Search Marketing Takeaways for Leaders: September 2026
Can you still tell what’s driving growth?
For years, search measurement has felt relatively straightforward. People searched. Brands appeared. Customers clicked. Marketing teams measured rankings, traffic, conversions and paid-search performance. That journey was never as clean as the reporting made it look, but it was understandable.
Now somebody might discover a product through TikTok, validate it on Reddit, ask ChatGPT to compare alternatives, encounter an AI Overview, watch a YouTube video, search the brand name and eventually convert through a paid Google ad. Which channel drove the sale?
The honest answer is increasingly: several of them.
That makes updates in September particularly interesting. Google has opened up more data around AI and visual search. Paid platforms are automating more of the execution while simultaneously giving marketers better ways to understand what happened. Microsoft is bringing advertising closer to CRM revenue. Reddit, TikTok and YouTube continue to blur the distinction between search, social, media and commerce. The leadership challenge is moving with them.
Can we still tell where demand was created, what influenced the decision and which activity genuinely drove commercial value?
Here are ten developments worth paying attention to.
1. Search measurement is finally starting to catch up with how people search
One of the most useful Google developments this month is relatively simple: businesses can finally see more of what is happening inside newer search experiences. From 31 August, Google rolled out dedicated Search Console reporting for generative AI features globally. The report separates impressions generated through AI Overviews and AI Mode and allows businesses to analyse performance by page, country, device and date.
Google has also rolled out its Search generative AI control globally, allowing website owners to determine whether their content and links are eligible to appear within AI Overviews, AI Mode and other supported generative experiences. Then, on 24 September, Google added reporting for multimodal search.
That includes searches initiated through Google Lens, Circle to Search, image uploads and Chrome’s “Search this image” functionality. For businesses receiving this activity, it can now be separated within Search Console reporting.
That matters particularly for sectors where people naturally search visually: travel, retail, home and interiors, property, fashion, automotive and experiences. We have spent several years talking about search becoming more visual, conversational and AI-led. The reporting is now beginning to reflect that reality.
Leadership takeaway:
Establish a baseline now. Understand how much visibility you currently receive through conventional search, generative AI and visual search, and how that differs by page and audience.
Those numbers may still be relatively small for some businesses. Having the baseline means you can spot when they become commercially significant.
2. Rankings can look healthy while search delivers less traffic
A useful piece of research from Ahrefs this month illustrates why traditional SEO reporting needs more context. When AI Overviews launched in France in July, Ahrefs was already tracking 963 French domains through Google Search Console.
Across the full group, click-through rate fell by 3.4% in the nine days after launch. For the websites with the greatest AI Overview exposure, median CTR fell by 23.1%, with 82.2% of highly exposed domains experiencing a decline. This is a short-term French dataset rather than something we should automatically apply to the UK, but the behavioural signal is important.
A business can maintain strong organic visibility while receiving fewer clicks from that visibility. At the same time, Google has been actively changing the underlying organic environment. Its August spam update ran from 18–21 August, followed by another global spam update beginning on 24 September which was still rolling out at the end of the month.
That makes diagnosis harder. A decline in traffic might come from rankings. It might come from AI answers satisfying more of the search journey. It could be a change in SERP layout, seasonality, competitors, brand demand or a Google update.
Leadership takeaway:
Stop treating ranking movement as the explanation for every organic performance change.
Review visibility, click-through rate, traffic and commercial outcomes separately. Annotate major platform updates and look at the complete pattern before making significant changes. The search result itself is increasingly part of the customer experience.
3. Google is automating more paid search, and adding more ways to inspect the decisions
Google’s transition towards increasingly automated Search campaigns continues. From September, campaigns already using campaign-level broad match or text customisation are being upgraded into AI Max. AI Max brings together expanded query matching, automated creative generation, brand controls and landing-page selection within Search campaigns.
For marketers accustomed to controlling individual keywords, ads and landing pages, that represents another meaningful transfer of decision-making to the platform. Interestingly, Google is also beginning to expose more of what that automation does.
On 23 September it announced new AI Max reporting designed to show the complete journey from search term → creative assets shown → landing page selected.
Its AI Brief beta also allows advertisers to provide more detailed natural-language context about the business, audience and messaging they want AI Max to understand.
The direction seems to make sense. If the platform is going to make more decisions, advertisers need enough visibility to judge whether those decisions are commercially sensible.
Leadership takeaway:
Automation changes the role of paid-search management. More attention should move towards the inputs automation depends upon: customer understanding, proposition, conversion data, commercial objectives, landing-page quality and measurement.
The question becomes less about how tightly every lever is manually controlled and more about whether the system has been given the right information and whether somebody is scrutinising the output.
4. Amazon bringing ChatGPT into its advertising ecosystem is a bigger signal than another new ad placement
On 10 September, Amazon Ads announced a partnership with OpenAI. Selected US advertisers can now extend Amazon Ads campaigns into ChatGPT as part of a pilot programme.
