This article was originally published on The Drum, on 4th December 2019.
It seems like a lot of digital projects are kicked off with the question 'what are the personas we’re targeting with this?'
Often the personas presented are designed to create a really well-rounded, fictional human being who has thoughts and feelings and aspirations. And two kids. And drives a BMW. And has a house in the suburbs. But lately we’ve been questioning how much this type of persona development is really helping businesses when it comes to their digital marketing.
So we decided to get Sean and Tom (our Client Director and UK CEO respectively) together to talk about whether - and if so, when - personas can really help us as a digital marketers.
Sean Philip: The way I’ve been taught to think about audiences is to ask 'what are the goals of a broader audience group? What do people want to achieve in the finite amount of time that they’re interacting with your brand?' So when I see personas that imagine a whole life for these hypothetical people we’re targeting… I really question how relevant that is for the type of work we’re doing.
Tom Jones: So you’re strongly against personas?
SP: I’m absolutely not against researching who your audience is and what they do in the context of their engagement with your brand. But I think you can spend a lot of time and money developing a beautifully creative vision of the sort of person you might want to be speaking to. But in reality, they don’t exist. And it’s too narrow a focus.
Personas from a digital targeting point of view... their day is done. Because when it comes to targeting online, the machines are better than we are at that.
TJ: I think researching the audience is obviously important. But I mean, I’m not strongly for personas. And the worst ones are just a pure exercise in post-rationalised thought. Personas from a digital targeting point of view... their day is done. Because when it comes to targeting online, the machines are better than we are at that.
You’re better off using Google and Facebook’s supercomputing power to improve your campaign targeting, and reinvesting the time efficiencies from that into strategy, creative planning and evaluating the impact of your campaigns
SP: We had personas from a client that they’d invested a lot of time and money on. They’d used them as part of their creative ideation. So they knew the sort of messaging they wanted to use for each persona group. It informed their TV ads and everything else they did. They gave that to us and said “...well we’ve done a lot of research around who these guys are. So we know we can also target them online. We know how old they’re supposed to be, whether they’ve got kids, where they shop…”
You can build that up in Facebook really quickly and easily, and then target those people. So we did that and we found that actually, it didn’t perform any better than if we’d just not really put any targeting parameters around it, and instead just told Facebook to find the best audiences. And I think that’s where the application of persona thinking to digital targeting just doesn’t work. I think we’re in a post persona world, when it comes to that bit.
TJ: Yeah, technology companies like Google and Facebook, increasingly their message is take a step back and go broader, to allow their algorithms to do the work. And that’s not about sitting back and ignoring the targeting aspect of campaigns. It just means we have an opportunity to focus more on testing different campaign types, experimenting more with creative and so on.
^^ A pic of Sean probably-not-looking-at-a-persona
SP: On the other hand though, in digital a lot of the time we’re saying “how can we optimize this campaign? How do we target this audience even more specifically?” But the more precise and specific you go, it can actually narrow your focus around who you’re looking to engage. And although that can be a good thing, it’s only really the case if you’re trying to get more of the same people to buy your product.
You can either increase the size of your market share, or you can work to increase the size of the market in the first place. I think the first one is what those tech platforms are good at doing.
There’s two things you can do from a marketing perspective. You can either increase the size of your market share, or you can work to increase the size of the market in the first place. I think the first one is what those tech platforms are good at doing. They’re saying “there’s a market out here for your product. We’re gonna get your ads in front of that market and convince them to buy your product.” But the tools we have, the algorithms we have, the way everything’s set up to be constantly optimizing doesn’t solve the second point around increasing the size of the market. And as marketers, we should be trying to do that too.
TJ: So maybe that’s where there’s more of an argument for these more creative or imaginative personas. Maybe they are useful from more of a ‘big idea’ piece, which speaks more to that second bit around increasing the size of the market for a product.
SP: Yeah, it’s almost about drawing an audience in such high fidelity that you can picture having the conversation with them. So you can tailor what you’re gonna say to them in really specific ways, that might help you develop the right creative for, say, a mass market campaign.
TJ: So you’re back on board with personas?
SP: I can see why they’re used by creative teams, from that perspective I think.
Should digital people be talking more to creatives about what we’ve come to understand about who the audience actually is?
TJ: Ok, so if that’s the case, is there an opportunity for better communication between digital campaign people and creative people? Should digital people be talking more to creatives about what we’ve come to understand about who the audience actually is?
SP: That’s a really interesting point. Taking that example from the client I mentioned… Ultimately we found that, regardless of all this persona stuff that they’d done beforehand, women over the age of 65 bought loads more of the product. And more consistently. And converted more than any other group. And that was unexpected. Their research hadn’t predicted that.
So feeding that kind of insight back up the chain potentially changes the approach from a wider marketing perspective. Maybe rather than saying “These personas are your targeting parameters, go and find these people using digital”, we should let the algorithms do the targeting first. See what’s working well. And feed that back into creative teams.
TJ: I think that’s something that probably needs to be done more often. It’s just, often the digital bit is seen as so executional that it doesn’t then circle back up to the strategic planning phase.
I also think personas can potentially be more useful if they’re built in a different way in the first place. Rather than focusing on characteristics or demographics, they should be asking, “What are this person’s goals? What do they need from this interaction?”
SP: Yeah. That’s going back to the motivation and need. What are the things that a person needs to achieve when they come through to our website? Or taking the website out, what information do they need from us to make a decision? Then we can think about what sort of content we want to show those people, that’s gonna help them achieve that goal. Not getting bogged down with, “...and this person has three children, they like going to salsa dancing at the weekend”, you know.
TJ: So basically we’re totally up for personas if they’re linked to people’s motivations.
SP: It has to be linked to their motivations, yeah.
TJ: And we want to experiment with a kind of… reverse persona development, where insights from digital campaigns ladder back up to creative work, to inform the personas that get created in the first place.
SP: Yeah, exactly! I think that’s an approach to personas I could really get on board with.