Innovation in action with Michelle Booth (Bonus)

PLEASE NOTE: This transcript has been created using fireflies.ai – a transcription service. It has not been edited by a human and therefore may contain mistakes 

 
00:03 
Dom Hawes 
You're listening to Unicorny and I am your host, Dom Hawes. Now, this is a bonus episode. It's free. Hang on a sec. They all are. Okay, this one is extra free. It belongs to part one and part two of Michelle Booth's amazing podcast because when she joined us in the studio to talk about marketing transformation, product incubation and agile and marketing in action, we close the show with a discussion around a pest analysis like we normally do. And pest, by the way, is political, economic, social and technological. Now, normally we edit the main show down, and we edit these bits down too, to make one crunchy show. But this time around, there was so much good stuff, we decided to roll it in full and turn it into a two part episode with a bonus.  

 
00:47 
Dom Hawes 
So this is my conversation with Michelle Booth, where we talked around a pest analysis. I hope you enjoy it. Not everyone likes talking about politics, and we're not going to talk about party politics, but clearly this is the year where there are more elections around the world than any other time in history. There are two in particular that I think impact our culture. The US and the UK, both at the moment are looking like changes of government. In the US, that probably means upheaval in many ways. Not sure it does for us, but as marketers, what would you be advising your fellow cmos and senior marketers to be thinking and doing this year, given that we've got normally one of the most disruptive political events of the cycle?  

 
01:28 
Michelle Booth 
I think it comes back to the conversation were having about uncertainty. I think uncertainty is now the norm. I don't think we're going to get to a place where we can all rest and go, oh, now we know. Having said that, I think that if we do have more long term leadership, that means that we can start to know the direction of travel in terms of things like policy, and also be clear in terms of where we're going. But I think that the most important thing is that we need to really stand firm in terms of our brand, what we're trying to do, and then be able to adapt to the climate that we're in. Never is it more true in terms of investments and market context.  

 
02:06 
Michelle Booth 
But I think that if you understand where you are and what you stand for, you can flow within that, knowing that you need agility, not just as a nice thing to say because it sounds good, but also because an 18 month marketing plan is never really going to stick, I don't think, ever again in the future.  

 
02:25 
Dom Hawes 
Yeah, I mean, I was asked by someone this week what our three year plan was. It's like, I don't know. I have no idea what's going to happen next week, let alone next year. So I think what's interesting is, and what I do actually is I think you're right, by the way, and I agree that uncertainty is here forever. There always was a bit, but we used to live in an era of three and five year plans because things moved more slowly. So I think I have things that are not going to change or probably not going to change on which we can rely and things that are going to change that we can't influence and just have to accept.  

 
02:58 
Dom Hawes 
If you separate those two things out, then I think that it probably provides a bit of stability to plan those things that probably are going to be secure and stable, like government being one of them. Like once we're through the election, unless something dreadful happens, we're pretty much guaranteed four to five years of government. Therefore we can pretty much forecast the type of policies that might come out of that government and we're through that risk. The economy, different issue. So we're technically in recession. I'm not sure that actually makes a difference because it felt like we have been for the last year anyway, but it always knocks confidence. What advice would you give to marketers about the next twelve to 18 months, given where we are economically?  

 
03:38 
Michelle Booth 
I think it's the impact on consumers. I mean, the word recession is loaded and the impact that makes in terms of people thinking more long term and also just the issues around cost of living and people just struggling to stand still. I think the opportunity is to think editorially. So being really clear on our anchor points, and I think that's a really important thing. Not everything is agile. You need to be really clear in terms of what you stand for, which all good magazines do, but also themes and issues in where you can play to support people as they sort of navigate the financial situation. And also it comes back to share of wallet. Ultimately, it's really making sure that your product and your positioning is fit for some of the needs that people are facing right now.  

 
04:28 
Michelle Booth 
And also as people make decisions about what to buy and what not to buy, you can really position the value exchange that you offer.  

 
04:36 
Dom Hawes 
I hadn't heard that think editorially before, but my brain's thinking about it and I love it. It's a really powerful way of structuring different parts of your activity. In business, you've got a news section, you've got an opinion piece, you've got your articles. And if you structure your own future thinking editorial. I hadn't thought about that before.  

 
04:56 
Michelle Booth 
That's how I think about brand. I see it as a sort of living brand book and I think that when you unpack how really good editorial brands work, they so understand their audience and yet they can flex and play with lots of different other subjects, but they know their position on it. I worked a lot with Cosmo as an example and they understood their audience so well. I think that's the thing, is that flex around sort of the messy middle in terms of product consideration is it allows you to be able to have more than just matching luggage, but really be clear in terms of themes that you can talk about and also the places that you don't go.  

 
05:34 
Michelle Booth 
And I think in a world of social media, it really helps you in terms of understanding how your brand acts and behaves in consumers lives. I'm a really firm believer that strategy is what you don't do, and yet it's the hardest thing, especially when you're given lots of juicy opportunities. And I think that saying no, I think we often think if we throw enough mud at the wall, some of it will stick. Whereas actually the most important thing is working out which arrows you're going to fire and committing to them.  