The immediate UK impact is limited, but strategic significance is much more interesting. Traditionally, the place where advertising was purchased and the place where the customer searched were closely connected.
Google search activity was monetised through Google Ads. Amazon product searches through Amazon Ads. Social discovery through the advertising products of those individual platforms. This partnership begins separating those layers.
Amazon brings significant commerce data, advertiser relationships and campaign infrastructure. ChatGPT provides a completely different conversational environment where people increasingly research and compare products.
The advertising platform, customer data environment and discovery interface no longer necessarily need to belong to the same company.
Leadership takeaway:
Think beyond individual media platforms when planning future search investment. The emerging ecosystem may involve: customer intent in one environment, audience or commerce data from another, and the eventual conversion somewhere else entirely.
That will make attribution harder, but it also creates new ways to reach customers at high-intent moments.
5. Microsoft is bringing search advertising much closer to pipeline and revenue
Microsoft’s September updates deserve more attention than they often receive, particularly for B2B businesses. Microsoft Advertising can now connect directly with HubSpot.
The integration allows businesses to connect advertising activity with contacts and deals, automate lead follow-up and see which campaigns are contributing to pipeline and revenue, rather than stopping measurement at clicks or lead submissions.
Microsoft has also made optimisation experiments generally available across Search, Shopping, Audience and Performance Max campaigns, making controlled testing easier. Behind the scenes, Microsoft is taking paid-search data closer to AI working environments too, allowing advertising information to be accessed through tools such as Copilot, ChatGPT and Claude via its MCP infrastructure.
We’ll explore that more deeply in this month’s AI in Marketing update. For search leadership, the CRM connection is the more important development. A £50 lead that becomes a £50,000 customer is considerably more valuable than a £30 lead that never progresses. Search reporting should increasingly reflect that reality.
Leadership takeaway:
Wherever possible, close the measurement loop beyond the marketing platform. Connect search activity to: lead quality, opportunities, pipeline, revenue and ultimately margin or customer value.
The more advertising becomes automated, the more important the definition of a valuable outcome becomes.
6. Reddit’s role in discovery is evolving in several directions at once
Reddit has become an increasingly important part of the search ecosystem. Ahrefs analysed 145 million US Google search results and found Google’s “Discussions and forums” feature appearing on 11.7% of SERPs.
Reddit appeared in 87.8% of those US results and 83.9% of the equivalent feature globally. Around 41% of appearances also related to searches with commercial intent. Reddit itself is also becoming more explicitly search- and commerce-led.
Its AI-powered Answers product has been incorporated into Reddit’s main search experience, while September saw its shopping test expand into Home & Garden and Fashion & Apparel. Users can increasingly move from conversations and recommendations towards product information and purchase links without leaving that research environment.
There has been some debate over whether Reddit is becoming less important as an LLM source.
Promptwatch reported that Reddit’s share of ChatGPT citations fell sharply during part of August, from 3.8% to around 0.5%. But the company described the finding as provisional. A separate Ahrefs September dataset still ranked Reddit as the most-cited domain in ChatGPT, with a 16.8% share amongst the sources it tracked.
The sensible conclusion is that AI citation behaviour remains volatile and highly dependent on platform, query and methodology.
Reddit’s broader importance is much clearer - it influences Google results, its own growing search environment, product research and, in some circumstances, AI answers.
Google’s expanding Preferred Sources and Search Profiles features point towards the same wider theme: users and platforms are placing more emphasis on trusted sources, rather than treating every result as interchangeable.
Leadership takeaway:
Community and reputation belong within search strategy. Understand where customers discuss your category, which sources they trust and what they say about you when your marketing team is not controlling the message. Those conversations increasingly surface elsewhere.
7. AI referral traffic is growing quickly, but averages can be very misleading
Similarweb’s latest industry analysis gives us a useful sense of scale. AI platforms generated an average of 770.7 million referral visits per month worldwide between June 2025 and May 2026, up 117.4% year on year.
But the differences between sectors are significant. Marketplaces generated the greatest absolute volume, while Beauty and Fashion were amongst the fastest-growing categories from smaller starting points. Travel produced considerably more cited AI responses than the overall average because those journeys often require a hand-off to another website.
This matters because industry headlines can create poor investment decisions. “AI traffic has grown 117%” sounds substantial.
For one business that might represent thousands of commercially useful sessions. For another it might mean moving from ten visits to twenty. Referral traffic also tells only part of the story. An AI assistant may influence a decision without generating a click at all.
Leadership takeaway:
Benchmark AI search against your category, audience and customer journey.
Track referral volume and conversion quality, but also look for wider indicators such as branded search, direct traffic, customer research and AI visibility around commercially important topics. Percentage growth without context rarely tells you where the next pound should be invested.
