Episode 33: Here's Where AI Fits into Your Content Strategy | With Natasha Vickery-Orme

Full transcript:

Natasha Vickery-Orme  00:00

There are a lot of people, I think, who might be slightly panicking about it – where they think, “Okay, so these machines can write stuff en masse, they can do it a lot quicker than the humans.” So those teams of writers that you mentioned, are all suddenly panicking that they're then being made obsolete.

But actually, I think it's kind of a shift. Rather than taking out those individuals that have those specialist skill sets, at the end of the day, the actual task of writing is very subjective. I think it's the shift to the idea that “Well, rather than writing it from scratch, this tool can save me time, and so what I can then do is I can enhance it.” Actually, rewriting and modifying and making those changes, at the end of the day, is a much harder task – and that is something that we can't put through an AI machine, no matter how much we train it.

David Bain  00:51

The Strategic Marketing Show is brought to you by Insights For Professionals: providing access to the latest industry insights from trusted brands, all on a customized, tailored experience. Find out more over at InsightsForProfessionals.com.

Hey, it’s David. Content is an essential part of building engagement into your marketing activities. But, it's getting harder to get your content seen by the right audience, and AI might be making this even tougher. So, what does highly engaging content look like? And how is this going to evolve in the future?

That's what we're discussing today with a lady who works with brands including, Indeed Sitecore, and AWS, helping shape their content strategies around online experiences that engage the right audiences. She’s a Content Manager who heads up all the strategic content production at Inbox Insight. A warm welcome to the Strategic Marketing Show, Natasha Vickery-Orme.

Natasha Vickery-Orme  01:51

Hi, it's good to be here.

David Bain  01:53

Good to have you here. Thank you for coming on, Natasha. Well, you can find Natasha over at InboxInsight.com. So, Natasha, how would you describe the purpose of content within a marketing strategy?

Natasha Vickery-Orme  02:07

Content has a bunch of different functions, and I think that's what makes it so difficult to get right. We have one side of it, where we're boosting brand awareness, we're trying to engage audiences and find those prospective customers. But also – and what I think many people forget as well – is that the essential point of content is that we want to establish authority, we want to demonstrate what we know, and we want to build that trust with prospects - or even people who aren't necessarily going to be coming into that funnel. Being able to kind of share that knowledge and establish ourselves in the industry can go a long way to building that further down the line.

It's such a multipurpose format, that it can be really difficult to get it right. For that reason, that's why we look at different content formats, platforms, and channels, and how it all comes together. That, in itself, is the biggest challenge of content.

David Bain  03:03

You talked about having different purposes behind content there, such as awareness, trust, authenticity, and simply being useful as well.

Can one piece of content actually provide all of those things? Or do you have to have different pieces of content honing in on those different areas,

Natasha Vickery-Orme  03:21

You can kind of do both. In theory, I would say you want one piece of content to achieve all of those things. But because of how difficult that is, we do often think about different pieces achieving different milestones.

Whilst you have pieces that are focused on authenticity, and driving your brand values, you can have others that are more focused on your product sales. But also, essentially what we're wanting to do is create an ecosystem where all of the content works harmoniously together in the same place so that you can create that environment for individuals to deep-dive into. So, by going from one piece to the next, they're actually able to get a sense of that from all of our content. It's like creating a really big checklist of all of the things that you kind of want to achieve, but also dividing that out. You can kind of go both ways with it.

Essentially, the really, really good pieces will achieve all of it in one. But to do that, you have to have so much behind you already. That might be ranking really high in SEO. The people that you see at the top that are already doing all of this, essentially, their single pieces can convey all that. But, at the end of the day, they've already established their brand; they've already built that trust and that awareness. Trying to catch up to that, sometimes it is simpler to just focus on one piece at a time. Having said that, you do have to be careful not to neglect a single area – and that is where it becomes really, really hard to try and see if you can get that balance.

David Bain  04:49

I love your mention of content working together. “Creating an ecosystem” was the word you use. That reminded me of Joseph Khan actually, on a previous episode of The Strategic Marketing Show, when he talked about how content should almost form a melody together. It should be a musical and be harmonious, as you say.

