Prompt
My boss keeps asking why traffic from search feels flatter even though rankings look okay. I suspect people are getting answers from AI tools instead of clicking through. How would you measure that, and do I need an AI visibility platform to prove it or can I piece it together with analytics and manual checks?
Latest observation
Yes—you can absolutely make a strong case without an AI visibility platform, but the best answer usually comes from combining analytics, search data, and manual AI-tool checks.
The core idea is to prove whether demand is still there while clicks are being intercepted or reduced by AI-generated answers, featured snippets, and zero-click search behavior.
What to measure
1) Search demand vs. click demand
Look for a gap between:
- Impressions staying steady or rising in Google Search Console
- Clicks and CTR falling
- Average rankings looking stable
That pattern suggests your pages are still being seen in search results, but fewer people are clicking.
Helpful cuts:
- Brand vs non-brand queries
- Informational vs transactional queries
- Queries with SERP features vs plain blue-link results
- Mobile vs desktop
- Countries where AI Overviews or similar features are more common
If impressions are flat but clicks fall, that’s often a signal of SERP answer compression.
2) Landing page traffic by intent
In analytics, segment organic landing pages by page type:
- Blog / informational content
- Comparison pages
- Product / category pages
- Local pages
- Support / FAQ content
If top-of-funnel content loses organic sessions while rankings remain similar, that’s a good sign that users may be getting answers directly from SERPs or AI tools.
3) CTR changes by query group
In Search Console, compare CTR over time for:
- Queries with long-tail informational intent
- Queries that are question-based
- Queries where SERPs now show AI Overviews, People Also Ask, featured snippets, or knowledge panels
A drop in CTR on queries that still rank well is one of the cleanest indicators.
4) Branded search and direct traffic
If people are using AI tools to learn about your category, two things may happen later:
- Branded search may rise if they remember your name from the AI response
- Or direct traffic / returning visits may rise if awareness increases
This is indirect, but it helps tell the story that AI may be influencing discovery, not just suppressing clicks.
5) SERP feature presence
Manually check whether your important queries trigger:
- AI Overviews
- Featured snippets
- People Also Ask
- Instant answers
- Video packs, shopping, maps, etc.
If the SERP is satisfying the query before the click, that often explains flatter traffic.
How to test whether AI tools are contributing
Manual checks
Pick your top 20–50 high-value queries and check them in:
- Google Search
- ChatGPT with web browsing / search features if relevant
- Perplexity
- Gemini
- Copilot
Look for:
- Whether your brand is mentioned
- Whether your content is cited
- Whether competitors are mentioned instead
- Whether the tool gives a complete answer without needing a click
This won’t give you a perfect quantitative measurement, but it’s very useful for pattern detection.
Use analytics to look for “assist” behavior
AI tools often act like top-of-funnel research rather than direct referral sources. So check for:
- Increases in branded search after AI-related content gets published
- Higher direct traffic or returning users
- Changes in assisted conversions from organic
- More “dark” or unattributed traffic patterns
You usually won’t get a clean “this came from ChatGPT” report unless the tool sends referral traffic, which is inconsistent.
Do you need an AI visibility platform?
Short answer: no, not to prove the trend
You can build a credible internal case with:
- Google Search Console
- Web analytics
- Rank tracking
- Manual SERP reviews
- AI tool spot checks
That’s enough to show:
- rankings are stable,
- click performance is weakening,
- SERPs are becoming more answer-heavy,
- AI tools are likely reducing the need to click.
When a platform helps
An AI visibility platform is useful if you want:
- Automated monitoring across many prompts
- Share-of-voice in AI answers
- Citation tracking
- Competitor comparisons
- A repeatable executive dashboard
It’s more about scale and convenience than proving the basic point.
If your boss wants a quick, defensible answer, you probably don’t need a platform first. If this becomes a standing KPI, then a platform can save time.
A simple framework to present internally
You can frame it like this:
Hypothesis: Search traffic is flatter because more searchers are getting answers directly from AI-enhanced SERPs and external AI tools.
Evidence:
- Rankings stable
- Impressions stable or rising
- CTR down
- Informational pages hit hardest
- SERPs show AI/answer features
- AI tools can answer key questions without clicks
Conclusion: Demand may still exist, but the click is being absorbed earlier in the journey.
A practical measurement plan
If you want to do this yourself, start with:
-
Export GSC for the last 6–12 months
- Queries, pages, clicks, impressions, CTR, position
-
Group queries by intent
- Informational, commercial, transactional, branded
-
Identify the biggest CTR decliners
- Especially where position didn’t worsen
-
Review the live SERPs
- Note which ones have AI Overviews/snippets/instant answers
-
Check top queries in AI tools
- Record whether your brand/content appears
-
Compare landing page trends in analytics
- Especially organic sessions to informational pages
-
Build a simple narrative
- “We’re not losing rankings; we’re losing click opportunity.”
One caution
Don’t over-attribute everything to AI. Flatter traffic can also come from:
- SERP layout changes
- Seasonality
- Better answers from competitors
- Snippet cannibalization
- Changes in search demand
- Tracking issues
The strongest case is usually multi-causal, with AI being one part of a broader zero-click trend.
If you want, I can also give you:
- a step-by-step dashboard template,
- a boss-ready slide narrative, or
- a list of metrics to pull from GSC/GA4.