8 min read
Content is worth creating when it helps a specific buyer make a decision they are facing right now, and when it gives them something a one-paragraph AI summary can't.
In practice, that means comparisons with real tradeoffs, first-hand experience, original data, and knowledge only your team has. Generic definitions and "what is X" explainers are the weakest bets, because search engines and AI assistants can answer them on the spot. Before you commission anything new, ask whether updating an existing page, or getting included in a source buyers already trust, would close the gap faster.
Why quick-answer content is the easiest to replace
Content that restates a simple fact is the kind search engines absorb most easily, and it is the least likely to earn a visit.
The strongest evidence comes from a causal study of real browsing data: 477,485 search result pages from 1,756 participants. It found that when Google showed direct answers and featured snippets, click-through to websites fell and people spent more time on the results page itself (Gleason, Hu, Robertson & Wilson, 2023). The study predates AI Overviews, so treat it as evidence for the mechanism rather than a measurement of AI answers. Its lesson still holds: when the page answers the question, fewer people need to click.
AI answers are shifting attention, but clicks haven't simply collapsed
The popular claim that "AI killed clicks" goes further than the controlled evidence does.
One lab study of Google's AI summaries found no significant difference in engagement or website traffic, although visual attention on the AI summaries increased (Rolon-Merette, Guo et al.). A 44-participant eye-tracking study reached a similar conclusion. Every participant looked at the AI Overview at least once, and it dominated attention whenever it appeared. Attention to organic results appeared lower, but that drop was not statistically significant (Pihlakas & Mägi, 2026). Both are small studies and cannot show what happens to traffic at scale.
The firmer conclusion is about attention. The AI answer is where buyers look first. Content that only feeds that answer a definition is easy to replace. Content that shapes what the answer says, or that gives people a reason to keep reading, still has a job.
The same eye-tracking study found that the type of task mattered more than whether an AI Overview was present. On transactional tasks, people scanned more widely and returned to results more often than on informational ones. The sample was small, so treat this as suggestive. It does fit the practical picture: people who are about to buy keep comparing sources, while people looking up a fact stop at the first answer.
Why publishing more makes you harder to notice
Buyers already do most of their research alone, and they already have more content than they can use.
Gartner found that business buyers spend only 17% of their buying time meeting with potential suppliers, all suppliers combined, and 27% on independent online research (Gartner, 2020, as cited in Salonen et al., 2024). DemandGen reported that 77% of B2B buyers consume three or more pieces of digital content before contacting a seller (DemandGen, 2020, as cited in Industrial Marketing Management). Your content is doing the work a salesperson used to do: framing the problem, comparing options, and reducing risk.
The same research describes buyers as overwhelmed with online information. In one survey, only about half of B2B marketers believed their own content stood out, and only 23% tailored it to specific individuals (Industrial Marketing Management). That is marketers rating themselves, so it is not an audit. Still, it describes the market you publish into. Another generic post does not raise your visibility. It adds to the pile.
What earns attention: relevance to the decision a buyer is making now
Timing and relevance drive engagement more than volume or funnel templates do.
In an experiment on B2B content, buyers who saw content they considered relevant to their current stage engaged more with the seller. The researchers also found no universally correct sequence for presenting content, because preferences by stage varied between individuals (Salonen et al., 2024). The study used a fictitious buying journey, so its results may not transfer cleanly to real ones. It still points to a better filter than keyword volume: is someone deciding something that this piece helps with?
Content that tends to pass that filter:
- Comparisons with honest tradeoffs, such as "X vs Y for a team of five," including where you are not the right choice.
- First-hand experience: what happened when you implemented, migrated, or tested something, with specifics.
- Original data from your product, customers, or support tickets that nobody else has.
- Risk-reducing detail, like implementation effort, pricing logic, integration limits, and what goes wrong.
None of this makes top-of-funnel content worthless. The point is that every piece needs a buyer, a moment, and a decision it helps with.
