Investigate your audience, and discover how user intent is changing

Becky says: “We are seeing huge shifts in how people search, and you need to stay on top of that, but not just by looking at what platforms are doing.

You need to understand what users are doing, and how their intent is changing where they search.”

How is user intent changing?

“We have more platforms than we've ever had before. At Reflect Digital, we’ve been doing research over the last 12 months, and we've committed to continuing this research and polling the public to understand how people are searching. What we're seeing is that search is more fragmented than ever.

That 18-44 demographic (which is a large chunk of the target audience for many brands) uses an average of 5 platforms in their search journey – and even more, in some cases. That can be anywhere from the beginning of their journey through to the end; from the awareness stage to filtering down their decision-making through to closing their decision.

In that process, they are looking across multiple platforms, and they're looking for different things. They're trying to understand the best option to buy, the best service, the best place to go, etc., but they're using different platforms – and all the while we've got more platforms popping up.

All of the LLM/AI platforms that people are using to search are making that more fragmented than ever, and businesses need to understand that because they need to be visible.”

Is there a certain age range or user type that you're targeting with this research?

“We're doing wide-ranging research covering all age groups, and we've got qualifying questions each time, which help us to slice the data to see what's changing and home in on different markets or different groups of people.

One of the things that's interesting with the advent of AI, and how people are using it to search, is that occupation is now becoming a predeterminer for how a user might search. If you work in IT or digital marketing like us, where you may be more likely to use AI or talk about AI at work, you are also likely to be using that more for personal search as well.

Whereas, for people who classified themselves as working in the performing arts, 0% were using AI for search because it’s not touching their world at the moment. It’s not something that they’re thinking about.

Whether you're a B2B or a B2C business, occupation is changing the behaviour of your audience. For some, you need to show up in a place where you might not need to be visible for others at the moment. That's going to change over time, as people become more aware.”

If your audience is experiencing your brand for the first time on platforms like ChatGPT, what research can you do to understand their behaviour?

“It’s tricky, and it is going to continue to be tricky – especially with things like AI overviews, where they might be seeing the answer and not even clicking through. In that case, you don't even know that they touched your business and saw your answer. They may or may not have spotted the brand, but that was still part of their journey, and you still want your brand to have been giving that advice where possible.

The key is understanding your audience and being led by that, first of all. Spend time talking to your audience in both a quantitative and qualitative way, whether that's surveys, an annual thing, or something you can do more frequently.

Obviously, things are currently changing at a speed that we've not seen in a long while. We've been charting this across 2025, and the September edition was the first time we saw the use of AI jump up by about 5%, and therefore, Google had moved down a little bit. In the previous editions, it stayed flat.

It's trying to understand what that looks like for your business. Yes, there are reports like the ones that we're doing, but those will only tell you generically what is happening. What you need to know is what's actually happening for your audience. They'll give you some idea and some thoughts, but you need to know who your audience is and how they specifically go about searching for your product, so that you can understand where those touchpoints are and what type of content you need to have.

At different points of that journey, people are going to be expecting to see a video that they can watch. At other points, they're looking for ‘Top 5 Reasons to Buy’. You need to understand what that looks like and make sure that it is still good for Google, but also for LLMs to use that content. Hopefully, they will then serve you in their answers, and pull your brand to the forefront.”

How do you paint the picture of who your audience is?

“There are a couple of different ways you can do this, but the easiest is probably to look at the customers that you have had (especially for an established business with a database of those people).

Obviously, that still needs to be the type of customer you want to target going forward, who fits where the business is going, but then you can get an understanding of who is already in your pool of customers. You can get demographic info from them and ask them those questions, but the most important thing when building out personas is understanding their motivation to buy.

Why did they interact with your business? These could be people in your database who haven't bought yet, and that's also telling and interesting to understand. Who is following your newsletter, following you on socials, engaged in what you say, but hasn't spent money with you? That might give you some research that helps you move those people from newsletter signups to actual customers.

You want to know what motivates them. Why do they follow your brand? Why do they buy your products? How often do they buy? It depends on what type of business you are – if you provide something that they would buy regularly, it’s a once-every-few-years purchase, a service, or a place they visit.

When you understand motivation, you can bucket people in that way, and use that as a springboard for creating content that ties into what their motivations are.”

Can you establish a motivation from data, or do you need to establish that from a conversation with your customer?

