Search everywhere doesn’t mean spend everywhere: how to budget for organic in 2027

Every CMO I speak to is being asked the same thing for 2027: do more in AI search and social search, without much more money. So the real question isn't how much extra to find. It's where to move what you've already got.

On November 5th, we ran AMA Everywhere, live on LinkedIn, YouTube and TikTok. Our Creator Partnership Strategist Aliya and I took on around 50 questions on organic planning and budgets for 2027. Some were submitted by marketers, others came from questions we'd received at events like BrightonSEO and Semrush Spotlight recently.

This post pulls the best of those answers into the plan I'd follow if I were sitting in the CMO's chair.

It covers three things:

1. How much budget should move from Google to AI search and social search (less than you think).
2. Which channel gets the first pound, and what to protect if budgets get cut.
3. How to measure it all when clicks keep falling and attribution keeps getting messier.

A quick note on language: I use "AI search", "LLMs" and "GEO" fairly interchangeably here. And "search everywhere" means everywhere your audience searches: Google, TikTok, YouTube, Instagram, Pinterest, Reddit and AI assistants.

TL;DR

If you only read one section, read this table. It's how I'd prioritise a 2027 organic budget, in order.

‍

‍

And the rules that sit around it:

• Search everywhere doesn't mean be everywhere. Show up where your audience already spends time.
• Aim to be recommended, not cited. Being named in an AI answer builds preference; a source link rarely does.
• Set an annual envelope, review it quarterly. Monthly is too often for anything to prove itself, but if you wait for an annual review you may miss the opportunity of being the first mover on AI/LLMs or capitalising on a social-search conversation opportunity.
• Make brand search tied to revenue your headline KPI. Falling traffic is fine; more qualified traffic should be the focus. If conversion rate, AOV, LTV and organic revenue are rising, you're in a good place.

The first pound of 2027 budget goes to digital PR

If I had one new pound to spend next year, it would go to digital PR. Not because it's my favourite channel, but because it's the only one that has the potential to move every search surface we care about.

• Traditional SEO still benefits from quality links and brand mentions.
• Social search gets amplified when coverage, creators and PR campaigns line up around the same conversation.
• AI search leans heavily on what publishers, listicles and product round-ups say about you.

What's changing is the purpose of digital PR. It's shifting from earning links for rankings towards earning recommendations and citations in AI answers. But it's still digital PR that gets you there, which is why it's the one budget line I'd absolutely lock in by January, with a clear view of what that budget is for and how it will be spent.

Is it still about links, or only brand mentions now?

"Only" is the problem with that question. Saying digital PR is brand mentions only implies links have no value, and that's wrong. Build quality links wherever you can; quality over quantity, as ever.

What's changed is the gap between the two. A few years ago links were miles ahead and brand mentions barely registered. Now they're much closer, because mentions are just as effective as links for showing up in AI responses.

We used to look down on brand mentions. In 2027 we should embrace them, for SEO, for direct traffic and awareness, and for AI recommendations.

How much should move from Google to AI search? 10–15%, as a test

Moving your whole budget to AI search right now would be shiny object syndrome. I wouldn't drink the Kool-Aid just yet. But ignoring AI visibility entirely would be short-sighted too.

Here's what I'd do: ring-fence 10–15% of the organic budget as an AI search test-and-learn pot. You accept a lower return on that money and don't hold it to the same standard as the rest. It lets you get into the conversations happening on AI platforms and get ahead of the curve, without walking away from traditional SEO, which is what's working today. Most people still use Google, so moving big chunks away from it makes no rational sense yet.

What return should you expect from AI search spend?

If you go in expecting nothing, you're in the right frame of mind. Brands making real revenue from AI search are few and far between, and they tend to sit in very specific niches.

Granola, the AI note-taking tool, is a good example. I found it through an LLM, and it has grown massively from AI recommendations. But its users are already on AI platforms looking for AI tools. For most brands, the shift from people searching in LLMs to people spending because of LLMs hasn't happened yet.

I believe it will. But expecting AI search to match Google organic numbers in 2027 would be foolish. Treat it as a test and learn, get into the conversation early, and you'll understand how your audience uses AI interfaces before it starts making you real money, likely from 2028.

The signal to scale it up

Use your own data. Set up an AI/LLM channel grouping in your analytics and watch it. The moment AI referral traffic starts turning into revenue, even in small amounts, that's your cue to grow the test pot.

