Q&A-gency: Loz Horner

We’re back for the second edition of Q & A-gency where Shape Insight’s Alex Holmes speaks with someone in creative development or media planning who is at the top of their game. Why? Well, it’s all about understanding how insight shapes their work today and tomorrow.

For this instalment, I spoke with Loz Horner. During his 30 years in the industry, he’s worked at Leo Burnett, Wieden + Kennedy and Mother, before settling into his current role as Strategy Partner at Lucky Generals. So let’s find out what he had to say for himself.

----------------------

Q. Curious to know, what change has had the most significant impact on what you do today?

When it comes to generating insights or finding insights to solve problems, I do think speed is the biggest change. The sheer speed at which we’re expected to find them, crystallise them, and then use them has ramped up.

I don’t think that’s because I’m just getting old. Or at least I hope not! I think it’s genuinely because things are getting faster. I also think that clients have less time to spend on strategic development, which means agencies have less time to work on briefs.

Another thing you have to consider is that, as I work in a creative agency, my end client is the creative team. I don’t want to have their part of the process squeezed because I’ve failed to deliver something they can work with. This means my colleagues and I can often find ourselves squashed between a client who has thought about the brief for quite a long time and a creative department that wants to get going and wants us to give them the starting point.

Sometimes it’s just a few days in which we’re expected to turn around the brief – and that obviously puts a lot of pressure on what type of research we can do.

Q. Devil’s advocate: hasn’t that always been the case? Every pitch deck begins with “the world is changing fast”…?

I don’t think I’m harking back to a time where we had a glorious 3 or 4 weeks to do some qual and mull it over – I don’t think those days ever existed. So, in a way, I do agree with what you’re saying, there has always been some form of pressure.

But I genuinely think there’s less time given to ‘Proper’ strategic development research. That limits your toolbox for finding insight. It means you might defer to AI to start shortcutting the process. It might mean you develop a few hypotheses and do a quick bit of quant to validate them. But in saying this, you can get to some brilliant answers very quickly, as my colleagues constantly remind me!

Q. Going broader then, what else has changed in terms of your day-to-day role as a planner?

The cliché is that everything changes and nothing changes. And like all cliches it’s true.

There are a lot of changes. Some are more cosmetic – like the way we build our ideas and how we present them to our clients. For instance, it’s much easier to visualise your strategic thinking or creative thinking to a client via AI now.

There is more stakeholder management these days too. I think there always was, but today you’re often having to help your client manage their boss. For instance, you might have to make the case for brand building – which is a long-term investment – because it’s at odds with the financial pressures for the board, who question whether the money spent on advertising could be better used on other things.

However, lots of things are the same. Human insight is still king. We’re still basically trying to understand people.

Q. So how has your approach to insight changed?

To answer this, it’s good to know what type of agency we are.

Lucky Generals is now a kind of medium-sized agency - so about 100 people, but we're not an BBH or a VCCP - we don't have big resources.

We came up as a startup, and when you're a startup, you learn to bootstrap a lot of stuff yourself. We keep that mindset today, that’s why we do a lot of our own research. Before COVID, everything was face-to-face, but now everyone’s comfortable jumping on a video call. That helps a lot. There’s also been a rise in self‑service tools that let you do quick, homegrown research work - great for getting a fast read on an insight or strategy and some evidence to support it.

Also, our approach to insight changes depending on the type of client we work with. You have the big established clients who have econometric data, brand tracking, and a big marketing department supported by an insight function. But often we work with smaller clients - perhaps that’s a startup that has kind of scaled and now needs to advertise more broadly, or wants to do a brand identity project. These types of clients don’t always have the budget or time to invest in what you and I would call ‘proper research’.

Q. Picking up on your point on self-service, you could argue that in a way your kind of marking your own homework?

Yes, I think you’re right, there is that risk. And I guess that’s the value that an independent external research provider brings – objectivity.

But, for different parts of the process, there are different factors to consider.

