Do you know why your content isn’t working? Semantic content analysis will tell you.

May 20, 2020

We know that exceptional content is what makes a brand. We also know that getting as audience-specific as possible when analysing data is crucial for great ROI. 

But we rarely put the two together. 

Despite the clear link between the two, it’s rare that we marketers use the data available to actually analyse the kind of content that’s working for us. If we’re aware of what’s working, we’re often a bit hazy on just why it’s performing so well.

It’s not just great content which gives us that winning edge. It’s also knowing exactly why that content resonates so well with our audiences. Luckily, the means to find out exactly that is already in our hands.

The ‘Data Boom’ (and what it means for marketing)

In the climate of the current ‘data boom’, audience targeting naturally takes precedence. According to  Econsultancy, 55% (i.e. the majority) of marketers said that ‘better use of data’ for audience targeting was their priority for 2019.

It makes sense. On a daily basis, marketers are bombarded with countless blogs, podcasts, speakers and everything in-between promising that if we perfectly optimise our targeting, our messaging will beat those daunting 0.9% CTR odds cited by Wordstream. 

And so, we dedicate hours and hours every week to creating personas, hypothesising about audiences, segmenting users and running lengthy A/B tests to find the Holy Grail piece of content that our audience will love. We add tools which analyse messaging success to our already-complex marketing stacks, and then we do our best to move in the direction the data tells us to.

But when we do find that winner, do we know why it works? 

Do we know exactly what features caused the higher CTR? Do we know how we’re going to recreate it in our next campaign (or, ideally, to make it better)?

The ‘data gap’ problem

We’ve got a ton of tools and techniques at our disposal, all of which can offer great insight. But still, we often lack a basic knowledge of just why our stuff is working the way it does. 

We at datasine call this lack of knowledge the ‘data gap’, because when it comes to understanding that all-essential ‘why’, there’s a gaping hole in our analysis.

Simply looking at results doesn’t give us the insight needed to truly understand why people like what they like. At least, not in an actionable way.

Analysing your content’s value

To crack open that data gap, we need to start conducting in-depth semantic analysis of our content. Only then can we begin to truly understand why some content resonates and some doesn’t.

Experienced marketers come with a deeply ingrained fascination with (and, hopefully, understanding of!)  psychology and our audience. So, on paper, we’ve already got the skills we need to analyse our content. It’s simply a matter of breaking it down into parts. 

Let’s take a look at this in terms of images and text:


If you want to perform a semantic analysis of your imagery, you should take all the image assets you’ve ever created and note down the particular elements you used in each. You can then check to see if there are any patterns which seem to correlate to ad performance. You may well find that better performing ads have elements in common.

For example:

  • Did you use a photo of your product outdoors? Or in the showroom?
  • Were people visible in the shot?
  • What was the size of the text, and the colour of any overlays or CTAs?

It may even be worth inviting a panel to judge your images on the emotions that they evoke, or asking photographers to assess the quality and composition of the shot.

Semantic content analysis


You can do the same for text content, perhaps approaching this by categorising how you describe your product or service. For example:

  • Do you highlight your product’s ease of use?
  • Are you emphasising how innovative you are?
  • Do you use particularly casual language? Or are you more formal? Technical, or conversational?

With this process, we can see which types of content are receiving the most engagement. And we can use these features to keep creating great campaigns – which we can optimise on an ongoing basis as our understanding of customer content preferences grows.

Semantic content analysis

Scaling marketing content analysis

If we have just a few campaigns on the go, content analysis is easier. But it gets harder and harder as we scale. It quickly stops being practical to expect humans to spend days, weeks, even months labelling what goes into each piece of content. 

Luckily, we’ve got machine learning and artificial intelligence (AI) to come to our rescue.

AI models can extract all of the relevant elements in seconds by analysing image or text semantically. Using semantic methods, they can ‘look’ at content like humans do. By using AI in this way, we can cut back on lengthy, expensive A/B testing and get rid of guesswork once and for all. 

This is a vision which we at datasine are working hard toward. Our platform empowers marketers with intelligent, data-driven insights which improves marketing performance. Our platform breaks down content into ‘atomised’ aspects, and then compares these aspects with performance trends to identify those all-important common elements.

This provides you with valuable insights not only about what’s working – but also why it’s working and what you can do to improve results. And this, as any creative marketer knows, is vital for creating the best and most innovative new content.

By embracing semantic content analysis and working collaboratively with AI, we can feel confident in understanding exactly what content is going to work (and why!) before we hit send.

Some people think that AI stifles human creativity. Here, at datasine we know that the truth is the exact opposite. 

Semantic marketing empowers human creativity by helping us to see creative content from many new angles. It enables us to spot patterns which we may otherwise have missed, and to use those patterns in new, innovative ways.

At the end of the day, semantic marketing methods help us to see our content through our customers’ eyes. And isn’t that what every marketer wants?

Related posts

Let's work together.

Book a demo to see how Datasine empowers marketers with intelligent, data-driven insights to improve the performance of creative content.

Share This