Authors: Laura Richardson, Partner – Paid Search, & Anna Darova, Partner - SEO
In Part 1 of this mini-series, we discussed how roles of search practitioners are set to evolve and expand, as the search ecosystem becomes intertwined and agentic search booms.
Now, we’ll unpick what this means in practice, with six recommendations for search practitioners to start doing today.
1) Build AI fluency across teams
As discussed in part 1, practitioners must be upskilled to broaden knowledge and understanding of the modern search ecosystem. AI knowledge must now be a baseline skill, not a specialist add-on, and learned from entry level. It’s important that search practitioners understand the dynamics of the rapidly-developing agentic search world, how content serves within this, and how consumers engage with content at each stage of their journey.
- Paid search: we anticipate ad layouts, formats and buying models to change significantly in the next 1-2 years as advertising becomes more prominent in AI answers. Specialists must understand how AI algorithms decide how and where to serve ads, how they contextualise brands, and how ads integrate into conversational journeys. Prompts are the new keywords for paid search; and specialists must recognise the changing user journeys, how users phrase requests to AI agents, and how those AI agents find and structure recommendations back to the user.
- SEO: specialists are expanding beyond Google and Bing into multi-modal search environments – ChatGPT, Perplexity, Reddit, TikTok to name a few. Understanding how brands are being interpreted by AI systems, what trust and topical signals are factored in, and how to optimise for both human users and machine summaries is now imperative. Total Search is no longer just an SEO and Paid collaboration, but an integrated discipline where SEO contributes to strategy and planning, generative LLM optimisation, organic social strategies, and brand building.
2) Create a best practice for generative content
Search practitioners must develop a clearly defined approach to optimise content for AI environments, to support discoverability, engagement, and purchase goals. This involves preparing ads, feeds, and landing pages so they fit naturally into generative search results.
- Paid search: specialists must develop technical strategies to influence algorithms and surface ads in relevant moments. Ads and feeds must be machine-readable. Practical examples include enhancing structured data feeds, optimising creative for natural language, or adopting more rich media ad formats. Paid search specialists need a mindset change to evolve and update traditional best practices, shifting focus from keywords and queries, to audiences, topics, and context.
- SEO: “Content is king” has been true for years, but in the era of generative search the definition has shifted. Authoritative, credible, and trustworthy content remains fundamental, but what differentiates brands now is how well their topical signals and reputation are understood by AI systems. SEO is extending beyond standard on-page and off-page tactics into PR affiliation signals, social platforms, community forums, and contextual optimisation across the wider web. The goal is to create a holistic footprint that shapes how LLMs and AI Overviews perceive and present a brand. has recently confirmed that strong SEO foundations directly correlate with visibility in AI-driven summaries i.e. good SEO equals good LLM visibility.
3) Lean into AI innovation
Search practitioners must be ready and willing to test and learn with new AI innovations, experiment with new channels and formats, and establish the ‘new normal’ for best practices.
- Paid search: teams must prioritise and accelerate the pace of test and learn initiatives across AI-powered products, particularly Google’s Power Pack: PerformanceMax, AI Max, and DemandGen. These products reflect the shift towards the agentic world and offer more dynamic, cross-channel, and visual ad content to serve in the right moment for a user. These products will become the bedrock for Paid Search accounts, and we anticipate much more AI-powered product innovation in the immediate future. We know control and insight is often the trade-off for using AI products, so getting the balance right between ‘trust the machine’ and ‘retain control’ is key.
- SEO: just as paid search is accelerating test-and-learn with AI, SEO is broadening its scope to incorporate insights from prompt engineering, cross-channel data blending, and new measurement approaches for AI interactions. Roadmaps increasingly include organic shopping optimisation, Reddit community audits, and TikTok trend mapping to capture cultural intelligence and holistic growth. As user search journeys often flow from Google or LLM into community spaces for validation, SEO is evolving into a more creative, experimental discipline influencing the full discovery process, and optimising for new formats.
4) Redefine search KPIs
Success metrics need modernising in line with the new era of search, and we must understand what is driving real long-term value. Search KPIs typically focus on on-site behaviour goals like sign-ups or purchases, but marketers must recognise the new importance of off-site metrics, as we begin to see more research, engagement, and even purchasing, happening directly within AI environments. This period of transformation will create the new-normal for KPIs and reporting metrics, and advertisers must be prepared to see radical shifts in traditional metrics.
