Impression is a multi award winning digital marketing agency that's known for its data-driven, creative approach to exceeding client goals and delivering exceptional ROI across all digital channels.
This promoted content is produced by a publishing partner of Open Mic. A paid-for membership product for partners of The Drum to self-publish their news, opinions and insights on thedrum.com - Find out more
Narrative fallacies through the lens of an SEO
26 June 2020 9:10am
Lately, I have become increasingly mindful of how pervasive narrative fallacies can be within the digital marketing industry. Granted, I may not have always been aware of the term - or just how much it's used as a rhetorical device - but the essence of the problem has always been there for me, subconsciously at least.
One of my first encounters occurred last year when a particular situation escalated to me. A client was hesitant to implement some of Impression's strategic recommendations, and after initial negotiations occurred, it was clear how both party's stances were hardline. We were confident in our recommendations, but equally, the client was also confident as to why we shouldn't pursue those recommendations. As any digital marketing professional can vouch for, this can (and does) happen from time to time. It's all part of the working relationship. And in fact, that part doesn't really matter. We did eventually move on from this, working on other campaigns that suited both parties.
What I found particularly interesting though was unpacking why both parties felt strongly about their stances. Both came from a perceived data and experience-perspective, so who was right? And who was wrong?
With the work never implemented, results weren't available to understand whose thoughts were correct and whose were fallacious. However, as an initial bystander, it was clear there were personal narratives at play that were influencing decisions.
What is a narrative fallacy, and what does it mean in the context of digital marketing?
A narrative fallacy is a reductive conclusion given to a situation when only a cross-section of the facts are considered.
In the context of digital marketing, this is often used interchangeably with confirmation bias and typically occurs before or after the fact when interpreting performance data. In a bid to create a strategy, we choose the data we want to see, to tell the stories we want to use.
To the uninitiated, narrative fallacies can be very difficult to avoid as we're psychologically hardwired to slip into them. We almost have a reliance on them from an evolutionary perspective to make sense of the world around us. By connecting the dots and telling stories, they make us feel more in control - however simplified these 'stories' can be.
Despite them being easy to slip into across all walks of life, there is hope for digital marketers. We can employ a disciplined mentality to ensure we're at least mindful that they may be occurring in a bid to avoid them entirely. After all, falling ignorant to narrative fallacy can result in us pulling the wrong strategic levers where we fail to achieve our desired KPIs.
The importance of data
The importance of data to avoid narrative fallacies cannot be understated. In fact, the beauty of data is how unassuming it is and how little it cares about any preconceived notions you have. It is, as they say, what it is.
However, the critical part is having enough data to arrive at any form of valid conclusion and not presuming too much when gathering and analysing the data.
As an SEO by trade, I operate in a space of digital marketing where a lot of data is available to me. However, it's a space that's also governed by specialisms, stakeholders, users, competitor activity, bots, algorithms, machine learning - you name it. What satisfies one may not satisfy the other, and it's this interplay that makes SEO feel like an ongoing puzzle to solve. It's also a vast discipline that covers many areas of digital marketing by its very nature. This can make it feel like a blessing and curse at times, especially when we examine it through the lens of narrative fallacy and data collection.
Generating enough data
By its very definition, an effective strategy needs to hinge on data. It's just about how much of the right data you can get.
As technology has advanced, the accessibility of data has come on leaps and bounds and so too has the accessibility (and affordability) of big data through SaaS solutions. This presents an opportunity digital marketers can leverage to build more informed pictures of their market and channel performance.
Within SEO in particular, many tools offer big data at the core of their product offering but understanding how these datasets differ to one another can make the difference between an effective arsenal that's thorough to a lacklustre one.
Creating aggregated and expanded datasets also need to be acknowledged here as these can serve to increase the quality and depth of your analysis.
For instance, Ahrefs, Majestic and Moz all provide backlink analysis solutions with each being able to fulfil that fundamental requirement (if this is an avenue you're looking to pursue). However, with these, the backlink coverage you see is inherently at the mercy of each provider's backlink index. Aggregating these, therefore, ensures your data is greater than the sum of its parts.
Expanding your dataset can then come from blending data sources, a tactic that has been popularised by Google Data Studio more recently. Though this can be very effective at painting a more informed, data-focused picture, one should be wary of correlation and causation here as this inevitably can influence a narrative fallacy later down the line.
A continued competitor focus also falls under the remit of expanding your data set. It is my narrative (fallacious or not) that leads me to believe how easy this is to overlook as a digital marketer, especially as strategies age and one becomes more comfortable and complacent with ongoing activity.