 
06:01 
Dom Hawes 
Thinking editorially is going to be one of my big takeaways from today. We also are living in a time when sociologically there's quite a lot of change. The post COVID working model, in my opinion, is changing a lot at the moment and is a threat to a number of different areas of business. How are you thinking about that kind of hybrid post COVID situation we have?  

 
06:24 
Michelle Booth 
I still think especially for what I do, it is really important to get real people around a real table, especially when you're working in a sprint model, because the challenge when you work remotely is you get into PowerPoint and you get into things that feel more finished than they are. So I think it's really about structuring those interaction points and making sure that the time that you are in the office is workshops, is kickarounds, is collaboration, but then there is that time for deep work that is often done best at home. But I feel like now modern offices need to be structured around that collaborative space and that collaborative time. Whether it's postit notes up on the wall or whether it's getting everyone in for a workshop, I think that's the most important thing.  

07:15 
Michelle Booth 
Because even when we work remotely, and I've tried to do a lot of work around innovation remotely, it's not the same and you don't have the same connection as you do when it's kind of face to face in that environment. So I think, as ever, it's a clever blend between the two, and we work in teams and very successfully. But the power of bringing people together, the power of making those connections and the power of building those relationships is still needed, and often that helps face to face.  

 
07:44 
Dom Hawes 
I'm a fan of all of those things for our own teams, although our hybrid teams, that sort of client agency relationship thing, too. I think a better hybrid agencies to think about might be to spend more time in their client offices and less time in their own. But also I think that post COVID model has made people realize that if someone can spend two to three days at home and work from there, if an agency's people can spend two to three days work at home, their reliance on agency or their reliance on their suppliers is probably diminished. So I'm starting to see people look at models that offer them better value, offshoring, automation, that kind of stuff, some of the more routine tasks, and I think we're going to come on and talk about technology very quickly.  

 
08:27 
Dom Hawes 
What I'm seeing is client reliance on agency at the moment diminishing for all things other than kind of strategy, creative ideas. It's the big stuff, the little stuff. People are either in housing, outsourcing, offshoring, or finding other ways of automating.  

 
08:44 
Michelle Booth 
I think that's fair. And the truth of it is it can often be quicker, it can often be more effective, but also it enables us to work with agencies in terms of where they're brilliant, which is bringing that fresh perspective, bringing that challenge, and the sort of day to day operational work. If we really are going to run an end to end always on, we need that pace of being able to change content quickly and effectively. And to be fair, the best way to do that is either internally or.  

 
09:13 
Dom Hawes 
Sort offshoring totally, and also having some consistency in tone of voice and having people who just get the brand, as you say, if you need to be responsive and fast. I totally think in housing or offshoring, if it's kind of your own facility is the best way of doing that. Also, actually, not many agencies really like that, like the low value, high volume stuff. It's messy and it's very risky.  

 
09:36 
Michelle Booth 
I think when I've worked with Kraft as part of a team, you spend all day talking about the customer and especially in financial services, finding ways to frame things in a way that is compliant and that working with risk to talk about it. Most copywriters and agencies don't want to spend that time at the sort of less sexy end looking at it. But also, you just spend all your time thinking about the same problem over and over again. And I think that it's very hard to transfer that knowledge onto an agency in day one. So I think some of that nuance and some of that detail really does come by spending all day, every day, thinking and working on it.  

 
10:17 
Dom Hawes 
And the economics of it don't work, frankly, for the production stuff, for the volume stuff. It doesn't pay to have an agency to do it. Technology, we love AI, but it's going to change. I think it's going to change the way we work. Lots of people talking about it, because Gen AI, particularly chat GPT and know Dolly and those things. The easy way to think is, oh, we're not going to need creatives anymore. But of course we are. But I do think it's going to have a dramatic impact on us. Are you playing? Are you looking? Are you thinking, like, how this might play a part in your future?  

 
10:48 
Michelle Booth 
I'm always curious. So one of my favorite things to do is to be on the mid journey discord group chat. And the reason being is more for the consumer insight. It is basically a wish machine, often boys looking at designing really beautiful women. But I think mid journey is a really good example in terms of how people are trying to articulate what's in their heads. I think it's a great tool and I think what it will do is take some of the bad creativity out of the way. It's like all art. There is stuff right at the top that we all aspire to. And then there's the generative, then there's the image libraries. Exactly.  

 
11:31 
Dom Hawes 
And then there's clip art.  

 
11:32 
Michelle Booth 
Exactly. And so if it gives us a better quality of clip art, I'm all for it. But having said that, some of the questions around prompts, in terms of what we're asking, what we're looking for, I think there's still a skill in terms of really understanding that. And it's really interesting when you do play around with it, what's in your mind versus what comes out is often very different. And I always find that quite humbling in terms of agency briefing, because as we've all written briefs, it's actually a great example of that challenge. And I think agencies do a far better job of translating client briefs than I think AI do currently.  