A four-question test before you create anything
This test is our editorial framework built from the evidence above. It is not a validated research instrument.
Scroll horizontally to see all columns →
| Question | If yes | If no |
|---|---|---|
| 1. Is there a real decision behind this topic? Can you name who is asking and what they're trying to choose? | Keep going. | Skip it, or fold it into a page that serves a decision. |
| 2. Can we add something a summary can't? Examples include data, experience, specific tradeoffs, or internal expertise. | Keep going. | Expect an AI answer or snippet to absorb it. Only publish if you can add depth. |
| 3. Does it connect to what we sell? Would a reader naturally move closer to evaluating you? | Keep going. | It may be useful, but it isn't a priority. |
| 4. Is a new page the fastest fix? | Create it. | Improve an existing page, or pursue inclusion in a source buyers and AI already trust. |
To use this in your next planning session, take the last ten items on your content calendar and run each through the four questions. Pay attention to how many fail question 2. Those are the pieces most likely to be replaced by an on-page answer, and usually the best candidates to cut or merge.
Sometimes the right content is not a new page
Question 4 matters because the gap you're trying to close often isn't a missing article.
If you already have a comparison page that's thin or outdated, rewriting it with current tradeoffs usually beats starting a competing page. If buyers and AI assistants rely on third-party roundups and reviews in your category, getting included there can matter more than anything on your own site. We cover when that is legitimate strategy and when it's just link building in Are Third-Party Listicles Good GEO or Just AI-Era Link Building?.
The hard part is knowing which response fits which gap. You would need to see what AI assistants currently say about your category, who they recommend instead of you, and which sources they draw on. That is the problem Tideflow AI is built for. It monitors AI answers and turns each opportunity into a gap plan: improve an existing page, create new content, or pursue inclusion in sources AI already trusts. Your team reviews and approves anything before it goes live.
Tideflow also keeps three signals separate that teams often blur together. The first is an AI crawler visiting a page. The second is a citation in an AI answer. The third is a referral visit that converts. A crawler visit shows discovery, not inclusion in an answer (Tideflow AI). No monitoring or publishing tool can guarantee a citation or recommendation (Tideflow AI vs Profound). The value of separating these signals is knowing which of your decision-focused pieces actually move buyers.
Frequently Asked Questions
Does content still pay off if AI summarizes it and nobody clicks?
Content can still pay off when it shapes what the summary says and when buyers who keep researching find depth there. Controlled studies of AI Overviews have not shown engagement disappearing, and most B2B buyers still read several pieces of content before contacting a seller. Content whose only value is a quick fact is the kind least likely to pay off.
Should we update existing pages before writing new ones?
Update first when an existing page already targets the same decision and only needs sharper tradeoffs, current details, or original evidence. A new page makes sense when no existing page serves that buyer's decision. Running the four-question test on your current pages is a quick way to find refresh candidates.
How do we find the decision questions buyers are actually asking?
Start with sources close to real decisions: sales call notes, support tickets, lost-deal reasons, and the prompts buyers type into AI assistants about your category. Our playbook on how to get mentioned by ChatGPT walks through testing what assistants say today, which shows where your content is missing.
Sources
- Gleason, Hu, Robertson & Wilson, Google the Gatekeeper: How Search Components Affect Clicks and Attention (June 2023)
- Rolon-Merette, Guo et al., Generative AI Summaries Do Not Increase User Engagement
- Pihlakas & Mägi, The Impact of AI Overview on Visual Attention Allocation and User Confidence Across Search Task Types (January 2026)
- Salonen, Mero, Munnukka, Zimmer & Karjaluoto, Digital content marketing on social media along the B2B customer journey: The effect of timely content delivery on customer engagement (February 2024)
- Industrial Marketing Management, Digital content marketing in business markets: Activities, consequences, and contingencies along the customer journey
- Tideflow AI, AI Visibility & Content Platform; Tideflow AI vs Profound: Which AI Visibility Platform Should You Choose?