Your agency is cutting-edge. Innovative. Progressive. You tout that your digital agency designs based on the customer’s needs, not what's hip, so it lasts. So, you’ve built them a detailed profile of their ideal customer, not just for their target market, but their target person, outlining everything about them.

Where they live. What they do. What they listen to. What they buy. About their family, favorite stores and publications. Who they like. Who they don't. Celebrities that matter to them. Books that influence them. It’s a brilliant portrait of someone detailed with skill, prowess, and poignantly crafted to paint the perfect customer so you can speak directly to them.

Except the problem is, this is completely wrong. 

Customer Avatars Dont Exist

This girl does exist. She's just not your customer.

I hate to say it but “Sophie, 27, from Chicago who likes Pinterest, teal, french bulldogs and shops at Whole Foods, watches Ellen (but only in Youtube videos) after her makeup tutorials and uses apps to search for ingredients and drives a smart car but usually bikes"–she doesn’t exist. 

In fact, no one does that you’ve created for your customer avatars. And even if they could, they’re from extrapolated brand journalism scenarios or research groups, not mass purchasing data. In fact, the data say your avatar is not indicative of every customer out there. There is no such finite subset and doesn't exist. 

In fact, not only do these customers NOT exist, customer avatars DON’T WORK according to brand science whatsoever. Please don't throw the computer at me, you'll want it for later, it's expensive. Also, hear me out.

I know this flies in the face of every marketing book, specialized online marketing course, and the first 10,000+ search results, but it's true. So let's dive in and talk about why and what this means for you and your work. 

“The difficulty lies, not in the new ideas, but in escaping the old ones, which ramify, for those brought up as most of us have been, into every corner of our minds.” John Maynard Keynes.


The year is 1959. A study (Evans 1959) was created to identify and discriminate the psychographics between Ford and Chevrolet customers to gather insight on how to market to those groups more successfully. Big execs were ready to capture this data and pounce on their marketing gold. Previous research had suggested that these customers were vastly different. 

The cars were almost identical and perfect substitutes–features, prices, models and the other features almost completely identical. The theory was that car customers in America took their purchases seriously and marketers had invested in marketing, the brand identity, and images. The goal was to discover the ways that customers thought, identified, and create respective customer bases out of them, and executives were drooling for the results. 

But that’s not what happened. And in fact, the personality traits and demographic profiles of those in the study were literally identical. The results shocked the world and arguing ensued. Then the findings were empirically confirmed repeatedly, again and again—even until today.

Over the next 50 years, most comprehensive studies in marketing science at the Ehrenberg-Bass Institute for Marketing Science at the University of South Australia would discover the profiles of brands in various categories, discovering that brands sell to the same people, not one unique specific, target market. 

Customer Avatars Dont Exist

Not much has changed, has it?

As the years have progressed, marketing has gone through as many iterations as mankind in its forms of rulers and kingdoms. In fact, their voices are quite similar resulting in some false beliefs of how things work. Typically, the loudest beliefs and fallacies tend to rule the most pervasively. Honestly, these companies still act the same way as they did in 1959.

Before ads arrived on TV in 1941, consumers were exposed to product-central ads, depicting stories, problems, and how the product was always the solution. Visit a Cracker Barrell, you'll see an homage to advertising’s past hung in their antique decor. If you’re a history buff, it was very much "Roman Catholic Church, pre-Martin Luther"–no one else could interpret the product for us, the company was always right, peddling a single, unwavering message to the people.

In later years, as radio, newsprint, and magazines merged with TV ads, the full-fledged media campaign was born, thus creating the ability to mass market. Now corporations could say what they wanted customers to believe en-force.

History proved that as time went on, customers discovered they had voices. Organizations, like the Consumers Union, in 1937, formed to protect consumers from fallacious claims and started reviewing products, later to be known as Consumer Reports. Over time, the FDA or even Congress stepped in when there were product tampering (c.f Tylenol, automobiles), and safety concerns. The products themselves were forced to be evaluated and messaging had to be created to sufficiently address the issues. Corporations had to be more careful that they claimed. 

And likewise, as time went on, so did the advertising industry that so depended on stereo-normative tropes, being forced to change to address means of communication, technology, and user experience.

Agencies were problem solvers, and addressing the most traditional issue of supply and demand, have consistently deduced that if it wasn’t an issue with the message going from sender to receiver, it clearly must be an issue with the receiver–therefore it needed to be clarified. They clearly weren’t targeting the right person. Little did they know, they were overcomplicating reaching customers, creating unrealistic expectations, soon-to-be doomed campaigned, and waste a lot of money. 

As a result, marketing transitioned through mass marketing, to product-focused marketing, ultimately to user-based marketing mixes, which resulted in the creation of the idea of super-focused target marketing. Again, a myth was propelled and everyone swallowed it up.

In fact, we still do. Countless "marketing gurus" make their living by selling ideas on how to market your goods better, get inside the human psyche, and unlock and unleash, yada, yada, yada, yada. 

It’s time to unlearn. 

Traditionally, marketers believe that their clients are a unique set of people, devoted to them that can be stolen, so then these customers must be educated, protected, and consistently nurtured. This is false.


Recently I had a partnering agency approach me about creating a set of Distinctive Brand Assets (DBAs) for their client. These are the set of sensory assets that include brandmark, color, type, tagline, etc. 

They started by sharing the locale of the company (local, regional, national, international), the complexity of the products/services (one, few, or several), the complexity and size of the organization, and its departments, as well as the unique factors of what they need (A car wrap vs in-store signage, etc), as these are the core factors of understanding value-based pricing, research, and determine the data from which you get started to research. 