What sequence, then, do you actually tend to suggest content is created and published in? Obviously, one piece of content can be published in many different areas nowadays.

Natasha Vickery-Orme  05:24

When we're crafting our content, we take a step back and think about the Awareness, Interest, Desire funnel. We are thinking about how we can appeal to different individuals, depending on where they are at different stages. Now, whether that's introducing a topic to them, whether that is going slightly deep into that topic and providing more depth to it or going even further and providing the solutions.

Thinking about the content in this way allows us to map out a journey where we can go from, “Okay, so this is where we start with a topic, this is how we go into more detail about it, and then this is actually how we solve that problem.” Doing it in that order means that we can create quite a linear channel, and linear journey, for that individual. It doesn't always work like that, obviously, because we can't predict exactly what it is that people are going to interact with – but it means that we have spaces where that is possible. So, should somebody want to go further into a specific topic, we have those avenues available. If they decide to switch topics, we have that as well.

Then we can branch out. Then that's where we include our cluster areas. You can jump around different topics, or should I say different subject areas within one topic, and each of those links into each other. That, again, comes back to the ecosystem idea. Because we're not all linear (because that linear idea is a little bit outdated, it's not necessarily achievable these days) creating that environment where yes, okay, we still have the Awareness, Interest, Desire structure, but we can still be quite flexible with that, and provide that in a variety of different ways.

David Bain  07:07

I love the fact that you do put the customer journey at the heart of it, but you also mentioned that it's not linear because not every customer journey is the same, but, you need some kind of initial point to start from, certainly, as well.

Do you have a favored marketing channel traffic driver for different stages of the journey?

Natasha Vickery-Orme  07:25

Yes, in a way. For us, we focus entirely on the content production side of things, so the channel distribution side of it is not necessarily what sits within my team. When we're producing our content, we're trying to think about how we serve that individual on the page, in the IFP environment, and how exactly that looks from the user perspective.

But when it does come to the different channels, we are always trying to think about how we can serve them in the sense of – so we have our email channels, how does this title differ from on-page? How could it work in email? How does that then translate into the subject line? Or, if we go down the route of SEO, How do we compete with those already ranking in that space?

When it comes to matching up the different stages against the channels, it depends on the campaign that we're shaping, and how the different teams are coordinating that between us.

David Bain  08:24

One massive change that's happening at the moment, in terms of the creation of content, is the involvement of AI.

So, what is a good use of AI in creating content at the moment? And how is it impacting content creation?

Natasha Vickery-Orme  08:40

Yeah, it's such a good topic at the moment. You're seeing a lot of things pop up on places like LinkedIn, where individuals are showcasing how they're using AI to create social content, how they're using it to help them create eBooks, and there's a plethora of examples now out there of how individuals are using them to fuel their content creation.

We've been dabbling in this and understanding how we could use AI and how we can leverage it in our production pipeline; what that looks like, what it's capable of, what it can't do, and how we can shape that to help to supercharge what it is that we're offering. It's been really interesting. The tools are amazing. We've come so far compared to a couple of years ago. What we're actually able to do now is create entire articles from scratch really effectively and really efficiently.

Yesterday, for example, I was working on a piece where I wanted to talk about general management topics, so nothing too specific. It was more of a general area. And I could put into this AI tool a few headings, a few bits of notes about each section I want to talk about, and then just hit “Generate”. Within like 30 seconds, this whole article has appeared – and then reading through it, it's actually not that bad. In the space of 20 minutes – so we've done our research, we’ve come up with our title, we've been able to construct an outline – we have a fully-fledged article that is workable. It has managed to cut down that pipeline massively, where we would work with our writers, we would brief it out to them, then they would produce it and bring it back to us, which could take weeks. We've been able to cut it down to like half an hour, which is amazing.

That being said, it's definitely not finished. We have snags in the system where they can't produce statistics. The AI tools aren't able to generate relevant statistics and URLs that we can access that are live links. We can't access general resources that can enhance the article. We try and always include sources where people can go and learn more, and see (if we're talking about current topics) who else is talking about it. And the tone of voice isn't quite right, it doesn't match us. That's to be expected; nothing's ever gonna match our exact tone of voice without a human going in and fixing it. But, it is kind of mind-blowing exactly what it can achieve in such a short space of time, which I think is amazing.

