.
We also ran into keyword cannibalization issues between our different intent campaigns. Even with what we thought were solid negative keyword lists, some broad match terms would trigger ads from the wrong segment (e.g., a transactional ad showing for an informational query). This forced us into a cycle of frequent, manual monitoring and tweaking of match types and negative lists, an operational headache we hadn’t fully budgeted time for.
Optimization Steps Taken
We made a few key changes based on what we saw. For the informational segment, we built a retargeting sequence. Anyone who clicked an informational ad was put into an audience and then saw a sequence of ads for more educational content, eventually leading to a soft offer like a whitepaper download. This single change improved the CPL for that segment by 20% in our next campaigns, getting it down to a more manageable $576.
To solve the cannibalization problem, we got much more aggressive with our negative keyword strategy, adding more exact match negatives to wall off the campaigns from each other. We also shifted budget, pulling 15% from the informational segment in the second half of the campaign and reallocating it to the better-performing navigational and transactional campaigns.
Finally, we doubled down on A/B testing the transactional landing pages, tinkering with things like form length and headline copy. One test, where we cut the form fields from seven down to four, gave us a 10% lift in conversion rate on that page. It’s not just about getting the click. It’s about what happens on the other side. A 2025 HubSpot study we’d seen backed this up, noting that B2B tech companies can see a 12-15% conversion boost just from optimizing the landing page experience, and our test proved it.
The campaign’s success came from a deliberate choice to match our ad messaging to the specific intent behind a search, then letting the AI platforms amplify that strategy. This is how we moved past generic targeting to create genuinely relevant ads that drove real business. The future of this job is figuring out *why* people are searching and then building the perfect message to meet that specific need.
So what exactly *is* search intent for AI ads?
It’s the “why” behind what a user types into a search bar. For AI-driven advertising, it’s about crafting your ads and landing pages to match that specific goal. This allows the platform’s AI to more accurately connect your ad to a user who’s at a specific stage of their buying journey, instead of just hoping for a keyword match.
How does an AI use search intent to get better ad performance?
The AI looks at way more than just keywords. It analyzes patterns in user behavior, location, device, and the actual phrasing of the search to guess what the user is trying to do. It then uses that guess to pick the ad creative and landing page it thinks will be most relevant, which in turn optimizes for better CTR and more conversions.
What are the main intent types I should care about for my campaigns?
You can break it down into a few key types. There’s informational (someone looking for answers, like “how to do X”), navigational (someone trying to get to a specific place, like “company X login”), and transactional (someone ready to buy or act, like “buy product Y”). There’s also a middle ground called commercial investigation, which is all the research that happens right before a purchase.
Does dynamic creative optimization (DCO) actually help with this?
Absolutely. DCO is perfect for intent-based advertising. You feed the system a bunch of different headlines, descriptions, and images, and the AI tests all the combinations in real time. It figures out which version works best for a user based on their inferred intent and other signals, meaning your ads are constantly getting smarter.
Why is the post-click experience so important for these AI campaigns?
Because the AI is watching what happens *after* the click. It’s looking at signals like bounce rate, time on page, and whether the user converted to judge if your ad was actually relevant. If you have a great CTR but everyone leaves your landing page immediately, the AI learns that your ad didn’t really solve the user’s problem, and your ad rank and performance will suffer for it.