She then shared something else. A customer avatar. Oh no. It derailed the conversation and honestly, while it was based on the client's firsthand experience, it was mostly all "Susie, 37, she likes this, that, and has this many kids"––quantitative data that may seem important, but in actuality, distracts from the actual success and understanding of how products and services sell in the marketplace. 


This isn’t uncommon with agencies is the customer avatar. It’s actually become standard practice. ideally, it’s meant to provide a level of clarity, sophistication, and focus on the client’s marketing, right? But it doesn’t. Here’s why. 

Using, and creating, and customer avatars have become the norm. They’re the natural extension of the phases of marketing that advertising has gone through, but they’re not a true, validated practice we can utilize metrics to. They’re quite hard to escape, let alone not implement, when their use is often tied to agency-learning courses, discussed by industry influencers and marketing “gurus.” They, in fact, complicate marketing, and focus efforts on a smaller portion of the market and limit sales altogether.

For example, when you buy a can of Coke or a piece of Nike apparel, are these brands utilizing a customer avatar to attract you? No. Actually, the majority of their sales come from light buyers–people who buy 1-2 items each year. You’re considered a heavy purchaser if you purchase 3 or more items a year. Empirically, heavy purchasers, at most, account for 30-50% of total sales of a brand’s annual sales. So, why would you want to only target half the market? You want all of the potential sales possible. Focusing on one small facet limits growth. 

According to evidence-based metrics, the only factors we can, in fact, organize our customers into, is their purchasing behavior and product-related factors, not their demographics. 

According to evidence-based metrics, the only factors we can, in fact, organize our customers into, is their purchasing behavior and product-related factors, not their demographics. While the idea of who your customers are is appealing and satisfying, bringing a sense of confidence into your client’s marketing, it's a false sense of confidence based on imaginary conditions. 

Instead, our customer groups should be thought of as

  • random/sporadic
  • light
  • mild
  • moderate, and
  • heavy purchasers

By focusing on the frequency of buyers you are, in fact, focusing on the complete possible buying groups possible, and thereby focusing on the most growth possible. This is the case for strategic mass marketing, which sells to all possible buyers by having many occasional buyers, accepting weak relationships


When you point the portrait you're trying to paint towards the company’s growth, goals, and product sales and away from a customer avatar, you begin to focus in the right direction. 

After the partnering agency handed me the customer avatar, it took some thoughtful language to reframe. This is the same language you can use too. Here are some questions and conversation starters you can use to really focus things.

      In this instance, you can ask the client (when they hand you an avatar or customer profile they believe is accurate) about the purchasing behavior, not the demographic information.

      “So, can you tell me, I don’t see what level of commitment or participation the client avatar has to the program, could you paint me a picture of that?” From here, you can begin a discussion on purchasing and identify that they thought only one type of buyer existed, the frequent, committed buyer and share how this person is only one of several types of buyers. The goal is to paint a cross-section of purchasing, not just one.

      CEP’s are the who, the what, the where, and why people buy. For champagne, it's with friends, coworkers, or romantic partners at home, at a restaurant, event, or workplace, to celebrate a personal, professional, or romantic occasion. Marketing honed in on these occasions will seem intuitive and natural. Getting a handwritten note on the bottle or accompanying secondary packaging would put it over the top. By understanding these CEPs, you can truly address the situations where, when, and why people buy and make a more strategic decision on the top types of marketing you should use with your clients. Typically, the goal is two proven forms, so be sure to A/B test and validate.

    3. GET DATA
      Most companies have some form of data on their customers. For service-based businesses, it'sits in the form of invoices, amounts, and dates. For product-related businesses, you’ll want to get reports on distribution channels, look at SKU performance, and identify the peak periods, sales, items, and channels that are working best for you. While this could be a separate topic all on its own, the goal is to get data and start interpreting. You can’t make decisions on the frequency of purchasing without it.

      This is one of the biggest mistakes I see every day. Most agencies don’t have access to data, know where to get it, or where to go. There’s a lot of powerful research firms out there. Depending on the size of the client, some may be out of your budget. Some major market leaders are IPSOS and Nielsen, while others such as Lucid or Trend Source. The most important factor is understanding what KIND of data you need. Finding the one that services your client’s industry is imperative. 


When you shift the conversation from customer avatar (false metrics of demographics) to customer frequency (evidence-based metrics) you’re making a strategic shift as an agency, a strategist, and designer/developer who is focused on your clients’ growth. The brands you service now will have a metric behind what they’re trying to achieve.

If you know that you have very few light buyers, you can now start asking questions about the CEPs your client has, and whether or not there should be more validation or development to their product line, sales experience, or even simply distribution or marketing channels.

When you focus on accessing a complete portrait of consumer purchasing, you’ll be moving your clients into a conversation about growth more naturally and most importantly, you’ll become a true agency partner––someone who partners with clients in not just their design and marketing, but their growth. 

This is a huge shift, and in return, positions you for increased success as an agency. Things will change. You’ll go from being a designer to a strategist; a web developer to a growth facilitator, a social media person to a CEP craftsman, and everyone will go from being seen as a technician to a valuable resource to your client. 

You’ll be able to focus on increasing the marketing channels that work, advising on the ones that don’t with your CEPs and be able to craft more sophisticated mass marketing that is profitable for everyone, including every type of buyer that actually exists.



Evans, F 1959, 'Psychological and objective factors in the prediction of brand choice Ford versus Chevrolet', The Journal of Business, vol. 32, pp. 340-69.