Tapping Into the Customer Mindset 

Generative AI is EVERYWHERE right now. Marketers are scrambling to figure out how they can add it to their marketing strategies, but what lots of brands aren’t yet considering is how to use generative AI to better understand the customer mindset. 

We can’t predict exactly what our customers are thinking or feeling at any given time, but thanks to data, we can get a lot closer to getting it right than ever before. In my role as Phrasee’s Director of Client Strategy, I work closely with brands to help them make smarter decisions about how to use the wealth of data available to marketers today, particularly when it comes to better understanding their customers and making sure they receive the right messages at the right time. 

So how can marketers leverage generative AI to more effectively tap into the customer mindset? 

What works and what doesn’t, right now

Our new research into what makes customers tick shows the extent to which language can make a difference to customer engagement and response rates. Different words, emojis, and sentiments resonate differently across brands, audiences, and at each stage of the customer lifecycle, and this changes all the time. It’s easy to see how marketers can feel bamboozled by the data and not know where to start, but Phrasee’s 2023 data shows what’s resonating with real customers right now. 

For instance, customer-centricity was found to be the most successful tactic across all channels (such as email, push, and SMS). Specifically, the phrase “thank you” appeared frequently in high-performing marketing messages, alongside “you’ll like” and “you’ll love,” showing that it pays to tailor your marketing to your customers. What’s not working so well is urgent messaging, which pressures customers to take action, and surprisingly, our research into offer messaging found that “50% off” excites customers more than “70% off.”  

Some of this goes against what we might expect, which highlights the challenge faced by marketers today – without data, how can we be sure that we’re not making decisions based on our own gut instinct or natural biases? 

Getting started with experiments for language optimization

In my role, I speak to a lot of clients who are just starting out with language optimization experiments.  

The ultimate practical advice I can give to brands is to follow the scientific method. Start with the hypothesis or key question to investigate, then as you make observations and collect data, you’ll have a “guiding light” to help you cut through the noise of the data and stay focused. This means starting with a testable statement with a clear outcome – something as simple as “longer subject lines will reduce engagement,” or “adding relevant emojis will increase open rates.” 

The next step is gathering the data. That means running a series of tests that will allow you to measure the impact of the changes in your statement. You’ll also need to be mindful of the impact of factors such as the audience you’re speaking to or the time of day. For most marketers, this is the hardest task! You’ll need to do quite a lot of testing to ensure you’re getting fair, meaningful results. 

You’ll then need to analyze the data by looking for patterns that demonstrate connections between important variables. When you overlay language optimization with things like templates, content, and personalization testing, there are a lot of moving parts to keep track of. It’s no easy task, and you’ll need to be mindful that there’s a lot going on in your data, so having a robust test-and-learn strategy is key to your success. 

Leveraging the customer mindset for better marketing

I’m a huge advocate for getting a better understanding of the customer mindset. I always go back to a bricks-and-mortar example:  

Say you’re in a department store looking at electronics. There’s another version of you in an alternate universe doing exactly the same thing. A store associate approaches you in both universes. In one universe, the associate says “Can I help you? What are you looking for today? What do you need it for?”. In the other universe, the associate says “We have 25% off menswear!”. Which is going to be more effective? 

Bringing a conversation element to customers and really tapping into what they’re looking for is key to driving engagement. Brands have access to more customer data now than ever before, and there’s a certain element of trust that you’re going to do the right thing with that data. Marketers need to make good use of that data and provide value back to the customers. In exchange for this added value, you will get more data from your customers, which you can use to make better-informed marketing decisions.  

Let’s consider the customer lifecycle stages. A prospect who has just signed up for a brand’s emails is in a completely different mindset to an existing customer who is already actively engaged and considering their next purchase. Alternatively, you may have a customer who recently purchased a product, but may not be ready to purchase again for some time, or a lapsed customer who has become less engaged in your marketing over time. Different types of messages will suit customers at different stages of the lifecycle, so understanding their motivators, pain points, and mindsets at each stage will empower marketers to get smarter about the messages they send. 

Doing more with less

Language optimization plays a huge role in driving customer engagement, and the impact differs hugely between brands. There’s an expectation that you’ll be talking to customers in a certain way, for instance, based on the brand’s vertical, and ensuring that this brand voice stays consistent across all your channels, such as email, push, and SMS, is essential to fulfilling a customer’s expectations. 

That’s why there’s no better time to be leveraging AI content optimization tools like Phrasee. We’re on the dawn of something very special with generative AI, and as a company that’s been in this space for eight years now, we’re very well positioned to support brands in their optimization journeys with language. 

From a brand perspective, there’s a lot of pressure to do more with less… but there is no one perfect solution out there. As a marketer within a brand, you should be reassured that not everyone has everything right – there are a lot of gaps in a lot of marketing strategies, and part of my new role as Director of Client Strategy (EMEA) is to help brands remove those blockers and help them on their test and learn journey. 

And finally … a note on ChatGPT

As our Chief Data Scientist Neil Yager recently wrote, the hype around ChatGPT has put generative AI, and particularly its applications for language, at the forefront of everyone’s minds – even my mother-in-law was talking about it over the holidays! It’s great that there’s so much interest in ChatGPT, and we’ve already reached the point where it’s no longer about the “when,” but about the “how.” There are brands all over the world leveraging using generative AI to optimize and refine content, and it’s one of the biggest marketing trends I’m seeing for 2023. 

The “how” of ChatGPT will be around creating clear business cases to show how it will add value to each brand. A key focus will be getting stakeholders across your brand on board – especially as this is a relatively new area in marketing, and a lot of people fear significant technological change, so knowing how to deal with pushback is vital.  

In particular, there’s a lot of concern that ChatGPT is going to take marketers’ jobs, which won’t be the case! Instead, it should be seen as a tool to further advancements in customer messaging. If you’re not embracing generative AI in your marketing now, then your competitors certainly will be. 

One thing to note is that if you’re only experimenting with tools that generate content (like ChatGPT), you’re missing a trick. Tools that also optimize your content (like Phrasee!) are where you will see real uplifts in marketing performance, whether that’s top-level metrics such as increased site traffic, or down-funnel KPIs such as conversions and repeat purchases. That’s something you can’t get from a content generator alone – you need that consistent feedback loop of learning, understanding, and optimization that feeds back into the generation for the best results. There’s only a limited set of tools that can do that right now, and luckily, Phrasee is one of them!