14 Dec Using “Old School” Personas is a Serious Marketing Mistake
Let’s be honest. There’s probably a beautiful document, maybe even a video, a large wall chart or even a digital dashboard that contains your personas floating around your company. For generations of marketers, the persona served as the guiding principle for nearly all communication strategies. Your personas were the result of psychographic, demographic and ethnographic research. You made serious investments in those personas and your agencies / partners worked for months to develop them. There’s only one problem. They’re completely irrelevant. Let’s follow the logic.
Personas are based on the following:
- Market Research
- Audience Insights
- Categorization of Attributes (Behavioral, Demographic, Psychographic, Ethnographic)
- Purchase / Decision Influences
- Response Criteria
- Messaging & Media
Personas are supposed to be representational of key audiences. They are the basis from which brand architectures are developed, messaging strategies defined, creative execution aligned and measurement defined. Sadly, they are expensive exercises that today, are completely irrelevant and essentially counter intuitive to marketing success. Let’s delve into the “why?”
Fragmentation, Pace, Interest, Engagement & Influence
Personas were based on data sets that effectively represent a “moment in time.” Having worked with highly recognized companies that specialize in research, marketing automation and media planning, it’s interesting to see how legacy approaches continue to permeate insights and drive decision making. As we describe it at Trade, in a world where data drives narrative, storytelling drives brand and technology facilitates participation – engagement is dynamic.
If you’re using personas, they’ll most likely be built from a combination of the following:
Demographics: 1.) Gender; 2.) Age (Range); 3.) HHI; 4.) Marital Status; 5.) Size of Household
Psychographics: 1.) Values; 2.) Interests; 3.) Attitudes
Ethnography 1.) In-depth visits to audience locations; 2.) Contextual interviews; 3) Product / service usage
How do you solve for fragmentation, pace, interest, engagement & influence?
- Start with data that accessible and affordable; build data models that incorporate paid, earned and owned media into a set of customer and prospect “views.”
- Combined with tools that capture “human generated information” (i.e. blogs, social posts and news sites) built off of DataSift’s core feeds such as Zignal Labs, Nuvi, Hootsuite, etc.
- Overlaid with paid, earned and owned (standard analytics from Adobe, Google, DoubleClick, etc.) with data sets from tools such as Salesforce, Pardot, Eloqua, Marketo, SAP, Oracle, Genesys and Avaya with
- Data sets from tools such as Ensighten and Tealium to deliver experiences that drive customer experiences on owned platforms such as Adobe, IBM, Microsoft and Oracle and provide additional data sets that make programmatic significantly more value rather than solely optimized for spend.
This sounds complex, but in actuality, it’s the basis for real-time / near real-time decisioning in support of marketing spend and it’s easier than it sounds. With the opportunity to capture insights from social channels where people aggregate into discreet communities that are in many instances, temporary – the model above provides the ability to build a marketing structure that operates more closely to a newsroom, (editorial) than a waterfall of research, brand messaging, asset creation, media buy and measurement. It affords your organization the ability to deliver the right content, to the right audience in the right channel at the right time. Your target audience(s) are constantly changing / evolving – hourly, daily, weekly. The question is – Is your marketing strategy dynamic? If it’s driven by legacy persona models, it will limit your ability to intersect to your target audience.
Don’t take our word for it – talk to our clients.
To learn more about how Trade can assist your organization, contact 404-900-5592
To contact the author
Robert Morris, Partner