For instance, if we’re pitching, we are literally paying for it ourselves. And so, on those occasions, we can’t always afford to pay an external partner. There is a total pitch budget of a few thousand and you can’t spend all of it on an external moderator.

Once you get further down the path, then I guess that's when clients would also insist on having an objective third party. For instance, it would be rare for a client to let us be the qual researchers of our own creative concepts!

Q. I’ve always admired Lucky Generals for doing your own research. It reminds me of ‘Truth, Lies and Advertising: The Art of Account Planning’ by Dan Steel, where planners really understood the research process.

I think what you're raising is that there is enormous value in knowing how to do research, not just understanding the output of research. How do I assemble a project? Who am I going to speak to, and how easy will it be to find these people? Right through to how to write a discussion guide or survey.

I always try to bring new planners in and sort of say, you should do your own groups because it's such a sort of excruciating learning curve, isn't it?

I remember the first groups I ever moderated, and I was so bad at it. I tried to fill every gap of pause in the conversation and ended up just talking at them. Afterwards, my boss, John Poorta, took me aside and very nicely said ‘You know, just wait for them to talk. You need to be okay with uncomfortable silences.’

We’ve actually just launched a brilliant new initiative, led by our CSO Damien Le Castrec, where we’re going to give our planners the chance to make a real documentary about a specific target audience. Why? Because documentary makers know how to make research feel like must-watch entertainment.

Q. I’m curious to know, what role does synthetic data (i.e. respondents) play for you?

Can I broaden the question, firstly, to be about AI in general?

I’ve seen some really good applications. So, for instance, when you’re doing your own qual, things like Otter AI are incredibly useful and definitely reduce the amount of time it takes to go back through the conversations to find the relevant comments or themes.

We’ve built synthetic respondents for certain types of shoppers for certain clients and then tried to give them the personality traits that would fit the brand in question. So, down-to-earth, hardworking mum who’s got 3 kids, lives on a tight budget. And you talk to them, and you get back what you expect.

I feel like synthetic respondents are another point of reassurance in the insight validation journey we go on. By which I mean, if you’ve already found an insight you think is interesting and intuitive, then you can test it or try to replicate it.

I haven’t yet had an interaction with a synthetic respondent where I was, oh my god, that's a completely fresh idea or unexpected observation. But perhaps that goes back to the heart of what current AI models are good at – or maybe I’m just not using it correctly yet.

Q. Listening to you, it feels like any creative agency can now generate its own insights, which puts pressure on research agencies to prove their worth by finding genuinely new ones. Is that fair?

I don’t love this idea that every bit of research has to lead to a never-before-seen revelation. I don't think it's fair or reasonable to expect planners or indeed research agencies to come down from the mountain with some incredible tablet of stone. It's not going to happen.

I’ve always liked Merry Baskin’s explanation of what a good insight is: “An insight is an undiscovered cliché”. It’s an insight that as soon as you hear it, it's like ‘Oh, that’s really ******* obvious’. But then it leads you to say ‘ohhhhh, that's nice!’ I think the way she sees it supports my idea that agencies don’t need to be unearthing completely new insights all the time.

Bringing it back to AI, I do think it’s very easy for AI to regurgitate lots of truisms about human experience because it's drawing from the internet, isn't it? And so it will quite quickly give you the 10 things that are obvious. Therefore, because AI can get to these facts in minutes, perhaps you should be wary of using them willy-nilly.

Q. This seems to echo Will Lion’s “Pentagon pizza” article about building an “insight edge” through curiosity, ingenuity and bravado, not just a subscription or first base AI.

Yes, that’s what I’m getting at, and I think he’s right. Though those words sound a bit intimidating.

I’m not sure I personally have enough of any of those characteristics to claim to have an ‘Edge’, but I think if you’re a good planner or a good researcher, you should want to know what the broader context is. Do your homework. Understand the client’s situation. Understand their problem.

By doing that, it allows you to put two things together and make a spark - that you can now see that this bit of information over here is in fact or can in fact be turned into an insight that solves the client’s problem. I think that what Will was getting at.

Q. Curtis Weir told me insight makes work intentional and that tensions should kick things off rather than block them. Do you see it like that?

