Learn who your audience is and find out where they are
Travis says: “Audience-first marketing has and will always win, no matter how technology advances.
It's not a new concept, but it's more critical than ever. With AI entering the space, we are seeing huge shifts with AI and agentic search. The only way to thrive is to really understand your audience and how the people you are trying to reach search and find brands.”
Do most brands do a terrible job of knowing who their audience is?
“Some brands do it better than others, but when you look at traditional persona mapping, you end up with wildly obscure personas.
There is a meme comparing Ozzy Osbourne and Prince Charles because, to a persona map, they look identical. They're old, they're English, they're rich, but obviously they're into wildly different things and have very different interests
It's really important to use the real-world data that we have access to in the search world to map this back to what your audience is actually searching for. What do they care about? How do their attributes ultimately show up within their search?”
How do you map search data to your audience, and how do you determine the sweet spot of establishing the correct audience size?
“A lot of brands approach it with focus groups and things like that. They take a focus group of 10 to 20 people and then extrapolate that data. I would not recommend only that, because you will end up with the wrong outputs. What I would do is really try to understand the entire customer search journey.
If you are a shoe company, and someone is a nurse and they're on their feet all day, they are going to have a very specific need: shoes that last a very long time. Then, you start mapping that to what that user might search to fulfil that need. You conduct all of the keyword research, then you map it back to that customer search journey funnel.
From there, you don't just look at Google. That nurse was probably influenced by a TikTok influencer, an Instagram influencer, or Amazon. Pull search volume data and search data from all these platforms. Really start to understand how that audience member is searching and how they were ultimately influenced in that search journey.”
Do you then tie it back to likely buyer intent and imagine how that persona would be interacting with your content at each stage of the journey?
“Exactly that. At Brainlabs, we call this ‘digital signals analysis’. The second stage of that is the strategy that comes from that data.
Buying intent is really important, but that's not always the goal. We know that decisions (purchase decisions, brain decisions, etc.) can be made even without purchase intent, if someone's not immediately in the market, for example.
Once you have all of that data, you can map it to the type of content and the type of call-to-action that ultimately makes sense for where that person is in that journey. That is how you see the best results.”
Do you advocate having multiple personas for the same buyer journey?
“Every brand is going to have multiple audiences, hopefully. In my opinion, what usually drives brand success is having multiple audiences.
Yes, I would break down that search journey and that keyword research by those audiences. Then, you can layer that on top of your strategy and build strategies specific to each of those audiences.
The latter part of your buyer journey may be similar, but because you're targeting multiple audiences, the initial stages will be different, with different content and different landing pages. This is so important in the realm of AI search because we know that personalised content really matters there.
Someone on ChatGPT or another LLM is typically logged in to that experience, and that AI search engine will remember every previous search. If it knows that the person is a nurse, they have disposable income, and they like to go hiking or running, it's going to recommend a different shoe than someone who just goes to Google and types in ‘nurse shoes’.
To me, that shows the value of knowing your audience through and through, and then building very personalised content to let the AI search engines know exactly who our audience is and who we're trying to go after.”
You say that SEO has the best data and should guide all marketing strategy, but how do you convince other members of the marketing department that it’s an SEO’s job?
“Influencing change is always challenging. One way of doing this is to really show them what data is even possible within the realm of SEO.
A lot of other channels and practices don't totally understand the power that we have. We basically have the world's largest search data set on how people think and their intent on the Internet. We have access to that. That's really powerful. When you initially show them what we are capable of and what we have access to, I typically see an aha moment where that person realises that this data can help advise what organic social content strategy we launch, what email content we should launch, etc.
Showing them our capabilities is number one, without a hard pitch, and then number two is understanding their goals and KPIs. You want to then align your strategy to theirs, and understand what their motivations are as an individual. Do they want a promotion to Director of Marketing, for example?
That’s how I approach getting buy-in from other teams.”
To extend your research beyond Google, what other data sources can give you a broader perspective of who your audience is?
“I always start with SparkToro, because it’s a great audience research tool and it'll allow you to see what social channels the audience is hanging out on and what AI tools they use. Then, you can use that to advise strategy.
When it comes to the sources that typically pop up when we do this kind of analysis, Reddit tends to be a really big one. We know that Reddit's rise has been huge within Google, and it's also typically one of the most cited sources within LLMs.
There are a few different ways to get data from Reddit about your audience. They have an API, and I highly recommend using that. You can run a sentiment analysis to see how people are talking about your brand.
Also, Reddit ranks. You can pull Reddit's rankings and then distil that down by brand name, non-brand keywords, etc., to really understand how your audience uses that platform and what subreddits people are interacting on related to whatever term you want to target.”
Why do you believe that getting prepared for agentic search means having radical clarity in your entire marketing funnel and audience intent across that, and why is it important?
“I have had the fortune of getting beta access to agentic search, like Perplexity's Comet, ChatGPT's agent, Google's agent, Copilot's agent, etc., and it dramatically changes the searching experience.
Rather than me needing to go out and do a bunch of research on shoes, cars, credit cards, banks, or restaurants, I can ask the agent to conduct that research and then show me a summary of what that research is. It's a bit of a fortune teller, if you will.
Once people go through that experience, you will absolutely see certain comparison searches shift over to agentic search rather than someone having 10-20 tabs open trying to conduct their own research in that way.
Brands can and should prepare for that now. Use good technical SEO. Don't have a tonne of JavaScript, so crawlers can easily find and access information. Have great personalised content so those agents (who will have the personalisation information from previous search queries) know it is a good site to search, etc.
Preparing all of that will make sure that brands are ready for that agentic search world. While it’s currently only about 2-3% of searches, even if that only increases to 5-15%, that’s a meaningful segment of your audience that you don’t want to miss out on.”
How do you measure the impact and value of AI search if the content that the user interacts with isn't on your website?
“It’s difficult. One of the biggest KPIs we've been looking at is shifting away from clicks and traffic, and thinking about it in that way.
Anytime an LLM presents trained data, there usually isn’t a source to click. We're seeing direct traffic increase on sites. We're also seeing branded searches increase, assuming you have good AI visibility.
Being able to layer data and really understand that search journey for those audiences allows you to tell a more complete story of how much AI search is ultimately impacting that search journey.”
Has this change in focus and the understanding of the value of audiences changed the way that content marketing is done?
“I would say so, for sure. In the past, a backlink was a backlink. We were just happy to get any of them, as SEOs.
Now, it's shifting more toward how we better target what the audience is going to be most impressed by and influenced by. This is why Reddit is so important. Your audience is already spending time there. Let's build the bridge between organic social and paid social strategy to make sure we're showing up where that audience is.
That's just one example, but then you start getting into digital PR, and we're no longer chasing links; we're chasing brand mentions. That has arguably been the better strategy for years at this point, but today, we know that AI search doesn't really care about backlinks. It's more interested in brand mentions.”
Travis, what's the key takeaway from the tip you shared today?
“We have the skills to prepare for the AI future, and we have access to data that allows any marketer to fuel growth and performance in their channels.
Build those bridges, get excited about the future, and make sure that we are harnessing the data that we can access to build those bridges.”
Travis Tallent is Managing Director (SEO and CRO) at Brainlabs. Find out more over at BrainLabsDigital.com.