Again, from an organic perspective, ongoing competitor analysis is one of the only sure-fire ways to validate your efforts and understand how you're being perceived by search engines algorithms - one that isn't so shrouded in mystery, at the very least. As SEOs, we often audit our sites, looking at how they're optimised for crawling, indexing and ranking, but how often do we benchmark these factors against competitors? The data is available to allow us to do this, so we need to ensure that we monitor competitor activity regularly to avoid short-sighted narrative fallacies and promote resilient strategy.
Avoiding preconceived notions when interpreting data
If you're interested in learning more about narrative fallacies in the digital marketing space, Andrew Anderson's article for CXL is a must-read. Though focused more from a CRO perspective, he details just how susceptible teams are to interpreting data with preconceived notions already in place - be that why the data is trending the way it is, or how.
His post explains how, as part of a training program he used to conduct, he used to ask team members to explain why every version of specific A/B tests were the best options and why they would win. Though trainees were initially confused by the notion of justifying results before they were generated, he found that eventually, stories such as those listed below were commonplace amongst the team's responses:
"Test A will resonate with audiences better as it improves the value proposition while removing content clutter."
"Test B adds more to the conversion path by reinforcing key USPs"
"Test B uses more colours while driving more product and brand awareness"
The next part of the training involved encouraging the team members to speak to the consultants who organised those particular conversion campaigns and to see just how quickly similar stories were also generated by them.
Indeed, extracting emotional and behavioural meaning on aggregate from such campaigns is near impossible. Such insight doesn't actually serve to improve the quality of the data either. Instead, Anderson argues its irrelevance and how the questions below are better placed when exercising critical thought around strategic analysis:
Can you act on the results?
Can you move on to the next test? And;
Was the original set-up created in a way to maximise efficiency?
It's this type of mentality in performance marketing that has undoubtedly increased the popularity of A/B testing; not just for CRO and paid media advertising, but now for SEO too. SearchPilot (previously, DistilledODN) were arguably the pioneers of A/B testing in the SEO space, illustrating its importance via their 2017 study that showed how a random coin flip is better at predicting rank than a well-versed SEO and their narratives.
Now, that's not to say that A/B testing is the only way to deliver an effective SEO strategy. Though it can undoubtedly help and would align with my earlier point around generating enough informed data, this may not be achievable within the confines of a business or agency's budget.
Instead, what's critical is exercising this idea of discipline and understanding exactly where narrative fallacies occur in digital marketing, irrespective of how much data you have at your disposal.
It’s at this juncture that I would like to expand on this concept of avoiding preconceived notions by introducing the narrative fallacy that’s likely central to it all, The Texas Sharpshooter Fallacy.
The Texas Sharpshooter Fallacy
The Texas Sharpshooter Fallacy concerns focusing your insight to support any pre-existing ideas you had while ignoring outlying information. The analogy derived from a story of a Texan that’s shooting the side of a barn. After firing the last round, they walk over to the barn to draw bullseyes around the sets of holes to make them look like a great shooter. In essence, cause and effect reversed, where the latter is only determined by what the former is ‘targeting’.
In the context of digital marketing, envisage a strategist who ignores data outside of the desired result they are looking to achieve.
Using this bias as the basis of narrative fallacies is arguably the pattern I see the most. Take digital marketing specialists as one example. Though their skills have specific applications by their very nature, they can unknowingly fixate on these when it isn’t required for them to do so. From an SEO perspective, the problem here lies with the notion of “optimisation” and how vast the discipline is (as mentioned previously). Several factors of a site can be optimised, and several factors off-site can be optimised, but it’s knowing which optimisations are worthwhile implementing that always proves the most effective. This is where a specialist’s experience counts and someone who can acknowledge whether you’re conducting optimisations for the sake of it. It’s an innate ability, for sure, and one that takes time to develop.
Rounding it up
Of course, removing narrative fallacies entirely from the marketing mix is unrealistic. However, how they’re ingrained in human nature doesn’t need to be accepted outright - only acknowledged. Being aware of their shortfalls is half the battle and challenging you and your teams to think more critically about strategy and market analysis is the key here.
Are you basing your strategic analysis on a complete dataset? Are there any processes or behaviours present that are causing potentially biases? Approaching research holistically can help you avoid any pitfalls caused by narrative fallacies.
The ideas I have detailed here are largely conceptual. Though I’ve focused them through the lens of SEO, I hope they resonate with you irrespective of the digital marketing circle you operate in. At the very least, I hope this has made you realise how misleading those silly growth hacking YouTube adverts really are; you know the ones...
X startup used this one tactic to get Y result!
They plague the internet, but there’s always way more to the story than exclusively one thing that increased revenue tenfold ;)