David Bain  11:12

You mentioned tone of voice there as well and, obviously, tone of voice is key for most big brands out there – they have a very defined way of talking and they have a particular persona that they use.

Is that something that you've tried to work on? Have you tried to key in specific examples for AI or specific things for them to think about prior to actually producing the text? Or have you just thought, “Well, it's not going to be possible for them to come up with the tone that we're looking for, so it's only something that, post-production, we're gonna have to focus on”?

Natasha Vickery-Orme  11:41

It's a tough one. We do have the capability to fine-tune exactly what kind of tone of voice we get back from it. We have an array of different things that we can choose from, and we can teach the AI model that these are the things that we like to see, and this is the tone that we're going for. But realistically, I think, without custom building an entire tool from scratch, and training over 1000s and 1000s of articles, I'm not entirely sure if we're going to get there with it. I don't think we're going to get to a point, at least not anytime soon, where we have a tool that can just auto-create all this content without the need for us to go in and fix it afterward.

What it does produce is something we can work with. We can specify, “Okay, we're not wanting a playful tone of voice, we want something that's a little bit more professional, but is more engaging.” We can get quite a way there, and then it's just a case of, we only need to go in and fine-tune. A lot of it, as well, comes down to nuances that we prefer. When you look at our style sheet, it’s the ways that we prefer to speak and how we prefer to shape that tone. That probably could be fed into the AI machine, but then there's also that sense of, a lot of the time, it doesn't necessarily understand the context. It can get confused, when you put a topic in, and what it spits back out again is maybe not B2B focused, because it doesn't know that you're talking in a B2B context.

There's a lot of that fine-tuning that is involved in both the prep work, and also the aftermath. I think we could get there eventually but, at the moment, that tools don't really have that capability. It all comes down to data. There's not enough data being fed through it yet. I don't know, maybe ask me that question in six months, and it might be a completely different answer.

David Bain  13:31

Yeah, I mean, you say six months – AI has certainly come through a massive step change in the last six months and I would imagine that, over the next six months, it's possibly going to be even bigger.

So, what does that mean, practically, for marketing teams? I’m thinking specifically of teams of writers, content producers, and perhaps big marketing departments that have teams internally as well. What should they be doing at the moment to take advantage of AI and to be able to plan what they're doing with their teams to be able to take full advantage of it in the future?

Natasha Vickery-Orme  14:03

There are a lot of people, I think, who might be slightly panicking about it – where they think, “Okay, so these machines can write stuff en masse, they can do it a lot quicker than the humans.” So those teams of writers that you mentioned, are all suddenly panicking that they're then being made obsolete.

But actually, I think it's kind of a shift. Rather than taking out those individuals that have those specialist skill sets, at the end of the day, the actual task of writing is very subjective. I think it's the shift to the idea that “Well, rather than writing it from scratch, this tool can save me time, and so what I can then do is I can enhance it.” Actually, rewriting and modifying and making those changes, at the end of the day, is a much harder task – and that is something that we can't put through an AI machine, no matter how much we train it.

For marketing teams that are looking to leverage it now, we can start simple with things like social posts; creating social posts for LinkedIn or Twitter or Facebook, or whatever platform people are using. That is the simplest way to start. We can get the AI machine to generate, say, five different posts based on a subject or URL link, and then fine-tune it. As we get more confident with how it works – in the case of okay, how can we roll it out? Can we use it to create email subject copy? Can we use it to create introductions to articles?

We started with actually using it as a paragraph generator. We were looking to enhance content we already had live on the site, that we knew was performing well. We wanted to improve it and update it, so we were reviewing it and seeing, “Okay, where are there gaps in this piece? What could we add into it?”, and then using the AI tool to help us create that, so that it was a quicker way for us to improve that content. If you look at the site, we have like 4000 articles currently live. It's a lot of content for us to manage. Finding ways where we can fast-track that production is always beneficial.

