Why Good Creative is Good for Business

It’s often been claimed that digital growth is killing creativity, with data-driven advertising trumping quality creative.

With powerful targeting and ever-increasing scale across the likes of Meta and Google, and brands vying for consumer attention in an increasingly competitive landscape, with shorter and smaller ad units, it’s easy to see why creative is often overlooked. 

This is something that’s repeatedly confirmed here at Brave Bison when we audit accounts, with one of the most persistent bits of feedback being a lack of synergy between paid media creative and performance insight. We know from experience that the greatest outcome is achieved when creative, technology and media work in unison to create a continuous feedback loop - ensuring a steadfast strategy that speaks to the right audience for your brand, at every stage of the customer journey. 

Here, some of our resident experts in Data, Social Strategy, Paid Media and Measurement will equip you with the knowledge of how to effectively engineer digital ad creative for your audience and, more importantly, prove its impact on performance. 
*Psst! If you prefer to watch or listen, you can tune into the video format of this article, here.*

Let’s start with this incredible statistic: 50% of campaign performance is driven by creative. Mark, can you kick us off by telling us a bit more about where this data comes from, and what it means in real terms for how we should be approaching campaign creative? 

Mark Byrne, Head of Paid Search & Social:

A few years ago comScore did a study on what drives effectiveness when it comes to generating sales: pricing, promotion, media planning, and so on. Creative comes out on top, accounting for 50%.

This highlights the importance of what can sometimes be either an afterthought or developed without the right care and attention – particularly for those of us working in Paid Media teams. 

You can have the most sophisticated, modern campaign structure, automated bidding, targeting high-intent audiences, but if your creative isn’t up to scratch then all of this falls flat. 

Consumers are increasingly savvy about what an ad is, and if you don’t hit the mark they can be cynical too - stock images don’t cut the mustard. We need to focus on authenticity. 

To ensure what we’re getting the most from our creative, we use a Google framework that we’ve incorporated into what we do when developing creative or working with our client creative teams. 

This framework is called the ABCD of Creative and there are four main pillars to keep in mind: 

Attract - from the start we need to attract attention - short videos and fast edits at the top of the funnel. 

Brand - making sure the brand is integrated meaningfully, either using graphical overlays or having the product in situ. To be effective, we need to include the brand within the first 3 seconds – and with one report suggesting that social media users scroll the equivalent of the height of Big Ben with their thumbs each day, it’s essential we connect and stand out! 

Connect - using emotional storytelling to resonate with our audiences - what do we understand about our audiences and what do they care about? 

Direct - the final pillar involves us making sure we’re aligning our messaging with the right stage of the funnel, using the right CTA and importantly repeating that CTA. 

These are relatively straightforward principles of storytelling. We need to remind ourselves of them during the campaign planning process and be aware of their impact on the effectiveness of creative - going back to that 50% number at the start. 
There’s a lot to think about there, but one thing that really stands out is the importance of understanding and engaging with your target customer, which presumably means we need to know who that audience is. Is it fair to say that this is a crucial foundation for creative planning, and if so how do we go about building this picture? Nav, can you elaborate? 

Navya Hebbar, Data Strategist: 

You're absolutely right – audience understanding is key. And usually, when we think of audiences, we think of segmentation. Before we dive into it, I want to touch on three thinking traps or misconceptions to avoid during your paid media audience research.

We tend to view audience behaviour as deterministic. Those in your segment will buy your product, those outside won't. But reality is a lot more probabilistic. There are a number of attitudes that affect audience buying behaviour outside of just demographics. This means that, in addition to intent-based quantitative data, we need to look at social listening-fuelled qualitative data. We need to aim to understand the attitudes that fuel the audience's behaviour. 

We tend to believe that we're the ones deciding the audience segments. But, while you define them, your audience is segmenting your brand. They decide if a car is a sports car. They decide if your beauty line is a premium luxury product. What you want to be doing through your research and data collection process is to look for clues on how to evoke the right emotions through your creative, especially at the top of the funnel - to align your view of the audience with their view of your brand. 

We tend to think of segments as homogenous. But it's not as simple as "young mums" being our segment and "older businessmen" being yours. Quantitative data tends to flatten people, but that's not a real representation of the rich mix of attitudes those people hold. They might be a mum of two, but they may also be the founder of three start-ups. Acknowledging this is important.  

Keeping all of this in mind when you get into the actuals of research methodology, it ultimately boils down to two important questions: What data do you use? And how do you interpret it? 