Social search: let your audience set the allocation

Social search is a different conversation from AI search, because the audience is already there. In plenty of B2C categories, TikTok is a major destination for top-, middle- and even bottom-of-funnel queries. For B2B brands, YouTube and LinkedIn are proving truly valuable. So don't pick a percentage out of the air. Work out how much of your audience is searching on each platform, and allocate accordingly.

Our Search Everywhere Conversation Index is built for exactly this. It shows which conversations are taking off in your category, which creators and brands are being mentioned, and whether someone else is already doing the work for you.

What's the minimum budget to do it properly?

For the typical brands we work with, it starts in the mid five figures. That covers creator fees and briefing, and gets you roughly 12–20 pieces of content a month.

Smaller brands can make a little go a long way. The cheapest trick is getting your audience to do the search marketing for you:

1. Find the platform where your audience is already showing search intent.
2. Identify the conversation you want to own.
3. Design an activation (a competition, gifting, whatever fits) that gets your audience creating content around it.

That costs far less than paying creators or producing everything yourself.

If you still can't afford social search? Run 90/10

I wouldn't drop what's working to chase something new.

Run your existing activity at 90% and use the other 10% to test social search and build a business case. Any bigger shift is then backed by data, not hype.

Remember your website is still the conversion destination. If your PLPs and PDPs suffer because you've neglected traditional SEO, your social search work won't deliver a return.

Keyword research works differently here

Start with intent, not keywords: the pain, need or desire behind the search. On Google someone searches "retro football shirt". On TikTok they search "blokecore". You only find that nuance once you understand the intent behind it, like wanting the best retro shirt to impress your mates in the pub during the World Cup.

There's also no reliable equivalent of monthly search volume on social. Tools that claim to have it use different methods and aren't comparable. Instead, scrape 12 months of content ranking for a topic, take the median views as your benchmark, and track how hitting it moves your branded search over time.

TikTok Creator Search Insights, YouTube Studio and Google Search Console (when connected to your 'owned' social surfaces) are all useful here.

Creators: you don't need the biggest influencers to be found

It was great to welcome Aliya to our AMA Everywhere, Aliya's first appearance since joining the business a couple of weeks ago. She absolutely killed it and shared some awesome insights. During the session she explained:

Traditional influencer marketing sits at the top of the funnel and is judged on reach and engagement. Search-integrated influencer marketing goes further: it puts creator content where people are actively looking for answers.

Aliya had a great example here: when she wants a recipe or a skincare product, she goes to TikTok for a video demo, not Google.

So, should some of the big-creator budget move to search creators?

Yes, some of it. You need both, because they do different jobs:

‍

‍

Aliya went on to assess budget fit. Here, an integrated pot works best because the two feed each other. If you have to choose, make it a creator budget. And make it fluid: start at X in Q1 and increase it if brand searches, landing page visits and UTM clicks move. That's a far stronger case than "brand awareness will grow".

Choose creators by topic, not follower count

Start with the audience: what they care about, what they search for, what frustrates them. Then find creators who already talk about those topics and whose followers trust their recommendations.

For one client we spotted a search uplift around canicross (running with your dog), so we partnered with creators already making that content. A smaller creator who posts consistently on a topic can be worth far more than a big one chasing viral moments. We regularly see smaller creators outrank bigger ones simply because their hooks, captions and descriptions are better optimised.

It helps that algorithms have moved from follower-based to interest-based. If you follow a running creator and you've just got a dog, you'll be shown their canicross content whether you follow them or not.

Top Tip: Make the budget work as hard as possible

Aliya made many great points on the AMA stream, but one I think is essential to planning and budgets is usage rights.  

Secure usage rights and repurpose. Put the video on money-making pages and FAQs, not just a blog post, and add video schema so Google and LLMs can understand it.

Onsite SEO: protect it first, then run it as campaigns

If your budget gets cut next year, protect traditional SEO and your website first. That's still where most of your revenue and conversions come from in 2027. Second, protect the brand-building you do through digital PR.

Always-on retainer or campaigns?

Both. Technical SEO and general housekeeping need to stay always-on; I don't believe in switching them off. Beyond that, a campaign mentality has a lot going for it, and it's the part of our offer I'm most excited about for 2027.

Take festival outfits for a fashion e-commerce brand. Under one campaign pillar you can run:

• Social search content with creators around festival looks.
• Digital PR, both reactive and proactive, tied to trends such as hype items.
• Traditional SEO in support: optimised product descriptions, a dedicated festival edit PLP, and internal links from guides and blogs to that page and between products.