It's a case of starting small and not feeling overwhelmed, because it can sometimes feel like it's a bit much to take on – this tool that can suddenly do all these things that we've relied on people for. But then, at the same time, I think marketing teams aren't shy of using martech. We've been using things like CMSs and automating our customer relationship tools for quite a long time now. Introducing new technology into that mix, if anything, I think the marketing departments are probably going to be the first people to jump on this, really, because we're so used to including that new technology to find shortcuts and to find ways that we can do stuff quicker and faster and more efficiently while still having the same output.

If we think back 10 years ago – when there was no technology available for this sort of thing and all our emails were blanket newsletters, and there was no personalization, there was no segmentation, and stuff like that – we’ve come a really, really long way. It'll be really interesting to see how that continues in the next few months. It's mad to think what the end of the year might look like, compared to how we've started the year. I think it’s got really good, exciting things on the horizon.

David Bain  17:09

Absolutely. Mad, and perhaps scary to a certain degree as well. But you have to jump on the bandwagon. You at least have to be aware of what's going on and how it may impact the way that content is created. One other form of content that appears to be being impacted massively is images. Something that I'm hearing quite a bit about is something called Midjourney, which is an artificially intelligent way to generate images.

Are you having conversations with many brands about using AI to generate images within their content as well?

Natasha Vickery-Orme  17:43

It’s not something that we've explored just yet. I've been observing a few conversations in different places, and I've been witnessing certain issues around copyright infringement laws, and things like that, that are being flagged. It's not something we've looked into at the moment.

Having said that, we are very particular about the type of images that we use, in the sense that we try and avoid very stereotypical images and try and go down the more emotive, surprising route. When you look at the IFP website, for example, we are focusing on images that you wouldn't necessarily immediately associate with that content – that make you go, “Oh, I get it! That's quite fun; it's caught my eye, it's different.” For that reason, I don't think that we're going to be in a position to use AI to achieve that quite yet. We're using that human element to think outside the box and think, “Okay, this article is about AI. But actually, we're going to add an image in that is not robots, it's not a factory, and it's not the stereotypical things that you would expect to associate with that title.”

The copyright infringement stuff has been really interesting to observe, and I'm not entirely convinced that they've managed to sort themselves out with this side of things yet. I'll be interested to see how that progresses.

David Bain  19:08

Yes, you have to certainly be aware of where AI is likely to be taking the content from and whether or not you actually have permission to take that content. It’s a story yet to continue, for many parts, I'm sure, in future episodes.

Let's move on and ask you, not necessarily thinking about what we've been talking about so far, what is the number one thing that marketers need to incorporate into their strategy?

Natasha Vickery-Orme  19:35

Okay, so if we're not going to use the obvious AI, I think we need to be thinking about new approaches. One thing that we've done this year with our strategy, particularly, is we have changed how we've thought about it. We've taken last year’s strategy, and we've kind of flipped it on his head so that this year we can think, “Okay, so that worked that last year. What are we going to achieve this year? How do we want to achieve it? And how can we do that differently?”

It's taking what we know, but then deconstructing it and building it back up into something different. I think it's really important for marketers to not be afraid to try something new and to tackle that new approach. Yes, it's worked in the past, and we can take elements of that. But actually, AI is a great example of how we can leverage new tools and new ways of working to really broaden our horizons.

At the end of the day, it takes a tiny little bit of a leap of faith to get there. I think this year it is all going to be about those new approaches and finding new ways to achieve what we're trying to do.

David Bain  20:37

So, include experimentation in your strategy.

Natasha Vickery-Orme  20:40

Yeah, 100%. We do an awful lot of it within our team. It's a core element. We have designated space within our strategy to experiment with different styles, topics, formats, and different ways of doing things – and all the time. Every year, it allows us to grow at quite a rapid rate. Without it, we wouldn't be where we are, basically.

David Bain  21:01

I've been your host, David Bain. You can find a Natasha Vickery-Orme over at InboxInsight.com. Natasha, thanks so much for being on the Strategic Marketing Show.

Natasha Vickery-Orme  21:10

Thank you for having me.

David Bain  21:12

And thank you for listening. Here at IFP, our goal is simple: to connect you with the most relevant information, to help solve your business problems, all in one place. InsightsForProfessionals.com.