 
12:10 
Dom Hawes 
Prompting is an interesting one because it's natural language. People think, oh, I can prompt, because I can speak English or I speak whatever language it is. But of course, prompting is a lot more sophisticated than that. And those, the people that do run sophisticated, we use some quite sophisticated prompt routines on this podcast, for example, to analyze the different sections of the show, to look at things that are surprising and or interesting, and then we create some content based on them. But I think that skill in prompting, I don't think, certainly I'm not seeing it, that's something that's being taught widely in marketing teams, and I think.  

 
12:40 
Michelle Booth 
It should be not yet. One of the great things when you work alongside execution is it gives you great ideas. And I think AI is exactly like that. If you start to play around with it and you see things that you didn't necessarily imagine. Back to my point of sort of execution is strategy, and strategy is execution. It can take you to new places in terms of your thinking. So I see it as an amazing tool for actually driving human creativity and human innovation, as a prompt, as a kickaround, as an idea generation tool. So I don't think it's something that we should be afraid of. And I do think it's interesting how that you can start to see with AI that sort of tone and look and sort of style is quite noticeable.  

 
13:22 
Michelle Booth 
So I think what's going to become really interesting is as humans start to subvert that, and the sort of diy aesthetic that might come out will also be really interesting.  

 
13:31 
Dom Hawes 
Some of the walled garden AIs that are coming out now using the large language models, but then training on ring fence data, so that instead of writing like the generic Internet, writes in the ever changing world of dot dot, which actually, someone pointed out to me the other day, the reason that chat GPT is crap at writing is because people are crap at writing, because it's just been trained on the Internet. So there are real people who write like that. And if you hire them, fire them, because they now look like chat GPT.  

 
14:00 
Michelle Booth 
They usually do posts on LinkedIn.  

 
14:01 
Dom Hawes 
Exactly. Most of them do. But yeah, they're humbled. They never use personal pronouns, I've noticed. Humbled by anyway. So I think what's really interesting about AI for me also is internationalization as marketers. So because the way that they build the language models is I'm doing an AI course at the moment, which is the only reason I know this, I'm really interested in what sits behind. And so they understand idiom. They understand not just idiom in English, but they understand an equivalent idiom in Chinese. And so if you've played, I don't know if you've played with eleven labs, which is voice, it will do you a simultaneous dubbing, like I say, of course, simultaneous translation. It's more like a dubbing.  

 
14:40 
Dom Hawes 
So you can produce a video, go on there, upload in English, turn it into 32 different languages, upload it to YouTube, and people can then choose what language they view in. And if in there you have an idiom, and the one I tested yesterday was a stitch in time saves nine, it will translate that into the equivalent idiom in Chinese. Because the way the memory is built and the way the algorithm is built, it's not doing a straight translation of word. It understands the context of what you're trying to say. That's mind blowing.  

 
15:09 
Michelle Booth 
That is so cool. I've now got something to go and play with as I go.  

 
15:15 
Dom Hawes 
Back now, I'm going to take the last five minutes that we've talked about on this and I'm going to turn it into Chinese. And at the end of this podcast, after we've said goodbye, I'm going to play it. So if you want to stick around and listen to the Chinese, maybe not five minutes, that might be a bit long, but I'll do some language hops and people will see how good it is.  

 
15:32 
Michelle Booth 
I think the final point on AI for me though, is also the ethics piece, and obviously the whole of financial services is looking at AI. But I'm mindful of a lot of the decisioning when things become a black box in terms of how you make decisions, whether it be like sort of lending criteria or whether it be whatever. The issue of ethics is a critical one, and I know it's something that the industry is looking at very robustly. But more broadly, back to your point about the reason why GTP doesn't write very well is down to the people that do the writing. I think we need to be really careful about the bias that we embed in that. And I think it's very easy to say ethics, but there isn't a clear, definitive view of what ethics mean.  

 
16:17 
Michelle Booth 
And I think that piece is important. And obviously the pace of change is going extremely fast. But I think what we need to be really careful of in financial services, and I know that we are, and that's one of the brilliant things about being a risk based discipline, is that we need to make sure that the decisions that we make have no bias in them and that they are fair for all.  

 
16:38 
Dom Hawes 
That's a really good point. I had seen the leap from I always talk in generative models, because it's easy to think about that. But from chat GPT into, for example, Azure GPT, where you can train it on your own set of data, but there's going to be bias in our set of data, whether we like it or not, particularly if you're going back beyond 1015 years. And then how you police that is a really interesting challenge. So the large language models, for example, at the moment they have rules in place that will help them police that. But if you're doing something yourself on your own data, how do you spot bias? I mean, it's an interesting one. Answers on a postcard, and there'll be a cheap and crappy prize to the winner selected at random. There you go. I hope you enjoyed our conversation.  

 
17:22 
Dom Hawes 
If you did, please follow us on your pod platform rate, review us on Apple, or maybe recommend us to a friend, because every little helps. That is it for this extra free bonus edition. See ya. You've been listening to Unicorny, and I am your host, Dom Hawes. Nicola Fairley is the series producer, Laura Taylor McAllister is the production assistant, Pete Allen is the editor. Unicorny is a Selby Anderson production.