By Jonathan Crossfield
Marketers love numbers. They constantly analyze data to determine what works and what does not, looking for percentages and metrics that can help them improve campaigns, boost engagement and increase sales.
But there is a problem that does not receive nearly enough attention: a number on its own does not tell the whole story.
Most marketing professionals are not experts in data analysis or statistics. Yet they deal every day with dozens of studies, surveys and reports presenting figures that appear conclusive, and they often use those figures as the basis for important decisions.
This is where the problem begins. Using numbers to answer a real question requires more than knowing the final result. We need to understand which data were used, how they were collected, the size of the sample, the methodology applied, what the result actually means and, perhaps most importantly, what the result does not tell us.
A Number Does Not Tell the Whole Story
One of the things that continues to surprise me about the marketing world is the popularity of claims such as: “Headlines containing odd numbers generate 20% more clicks than headlines containing even numbers.”
At first glance, this may sound useful. In reality, however, it tells us very little.
If headlines containing odd numbers really do perform better, the important question is not whether the number is odd or even, but why.
Do odd numbers appear more realistic? Do they give readers the impression that a list is more detailed or carefully researched? Is their use associated with a particular type of content or audience? Or is another variable entirely responsible for the result?
These are the questions worth answering.
Saying that “headlines with odd numbers generate more clicks” is like saying that most leaves are green. The statement may be true, but it does not tell us why.
Nevertheless, this kind of statistic has become the foundation for countless marketing tips circulated through blogs, newsletters and social media, eventually taking on the appearance of established rules.
The problem is that marketing does not work that simply.
How Numbers Create an Illusion of Precision
Perhaps the explanation for the preference for odd numbers is simpler than it appears. It would be unreasonable to assume that audiences have some mysterious bias against even numbers.
A more likely explanation is that certain numbers appear more specific and less artificial than others. A headline such as “100 Tips to Increase Sales,” for example, may seem overly neat and may give the impression that the headline was conceived first and the content assembled afterward.
By contrast, “87 Tips to Increase Sales” may suggest that the writer researched the subject in detail and simply could not find another tip worthy of becoming number 88.
But even this explanation does not justify turning the observation into a universal marketing rule.
Does a headline containing “88 Tips” really seem less comprehensive or less precise than one offering “87 Tips”? I doubt it. It is also difficult to imagine a study capable of conclusively proving such a rule across different markets, audiences and types of content.
And this is precisely where the problem lies: we confuse the existence of a relationship in the data with knowing what caused it.
Two variables may move together, but that does not mean one caused the other.
Where Did the Statistic Come from in the First Place?
If we want to take any marketing statistic seriously, we should begin with the simplest question: Where did it come from?
Take the statistic about headlines containing odd numbers. It has been cited by numerous marketing websites and blogs, then reproduced in infographics and other articles until it has come to resemble an established fact that no longer requires scrutiny.
But tracing the statistic back to its origins reveals a different story.
Some references lead to an older study by a company specializing in content recommendations, based on an analysis of approximately 150,000 headlines. That may sound impressive, but the number alone is not enough.
What kind of headlines were included? How were they selected? What period did the study cover? How were the results compared? What other factors did the researchers take into account?
And was the original study published in sufficient detail for others to examine its methodology and findings?
Without answers to these questions, the sample size becomes just another number used to create an appearance of credibility.
The problem becomes even greater when we discover that a statistic is outdated but continues to be recycled years later as though it still accurately describes audience behavior today.
This is particularly important in digital marketing, where platforms, algorithms and user behavior change at remarkable speed. What was true five years ago may no longer be true today.
And what works in the US or European markets will not necessarily work in Saudi Arabia, Egypt or the UAE.
Context is not a secondary detail. It is part of the statistic itself.
Numbers Without Definitions Mean Nothing
There is another problem that is far more common than it should be: the use of terms that sound precise but are actually vague.
It is easy, for example, to come across a statement such as: “High-quality images receive 121% more shares.”
But what exactly does “high quality” mean?
Are we talking about image resolution, design, aesthetics, relevance to the content or a human assessment of quality?
If we do not know how “quality” was defined, the highly precise number at the end of the sentence does not give us genuine precision.
The same applies to words such as “engagement,” “virality” and “reach.” These terms have become fundamental to the language of marketing, but they can mean entirely different things from one study to another.
This creates a dangerous trap for marketers: they may be dealing with an extremely precise number built on an extremely vague concept.
Don’t Confuse Clicks with Results
One of the clearest examples is the tendency to focus on the number of shares or visits as direct evidence that content has succeeded.
A piece of content may generate thousands of shares, but how many people actually read it? How many remembered the brand? How many took a step toward making a purchase? And how many ultimately became customers?
These questions are far more important than the number of shares alone.
Focusing on a single metric because it looks positive, while ignoring other metrics that are more closely connected to the campaign’s actual objective, is like choosing the evidence that supports the conclusion we have already decided we want.
That is not data analysis.
It is data selection.
The Obsession With the “Magic Formula”
The problem is not limited to one statistic, one company or one study. It reflects a much broader problem within the marketing industry itself.
We are constantly searching for a magic formula that will tell us exactly what to do.
- Should an article headline contain eight words or 14?
- Should an article be 250 words or 2,500?
- Is it better to send an email newsletter on Monday or Tuesday?
- Should content be published in the morning or evening?
There is nothing inherently wrong with asking these questions. The problem begins when we treat the answers as universal laws, because marketing is not a fixed equation.
Audiences differ. Markets differ. Platforms differ. Products differ. The stage of the customer journey differs. Even language and culture can change the outcome.
What works for an audience in the United States may fail completely with an Arab audience. What works in the technology sector may not work in financial services. And what works for one generation may not produce the same results with another.
Therefore, the right question is not: “What number proves that this approach is better?”
It is: “What are these data telling me in this specific context?”
Don’t Let the Numbers Do Your Thinking
Numbers matter, and they may be more important than ever in the marketing industry. But greater reliance on data should not mean abandoning critical thinking.
Quite the opposite. The more data available to us, the greater our need to understand where it came from, examine its context, test it and compare it with other data.
A statistic is not a fact simply because it appears in a polished infographic. Nor does a study become reliable simply because its sample size is large.
Before using any number to guide a marketing strategy, ask:
Where did it come from? When was it collected? How was it calculated? Does it demonstrate a causal relationship, or merely an association? And does it apply to my audience, my market and my platform?
If we fail to ask these questions, it is easy to move from being marketers who use data to marketers who are driven by data.
And that is the greatest irony of all: the numbers that are supposed to make our decisions smarter may themselves lead us to worse decisions if we treat them as ready-made answers rather than tools for understanding reality.
Jonathan Crossfield is a content marketer, writer and speaker based in Australia. He specializes in content marketing and writes about digital media.
Related Topics:
Inside the Six Essential Steps Behind Successful Marketing Campaigns
From Trends to Timelessness: How Brands Can Create Marketing That Lasts