“You can do a bit of both. We've got a behavioural team, and I am very grateful for them because I've learnt a lot about how we, as humans, can write really biased questions sometimes.

It's best to get someone who knows how to write questions that aren't biased, because we often put the answer that we're expecting into the question without meaning to, and bias the answer. You want to make sure they're unbiased questions, and you can do that in a survey format that’s data-led, but with some qualitative questions in there.

Alongside this, you should also try and do some more qualitative research – whether that's getting people in a room around a table and trying to talk to them, or it's one-on-one interviews that you do on the telephone. You'll get some richness of that data, which no doubt will complement the more number-driven data that you've got through doing a survey, and you can start to feel the depth of that, the understanding, and therefore see what it means.

Sometimes you need to see them interacting with your website or your product to realise what is unspoken. There'll be some things that you'll get from a conversation that are deeper than what they've actually said, and sometimes it might even be seeing them interact with your product.

There's a fabulous case study of Heinz and ketchup, back in the ‘50s, where they brought in families, served them a meal with ketchup, and watched what they were doing. This was the turning point for them to bring in plastic bottles because they realised that, with a glass bottle, the child would go to pick it up and the parent would take it from them, because glass is dangerous. Therefore, they would get a parent's rationed amount of ketchup, which is very different to what a child would like to put on their plate.

They realised that they needed to change their bottle and their product by seeing something that they would never have seen in a survey. Sometimes, seeing users interact with your business will allow you to spot something that tells a story about their journey. It might be seeing them go and search, and seeing how they move about online.”

Do you create a singular customer journey that represents the majority of your users, or do you have multiple journeys?

“We would do that per motivation group. You've grouped these people together on motivation, and then you look at what they do and how they do it differently.

Some of it will be the same, or points might be the same. It might be that, close to buying the product, nearly every group will go and look at reviews, or they'll search (whether on Google or ChatGPT) ‘What do most people think about buying this product?’ and try and get that review data. That might be consistent for everybody, but the up-front piece might be different.

I’ve said before that demographics aren’t so important, but one thing we are seeing with AI is that younger people are more motivated to use it, and they feel more familiar with it than older generations. You do have that split. Then, there is the occupation split, which will probably go away at some point.

At the moment, while we're on this change curve and people are quite early in adopting how they search using LLMs and all the different platforms that have appeared as day-to-day options in the last 12-24 months, occupation is having an effect. Over the next year or two, that will fizzle out and become less of a determinant.

When something so new is happening, you do have these different effects. Earlier this year, we saw the craze where people were making models/action figures of themselves in AI. They were making their personal version of a Barbie or whatever, with their face on it. That kind of thing helps get it more into the masses.

People who use social media and don’t work in a sector where they've been playing with AI suddenly hear about this tool and go, ‘I want to go and play with that!’ They have a go, find it interesting, and then think, ‘Can it do other things for me?’ Things will hit the media that will bring more people on board, and over time, it will change.”

How do you ensure that reviews truly reflect your business, and how does that impact the user journey on LLMs?

“Reviews are going to be true to who your business is, so you need to smarten up your business if you don't like the reviews that are out there. You also need to be encouraging people to leave reviews and make sure that you've got a plan around trying to get that data.

An LLM is trying to act like a human and speed up the process a human would follow. When we ask, ‘Where's the best place to buy X?’, it's trying to do all the different things that we would do. As part of that, it will go and look at what other people say, and one of those routes will be looking at reviews. Another part will be looking at press. Who is this business? Do they talk about what they do? Do other people talk about what they do?

It’s similar to how search engines have worked, but in a slightly different way because it can play that information back to you instead of giving you the blue links. It's pulling all of that information together and crafting you a story to answer your question, and trying to speed up what you might have spent quite a while doing yourself.

You need to focus on getting reviews, think about PR, and think about the full marketing strategy and how important that is to the search journey and to being found wherever people are searching, because SEO is much bigger than Google.”

Becky, what's the key takeaway from the tip you shared today?

“You need to go and understand your audience. You need to talk to them. You need to find out what's happening.

Don't just do this once. You need to keep revisiting this because it’s going to keep changing. Then, use that information to map customer journeys and understand how to adapt your strategy in a world where search is changing.”

Becky Simms is Founder and CEO of Reflect Digital. Find out more over at ReflectDigital.co.uk.