If you're trying to get a strategic, deep understanding, then I would encourage everyone to become "data omnivores." This means letting go, temporarily, of analytical mindsets around statistical significance or the rigour of measurements. It's about embracing data from everywhere - social listening, surveys, a bit of googling, Reddit threads, or even Mumsnet. 

What this leads you to is a holistic and sophisticated framework. And once you've populated it with data, it's about making the techniques work for you - you can use a number of approaches here. We can analyse conversations, broader concepts and narratives, sentiments or content formats-whatever suits your brand needs. 

Here, I'm pulling out an example of how something like this might play out. We're looking at a fashion brand and what its audience might resonate with.

We used Meltwater, a social listening tool, to analyse the social conversations of the audience and cluster them. When we did that, we started to get a picture of audience personas and how to talk to them. 

Take, for example, the section in the middle: Vintage Aficionados. They're a segment of the broader audience interested in all things vintage and thrift shop related. As I mentioned earlier, this is only probabilistic - there will be a wide range of topics or themes to choose from. 

But identifying and zeroing in on a couple of personas means that we can draw out our reasoning into the creative space. Here, the vintage angle means that our creative can start speaking to rare, unique, or heritage aspects of our brand in a way that relates to the vintage aesthetic of the audience - and there's a higher probability that they'll pay attention to it than a stock image. 

From there, it's about going and testing the messaging across your funnels. This logic is foundational for both understanding audiences and setting up an approach where you can experiment, test, and iterate on your paid media creative. I hope that answers your question! 

Some really interesting stuff there, and a clear run down of the robust logic that can be followed to understanding a brand’s audience, and what makes them tick - a sometimes-overlooked first step when it comes to thinking about digital creative curation. Going back to Mark’s four pillars, the logical next step is to figure out how we attract and connect with them. Dani, if we think about overall best practice, is this where the content itself comes in? How do we maximise our chances of producing content that hits the mark?

Danielle Dullaghan, Social Media Strategist: 

When you have your audience identified, it's time to start to think about creativity and how we're going to lure in our audience.And really, what it comes down to is ‘how are we going to grab the consumer’s attention’? 

It's getting more and more difficult to grab attention on social media. I think this is the biggest challenge for us, as social advertisers - and I don't think this will be news to anyone, but our attention spans have shortened significantly.  

Microsoft ran a study which showed that our attention spans have dropped to eight seconds. As well as this, Meta ran a study that showed 94% of us keep our phone in our hand while we're watching TV. So not only are we as brands competing for ad space and brand recall against each other, but we're also coming up against other forms of media and entertainment. We're always consecutively consuming media at any given time, so it’s getting more challenging to actually grab the attention of potential consumers. 

So,how are we going to gain user's attention at every stage of the funnel? I think we need to give credit to the consumer on this one. They are very clever, and they have caught on to what an ad looks like on social. Over the last 10 years we've been bombarded with social advertising – particularly on platforms like Meta – and really, it's not a great user experience, because when you are logging on to social media, you're not logging on to buy something (unless it is for that specific reason).  

Most of the time, you're logging on just to relax or as an escape, just to watch some content. So it's not a great user experience when you're scrolling and you just get bombarded with ads that are asking you to ‘buy this’, ‘50% off that’ and ‘buy now’, and it's very distracting to your social experience. Users have built up a resistance that kind of messaging and creative and there is a type of creative fatigue that comes with it. They’re able to know within the first few seconds whether it's going to be an ad or not, and they can just scroll by, which isn't great for ad creative and for brands. It's not an effective way to run ads. 

So, what we propose doing is producing ads that feel like native content and not a sales pitch. But what do we mean by native content? 

We've put together a couple of examples of brands who are doing this kind of native advertising approach and doing it well. And what we mean by native is it feels just like you're watching unbranded content, like scrolling on or Tik Tok scrolling on Instagram – it doesn't feel like an ad and it doesn't disrupt the user experience. That's important: if a user or your new potential customers is scrolling and they come across one of your ads, that feels like content, they're going to sit and watch it and they're going to watch it for a little bit longer. 

We've applied these example ads to our ABCD framework that we talked about earlier. There are a couple of different narratives and arcs that you could potentially use if you’re going to go to the native advertising route. 

As an example, we’ve looked at the fashion industry. When you think about fashion brands and the shoots that they produce, they're quite premium and glossy, which feels very advertorial and doesn't fit the native-ness of social.So, what ASOS and Never Fully Dressed have done is use a behind-the-scenes look and feel shot on iPhone to create an ad that does feel very native and, if you were scrolling by this, you'd be more inclined to click and stop.