I think what Curtis maybe means by the word intentional is that the insight has a purpose. It’s not just a pointless observation; it’s helped unlock an issue and shown you a fresh way to solve it.

But if I’m brutally honest, ‘tensions’ is a bit of a new buzzword. People are now obsessed with defining insights as tensions, cultural tensions or consumer tensions – so that the brand can resolve that tension.

I slightly struggle with it because I think it's pushed us to obsess over tensions, when tension is really just one flavour of insight, and insight can come in many different varieties. Sometimes they are a tension resolution yes, but other times they're funny jokes off the internet. But good insights all share one common thing – an ability to help you see the problem in a different light and find a solution.

This is the same way as Jeremy Bullmore describes insights. He said a good one is like a refrigerator: ‘When you open it, a light should come on.’

Q. We’ve discussed before the importance of work being “culturally aware” or at least have the ability to “resonate more broadly”. How does insight (and agencies) help you do this – if at all?

My thought on this is that ‘Culture’ is a big word. Because when someone says, ‘Let’s impact culture’ it can feel like, ‘Oh, let’s change the world’. It’s like the word ‘Viral’ which used to be thrown around a lot in my industry. ‘Make us go viral’. And you’re like, ‘Well, that’s really hard guys, but we’ll do our best!’

I think the interesting thing about pursuing ‘Cultural’ insight is that it can sometimes lead to quite narrow and specific trends. And by that, I mean you can end up quite quickly with a trend report about really cool but very niche behaviours or subcultures which are interesting, but often aren’t useful for the brands we work on which are quite mainstream and broad.

I think it’s much harder to find genuinely mass culture insight – or a way to move or reflect culture at a mass level. The McDonald’s Raise Your Arches campaign is rightly celebrated because it achieved that – that’s a ‘mass insight’. 

Q. I’ve previously discussed whether or not tools from System1 et al, who use Facial Coding and Emotional Response measurements, could lead to homogeneous creative output – what’s your view on this?

I think it’s easy to blame System1 because they are the market leaders. I remember arguing for facial expressions and emotion measurement because back then it was all Millward Brown link tests and a Persuasion Score, and we were saying, hang on, not all advertising works like that; a lot of it operates emotionally!

But now I'm just worried about one single definition of effect winning - because that's what leads to the homogeneity.

So we'll go back to System1. Emotion is good. We all know we need to create it. But picking the smiley face versus the sad face from a list whilst watching an ad is a fairly blunt way to assess it.

I'll give you a real example from Lucky Generals’ history. We made this Funeralcare ad after COVID. As you can imagine, rather sadly, the funeral care industry did very well during COVID because lots of people died. And then after, thankfully and happily, deaths went back to pre-COVID levels. But that meant if you were going to try to grow your business, you needed to almost get people to think about Funeralcare before they needed it.

That became the brief. And so, we made this campaign and it encouraged people to have a conversation about their funeral and what they would want – it was very emotional, but very sensitively done. And then we put it through testing where a person is being asked to tick a box about how it makes them feel – and of course they didn’t always pick the happy face – it’s an ad about funerals!

I know I'm being overly simplistic, and I'm using an example that is an extreme, but I'm using it to make the point. I do think advertising needs to have the scope to work in different ways based on the client, the problem, the market, who you're talking to, what you're trying to sell, etc. I'd rather have some nuance to the methodology based on the nuance of how the advertising is intended to do its job.

Q. Last question, do you think AI will make insight better or just faster?

Is it an aid that speeds things up? Big tick. As I said, transcribing conversations and pulling out themes is something it’s great at. Same with speeding up desk research.

Beyond speed, it’s another useful tool which supports your thinking. The things you naturally did as a researcher or as a planner, which is, once you’ve had a few conversations and a theme starts to emerge, you often go and look for more evidence in the real world. You find it memes, or reddit communities. I feel like AI can definitely be used in the same way.

Can it help me be better at my job? Definitely.

But, will it be a replacement for humans? Not sure yet. We will see!