Campaigns are also easier to track. You watch one bucket of related terms, searches for brand + festival edit, and AI recommendations for realistic prompts that include the topic.

And they make implementation easier. "We need this live now" becomes "we're implementing this over the next four, eight or twelve weeks... could we get a relevant PLP live at Week 6 of that implementation?", which is a much easier conversation with the brand.

Organic has never had more to go after

What share of total marketing should be organic?

It depends on your industry; some categories rely so heavily on paid that you have to accept that it will naturally swallow up a lot of the budget. But I think this is the biggest organic opportunity in five to seven years.

In the old model, once you were in the top three for the majority of your trophy terms - with supporting rankings all on page 1, more organic spend didn't move the needle much. Now there's more to win.

Already ranking well? Bring more people to those PLPs through TikTok, YouTube, Instagram and Pinterest. Digital PR is justified beyond links. And if AI traffic starts bringing in revenue, you can double down before the mass adoption we expect over the next two to three years.

In AI search, aim to be recommended, not cited

This is the distinction I'd want every CMO to take into 2027 planning (credit to Meg in our digital PR team for framing it for me at our summer social event).

• Cited means you're part of the answer, but you might not be named. Someone has to click the source icon, then your link, and by then the AI has already met their need. No awareness, no preference.
  
• Recommended means your brand is named in the answer itself. Ask for the best black hoodie for a particular use and, if listicles, product PR and creator styling content all name the same brand as the go-to, that brand gets recommended. People then go direct or search for it by name.

Two levers that get you recommended

1. Long-form YouTube, cut into Shorts. Google-owned Gemini and AI Overviews are particularly likely to cite YouTube. We create long-form content on owned channels or with experts, then cut it into Shorts for TikTok, Instagram and YouTube. In our latest QBRs we're seeing YouTube Shorts picked up by AI Overviews too. We've seen a big B2B benefit from this, even as a mainly B2C agency.
   
2. Digital PR built for recommendation. Listicle mentions, product PR and campaigns that get your brand named as the answer, not just linked as a source.

Don't game it

Monitor groups of realistic, non-personalised prompts, see what's consistently showing up for your brand, and double down on that. Follow industry trends too: if Reddit drops out of ChatGPT's sources, do less Reddit; if Gemini keeps citing YouTube Shorts, do more Shorts.

But don't try to game AI visibility the way people used to game the SERPs. Ask whether the activity makes sense for your brand and audience, even if you removed the AI visibility gains from the equation.

If it does, AI visibility is a welcome bonus. Jumping on Reddit purely for AI, when it was never valuable to your audience, will waste budget the moment the rules change - and your audience may never care anyway.

Which AI tracking tools are worth paying for?

Honestly? None of them, right now.

It's easy to engineer a prompt so specific that your brand is almost guaranteed to appear, but step back and you realise... you have no context of how many people that prompt is being used by or how many people you are visible to as a result.

These tools are also very expensive, and the investment needed to get accurate visibility data doesn't justify the cost yet.

Set the budget annually, review it quarterly

It's getting harder to set a budget in January and leave it alone for twelve months. AI is moving so fast that we could see mass adoption of AI search within three to six months.

Your budget needs room to move.

So set an annual envelope and review it quarterly. If you expected Instagram and TikTok to perform equally, but TikTok is clearly driving brand searches and Instagram isn't, you need to be able to shift most of that content to TikTok. Lock everything in annually and there may be nothing left to act with.

Monthly is too often. Switching between demand generation and demand capture every month gets messy, and nothing has time to work.

What to lock in by January

• Where you want to make an impact in 2027, for example TikTok, YouTube and AI search, while supporting traditional SEO.
• The conversations you want to own. If you know who you're speaking to, where they spend time and what they search for, you can test fast and redeploy budget.
• Your digital PR budget. The one line I'd set in stone, because it benefits every part of the search universe.

Budget by opportunity, not just by channel

You can also assign parts of the budget to specific conversations or campaigns. If festival edits is a £1m opportunity you think you can unlock with £100k, test it, learn from it, and adjust the budget next time that campaign comes round.

And judge organic over the right timeframe. Brands often expect organic to return the same as the same money spent on paid. With paid, you pay for every click. Organic keeps delivering long after you invest, so judge its return over 18 to 24 months, not like-for-like with paid.