- Paid search: traditional success metrics like CPA, CTR and CPC won’t disappear, but they need to be reframed and expanded because user journeys and ad placements are changing. Visibility and impression share metrics will become more important with the rise of AI Mode, creating new opportunities for ads to be served, and brands to be discovered. The search landscape is also becoming more visual, so it’s possible we could see CPM models for ad placements, creating new metrics and KPIs to report against. We already see this in some search platforms in the Eastern world, like Naver and Baidu, where visual search is much more prominent. There’s still a lot to develop here, but we must start monitoring a broader range of metrics now.
- SEO: the introduction of LLMs has made SEO measurement more complex and far less transparent. Google has merged AI Mode and AI Overviews data within Search Console, without a separation from other search types, meaning query-level attribution from AI Searches remains unavailable. At 兔子先生, we’ve developed a proprietary tool, Deibu, to analyse brand sentiment and share of voice in LLMs, complemented by third-party solutions that provide directional visibility insights. To put this into practice, our SEO specialists built an LLM dashboard that tracks traffic and revenue attribution across platforms such as ChatGPT, Perplexity, and Gemini, enabling brands to monitor growth over time. While no dataset is flawless given the personalisation of results, these tools offer some insights into presence in AI ecosystems. Core SEO KPIs like traffic, conversions, and engagement still matter, but they’re no longer the whole story. In the new era of search, brand trust and recognition are becoming the decisive factors shaping user behaviour and conversions. Visibility across AI results, social platforms, and video ecosystems is just as important as traditional rankings and SEO specialists must understand that rankings alone no longer guarantee traffic.
5) Bring specialisms together with shared goals and outputs
As we explored in Part 1, silos need breaking down between SEO and paid search teams to form agile, multi-disciplinary talent, that can pivot quickly to lead the way in the AI space. The evolution of search is reshaping workforces from keyword tacticians into hybrid strategists, technologists, and storytellers. Here’s a few examples of shared outputs, but by no means an exhaustive list.
- Create a shared vision around the user, not the channel: focus on the user journey and experience, not channel metrics. Define and measure shared metrics, such as qualified traffic, new customers, or total revenue, and focus strategies on how both channels can work together to help the right customers find and choose your brand.
- Report on Total Search performance: move from channel metrics to business outcomes by building integrated reports to monitor metrics like Total Search visibility, Total Search conversions, and blended ROI.
- Build structural alignment and communication: this requires operational changes to connect SEO and paid search specialists. How about weekly check-ins, chat groups (e.g. on Teams), or cross-team buddying?
- Create a shared knowledge repository: a single library of information, perhaps containing brand guidelines, goals / objectives, or best practices.
6) Monitor brand safety
It’s important that teams are monitoring and interrogating AI outputs, to help preserve brand-safety. The ability to track and report information being surfaced in AI environments is still incredibly limited, but tools and manual observations can help audit and identify if brands are being misrepresented. Things to watch out for could be: incorrect facts, bias, associations with unsafe topics, or harmful statements. To reduce the likelihood of this happening, brands must ensure their own data / information is always up-to-date and accurate, ensure external sources used in AI responses are up-to-date, and where possible encourage positive testimonials across review sites and social media. Upstream accuracy = downstream AI safety.
To conclude, we face a period of intense change as AI becomes embedded throughout Search, both for the advertiser and the end-user. There’s a lot of speculation around how search products and user behaviours will develop, and still many unknowns. One thing is for sure, those who lean in now will have the competitive advantage and will flourish in this next chapter of Search.
Want to know more about how your brand is surfaced and represented in AI environments, along with optimisation recommendations? We have built a robust approach to LLM optimisation, spanning four key pillars:
- LLM-visible: ensure bots can find, access and cite your content
- LLM-readable: help LLMs understand and interpret your content accurately
- LLM-credible: ensure LLMs perceive your content as trustworthy and authoritative
- LLM-validated: audit how LLMs represent your brand: track sources, monitor appearances, test formats, and optimise to influence results
For more information, reach out to us at and we’d love to discuss our LLM audit and recommendations service with you.