In the example from Stitch Fix, they have produced a premium yet native creative. It’s a native take on a traditional testimonial, but they've used the fonts that are native to the platform. So again, if you're scrolling, you're not going to automatically see this as being super salesy content and you might be more inclined to click and watch as a form of entertainment. 

UGC (user generated content) is content that is shot by your customer or audience. Or curated content that looks like it is shot by a customer, your audience. Again, it brings the realness and the authenticity that users love to see in content because again, it doesn't feel like you're pushing too hard on a sale.

So, to summarise then the key components to the perfect parts to a native ad: 

Use low-fi video where possible – so that's shooting on iPhone; 

Be inclusive in casting, because your audience will love to see themselves in advertising and they'll be more likely to click on the content if they see somebody like themselves in your ads; 

Use user generated content or products with faces and testimonial;  

Have interesting copy which will help users to stop when they're scrolling; 

Most importantly, make the content feel like content. It's important just to remember that this is not a sales pitch and it is a piece of content. 

So, we’ve used our performance success framework, studied our audience, and we now know – in theory – what makes for the perfect ad. But Kit, the million-dollar question is whether this really drives incremental performance. what processes can we put in place to prove and measure this?

Kit Bienias, Performance Director - Growth 

There are three solutions that you can use to prove the incremental value of your activity.  Our preferences areGeoLift studies and Data Driven Attribution. 

First up is GeoLift– Meta & Google’s solution to incrementality testing.It is the notion of splitting a group of individuals into a test and control group, showing ads to one group and switching off ads for the other group, then measuring the incremental uplift in performance for those who were exposed to ads. 

The reason we like using Geolift is because it removes seasonality challenges, as you can run this test at the same time – rather than having to switch off activity for a month and then back on for another month, which can lead to seasonal inaccuracy and a significant loss of revenue.And the tool does this through finding locations that see a similar level of conversions, which then form the test and control groups.

Then we can measure the incremental uplift on our bottom line through showing ads vs not showing ads. 

This is the gold standard when it comes to determining the true value of your activity because there is no bias involved: you’re literally switching on and off activity, to see whether your channel or campaign can deliver incremental value. 

Next up is Mixed Media Modelling, which is our least favourite solution because it is based on historical data rather than real-time data.But it can be useful if you are running lots of offline activity and want to understand the correlation between your offline and online activity.

How does MMM work? 

You ingest all your historic channel data into a model that has been built by a data science team. 

MMM uses regression analysis to determine the relationship between several metrics to predict business outcomes, a similar approach to how we carry out our forecasting, though the big nuance is that MMM does this at scale using an algorithm and automation. 

So MMM is good for conducting forecasting and looking at the correlations between online and offline activity.  However, the challenge with this model is that it is based on historical data, so it cannot keep up with latest trends and every time the data is ingested, the algorithm must re-learn. That’s why it’s not our first port-of-call when it comes to setting up a measurement framework. 

Finally, another favourite of ours: Data Driven Attribution. With DDA, Campaign Manager gives us visibility of de-duplicated, cross-channel attribution using floodlight tags.Channels can be tagged and ingested into Campaign Manager through impression and click trackers. 

Not only can we report on cross channel performance, but we can also create a DDA conversion goals that can be paired with automated bidding in search. So, the automated bidding will start to consider all assisted touchpoints in the path to conversion rather than solely look at last click last click, helping increase conversions.

So, to finish back where we started: is today’s love of data-driven advertising a death knell for creativity? Clearly, data has a critical role to play in effective online advertising. It’s crucial that we use robust hypotheses, tooling and testing to understand who our brand’s audience really is. We need to stack these insights onto a smart campaign structure that utilises intent-driven targeting and clever automation. And it’s absolutely essential that we use data at the end of the journey, be it any one of Kit’s three suggested measurement frameworks, to understand whether our campaigns are working for us, how we should iterate our strategies, and where we should focus our investment next. 

But all of this is only half the picture – literally! 50% of your advertising success hinges on the creative that you deploy off the back of these insights. It’s the creative that will give your ads a fighting chance of capturing an 8 second attention span, interrupting a Big Ben-height scroll, and fostering a genuine connection with the audience that you’ve worked so hard to suss out.  

The greatest success is to be found in the effective melding of data and creativity – and we hope we’ve given you some helpful ideas about how to do just that.
Curious to see how we can help support with your creative or digital performance strategies? Slide into our DM's.