Eye-Catching and Sales-Making
I spent two decades of my life honing my copywriting skills – writing everything from classified ads for real estate, to political campaign copy, to scripts for online infomercials, and just about everything in between.
I wholeheartedly believed that the words a salesperson says (or copywriter writes) make all the difference.
In some ways, I believed that – through words – we can hypnotize. And that if you say the right magic words in the right way, people will automatically buy.
Most copywriters draw insights from the world of Neuro Linguistic Programming (NLP), psychology texts, and even stage magicians as much as they draw from the world of business, salesmanship, and marketing. But the best insights come from testing ideas in-the-wild to see how they perform: split testing.
In recent years, the insights gleaned from split testing shattered my illusion that it’s the words on the page are what convince people to buy.
As I began to test more rigorously, and at larger scale, I began to notice that the changes I made to words had less of an impact on sales than other (more visual) changes.
Eventually, as I was reading Sapiens by Yuval Noah Harari, the reason dawned on me.
In the opening pages, Harari provides a brief history of human history, which I’ll summarize below:
3.8 billion years ago – the beginning of biology
521 million years ago – the first creatures with brains
55 million years ago – the first primates evolve
6 million years ago – the last common grandmother of chimpanzees and humans
3 million years ago – our ape ancestors begin walking upright
2.5 million years ago – the first stone tools
2 million years ago – our pre-homo sapiens ancestors leave Africa for Eurasia
500,000 years ago – neanderthals evolve in Europe and the Middle East
300,000 years ago – our ancestors are using fire
200,000 years ago – our species (homo sapiens) evolves in East Africa
70,000 years ago – the first use of verbal language
5,000 years ago – the first written languages emerge
The evolution of our brains can be traced back 3.8 billion years ago. And (to oversimplify it) every now and then, we get an “upgrade” – a little bit of extra brain tacked on top of the existing brain bits we have. 3 million years ago, we got the walking upgrade. 2.5 million years ago, we got the stone tools upgrade. Just 300,000 years ago we got the fire upgrade. And only 70,000 years ago, we got the language upgrade. And it was all plugged into a 521 million year old “base model” brain computer.
As a result, the way we process information is through the oldest “lizard” parts of the brain first (threats/opportunities). Then newer “mammalian” parts of the brain (patterns and metaphors). Then the newest “human” parts of the brain – like the neocortex (which is the part of the brain that handles higher level reasoning and language processing).
Because we’re “visual first” creatures, the visuals on the page are actually more meaningful and impactful than most people give them credit for.
Proof #1: Road Signs
Without speaking Chinese, you can tell exactly what this sign says, right?
Road safety signs are a masterclass in visual cues. They’re universally understood, and rarely use text.
Lesson: Visual cues by themselves have a strong “conversion rate”. We use them, and understand them, ahead of text cues, in life-and-death situations.
Proof #2: High Fashion
Every high fashion brand website runs on minimal text. All “sales messaging” happens visually. Many (including Burberry below) don’t even describe their product – dimensions, material, etc – unless those details are in the title of the product.
Lesson: We can sell very valuable things without leaning on text.
Proof #3: Foreign Websites
TaoBao has no non-Chinese version of their website – but a global user-base. Despite the fact that a portion of their users cannot read the text, there are whole non-Chinese-speaking communities (notably the “replica fashion” community) who buy from it.
You can see this looking at the Google Trends for TaoBao including in search volumes in Arabic (Saudi Arabia), Russian (Georgia, Russia, Kyrgyzstan), Portuguese (Brazil).
In this image, if you were looking for a bag, you’ve got the visual cues on the left. Wanted to search for something, it’s clear where the search bar is. Want to complete a purchase, you’ve got the cart icon top right.
Lesson: Being unable to read text WON’T bring conversions down to zero – IF there are visual cues.
Proof #4: Textless Games
We don’t have to look far to find video games that exist seamlessly without the need for text or dialogue. Missing Translation (above) is just one of these.
But even in games that use text and dialogue, the use of these tend to be secondary to visual cues.
Lesson: The usability of video games is driven by visual cues ahead of text cues.
Proof #5: Studies on Verbal vs Non-Verbal Communication
Since 1971, psychologists have been studying the role of both verbal and non-verbal communication in various contexts – work, service delivery, sales, perceived leadership performance, and more.
50 years on, there are over 536,000 published papers on verbal and non-verbal influence.
Almost universally, they point to non-verbal cues being as-powerful, if not more-powerful-than verbal cues in eliciting positive responses from test subjects. As one study from the European Journal of Marketing notes: “Non‐verbal communication has been extensively studied in the psychology and psychotherapy disciplines and has been shown to have a central effect on participants’ perceptions of an event.”
When we’re left to rely on text cues (such as in remote working environments), or when we say one thing but demonstrate a conflicting non-verbal signal, we tend to feel negative emotions.
Similarly, we see that gestures cross cultural boundaries, and prove to be useful and effective in achieving outcomes despite cultural and language barriers in situations like usability testing.
Lesson: We respond stronger to non-verbal cues in human interactions than verbal cues.
Proof #6: Virality on social media
The vast majority of viral posts on social media are visual.
Of the top 20 most viral tweets of all time, 8 were visual (images or videos) with little or no text.
The only thing that is more retweetable than something visual on Twitter is an offer. (i.e. “If I get X-many retweets, I’ll do Y”.) 10 of the top 20 fit this.
Similarly on Reddit – only 2 of the top 50 posts were text-driven. The rest were visually-driven.
Lesson: We respond stronger to visual cues than text cues.
Proof #7: Business Logos
Increasingly, large businesses are saying it without words – MasterCard being one of the most recent converts. Apple, Nike, Twitter, Pepsi, Starbucks, Chrome, Total, WWF and more have joined a growing number of organisations ditching ALL text from their logos, and creating “Figurative Marks”.
Lesson: We don’t need text to recognize brands. We just need visual cues.
Proof #8: Conversion Testing
2016/17 was the year I discovered visual cues outperform text. That year, the top performing CRO tests included: making add-to-cart buttons larger, removing logos from a website, increasing the size of star rating icons, and changing the colour of add-to-cart buttons.
Only 1 of my top 5 tests that year involved text. Three different offers were added to the “Add To Cart” button. The “Buy Now, Pay Later” option outperformed the null, “Free Shipping” and “Fast Shipping” alternatives.
The best part about this is ALL of these tests is that they can be tested site-wide to see the universal effect on a whole ecommerce store. Copy changes can only be tested one page at a time.
Lesson: Non-textual cues outperform text cues on websites.
When you’re thinking about your messaging, try removing the words or putting them in a language you can’t read.
If your promo is still convincing and easy to buy, you’re onto a winner.