8. TikTok’s search proposition now looks increasingly like search advertising
TikTok’s evolution from social discovery platform towards genuine search environment continued in September. Its new TopSearch product allows advertisers to bid for the first sponsored position against selected keywords within TikTok search results.
TikTok has also introduced familiar search-media measures including impression share, while advertisers can target brand, category, competitor, seasonal and campaign-related terms. Availability varies by account and market.
The mechanics are becoming surprisingly recognisable: Keyword, Search intent, Auction, Top sponsored result, Impression share.
That does not make TikTok another Google. The reason somebody searches there, the content they expect and the influence of creators and communities remain different.
Leadership takeaway:
Search strategy should increasingly be based on search behaviour, rather than a fixed list of traditional search engines. If an important audience actively uses TikTok to research your category, it belongs somewhere in the discussion around discoverability, content and potentially paid search investment.
Channel ownership inside the marketing team should not get in the way of following the customer.
9. The advertisement itself is becoming somewhere customers can search
Another September development illustrates how quickly the boundaries between advertising, search and customer experience are blurring. Google has begun testing Business Agent for YouTube Ads.
Viewers can ask questions about the product or brand alongside eligible video advertisements and receive conversational answers without leaving the viewing environment. Think about what that changes. Traditionally an advert attracts interest and sends somebody elsewhere to research. Increasingly the advertising environment itself can answer the next question. A customer could see a product, ask about specifications, compare suitability and move closer to a decision in the same interaction. That raises the importance of the information supporting the advertising.
Product feeds, accurate service information, FAQs, pricing, availability and brand knowledge become inputs into a conversational sales experience.
Leadership takeaway:
Start thinking about advertising as an interactive information environment, not simply a creative placement. Ask what someone might want to know immediately after seeing your proposition and whether your underlying data and content can answer those questions clearly.
This has obvious implications for ecommerce, but the principle extends into services, travel, property and B2B.
10. AI assistants are already influencing buying decisions, even where Google remains part of the journey
A September survey from Semrush and Exploding Topics provides some useful behavioural context. The research covered 2,338 US consumers. Among people already using AI, 65% said they had replaced at least some product-related Google searches with AI conversations.
More strikingly, 57.5% of AI users said information from a chatbot had led them to decide against a purchase. Yet Google had not simply disappeared. Search engines remained the most commonly used research source amongst weekly AI users, ahead of reviews, AI assistants, friends and family, social media and retail websites.
That combination feels important. Customer journeys are expanding rather than neatly migrating from one platform to another.
Someone may ask ChatGPT which products to consider, check Reddit for real-world experiences, watch reviews on YouTube, search the chosen brand on Google and finally buy through a Shopping ad. Each interaction adds or removes confidence.
Leadership takeaway:
Measure search as part of the decision journey, rather than expecting one channel to receive all the credit.
Ask:
Where was the brand discovered?
Where was it evaluated?
What created trust?
What removed objections?
And which interaction finally converted existing intent into revenue?
Those questions matter more commercially than arguing over which platform technically “owned” the sale.
Leadership Perspective: visibility, influence and value are becoming different things
September’s developments point towards a fairly clear challenge.
Search is becoming easier to automate while becoming harder to attribute.
Google can decide which queries to target, which creative to serve and which landing page to use.
AI assistants can recommend or discourage a purchase without sending a website visit.
Reddit discussions can influence Google, AI platforms and customers simultaneously.
TikTok can behave like a search engine.
YouTube advertising can answer questions.
An Amazon advertiser can appear inside ChatGPT.
None of that makes traditional search measurement redundant. It does mean leaders need to distinguish three things more carefully:
Visibility
Are customers encountering us when they research?
Influence
Are those interactions helping or harming the eventual decision?
Commercial value
Is that visibility ultimately contributing to profitable customers?
The three increasingly happen across different platforms. Strong marketing leadership therefore needs a view that joins them back together.
Leader’s Checklist
A few useful questions to ask this month:
Are we separating visibility, traffic and commercial performance in search reporting?
Have we established baselines for AI and visual search within Search Console?
Does paid-search conversion data reflect lead quality and revenue, rather than simply volume?
Do we understand how Reddit, TikTok, YouTube and AI assistants influence research in our category?
Can we explain where demand is being created, even when the final conversion happens somewhere else?
Final Thought
Search remains one of the most commercially important areas of digital marketing. The difficulty is that “search” describes less and less of a single channel.
People search with words, images, video, communities and conversations. Platforms increasingly answer questions before sending a click.
Advertising can appear inside research experiences. AI can influence the decision without appearing anywhere in conventional attribution.
That makes measurement harder, but it also forces a useful discipline. The objective was never really to generate the maximum number of search clicks.
It was to make sure the business was visible when customers had a need, credible while they considered their options and easy to choose when they were ready to act.
The platforms surrounding that journey will continue to change.
For leaders, the more durable question is: Can we still tell what is genuinely driving growth?
Useful Links & Further Reading