How to prove it's working when clicks keep falling

Clicks and traffic are falling, no question. But most of the clicks you're losing are informational.

I haven't sat on the fence with this one... I think it's becoming clear that ranking for those informational queries never made you money, and they often came from keywords that weren't closely tied to your customer - visibility for the sake of visibility, traffic only to appease an SEO's KPI.

If conversion rate, average order value, lifetime value and repeat visits are growing, your traffic is falling but getting more efficient. Being able to tell the board "traffic is down, but revenue is up" is actually a strong position, because it's becoming evident that everyone's traffic is down.

The same logic decides whether SEO budget should go up or down. Traffic falling and revenue flat? The SEO work was probably low-value noise. More investment may now be needed to ensure your SEO becomes effective and efficient.

Traffic down while revenue, conversion rate, AOV and LTV rise? You could see that as a good position to hold the line... The most ambitious brands will see that as a case for more investment still.

CMOs should know what their SEO team is KPI'd on. If it's traffic for its own sake, that's the wrong goal in 2027 and beyond.

The metrics on every 2027 organic report

1. Brand search tied to revenue. This is the headline. If more people search for your brand in 2027 than in 2026, and you can link it to your activity (say, rising searches for brand + festival edit after a festival campaign), your work is having the impact you want.
2. Presence with commercial value. Growing visibility in TikTok and YouTube results and in AI recommendations, stress-tested so it comes from keywords that matter to your revenue, not terms that inflate the numbers.
3. Efficiency. Conversion rate, AOV and year-on-year organic revenue.

Share of search

Share of search is your brand's share of all branded searches across your competitive set. If Airbnb, Hotels.com and Booking.com together get 120,000 branded searches a month and Airbnb gets 40,000, Airbnb has a third of the share of search. The more people search for you by name, the less you have to fight for competitive non-branded terms - your brand is the preference.

Linking sales to platforms that don't send traffic

• Brand + term searches. Create festival edit content, watch searches for brand + festival edit rise, then show the page ranking for that term getting more clicks, conversions and higher order values.
• Direct traffic to landing pages. Useful, but treat it with caution, because other brand activity drives it too.
• UTM links wherever you can, especially in YouTube descriptions.

On platform, look past views. Saves and shares signal deferred intent, comments show what people took away, and watch time tells you whether they found it useful.

Track views over time as well: viral content spikes then falls off a cliff, while content that ranks for a search term, keeps getting found.

Who owns the budget, and the results?

AI search: ultimately the SEO team, but digital PR has to sit with it, ideally by bringing the digital PR team or agency relationship under SEO. That combines onsite, technical, brand amplification and product PR, so the team can fairly be held accountable. If digital PR sits outside search, it's unfair to hold SEO entirely responsible for AI visibility, because PR contributes so much of it. For B2B brands leaning on YouTube, there's a case for the social team sharing accountability too.

Social search: a collaboration. Search finds the opportunity and brings the data. Social brings the hooks, the creators and the ambassadors, and decides where content fits the roadmap. Brand oversees it all.

Brand teams used to frustrate me. I've come to see them as custodians of the brand rather than gatekeepers. For years SEO chased visibility and put content live that did nothing for the brand - looking back, I can totally understand why they didn't see the value in our work, to be honest.

Now search can look to brand for guardrails, and for clarity on who we're actually speaking to, the types of conversations we want to participate in and own, and where we want to represent our audience community - with this mindset, brand will truly understand the value of what we bring to the table. AKA getting the messages they spend so long developing in front of the right person, facing the right intent problem, on the perfect platform at the perfect time.

Funding social search from the search budget is easiest, because it carries search metrics. But there's a good argument for splitting it across search, social and brand, since all three care about brand + term search. Making that case can unlock more budget overall.

Plan together, on shared KPIs

With a few of our clients, for the first time in my career, all-agency meetings have felt genuinely collaborative. That's because everyone is working to the same KPIs, so growing one thing helps everybody. As a marketing director you then get one QBR showing how the digital team grew brand search, share of search and revenue, instead of three separate ones.

My final advice to CMOs planning 2027

Don't think about your organic budget line by line, channel by channel.

Instead, plan it around the conversations you want to own, the moments you can connect with your audience, where you can be influential and the brand searches you want to grow.

Protect what makes money today, put the first new pound into digital PR, give AI search a small pot to learn with, and review it all every quarter.

‍