Think you know your target customer?
What Is It?
If you’ve narrowed it down to one type of person with one set of habits, you are likely missing out on a world of potential expansion. Any product on the market likely has multiple customer archetypes who want the item for a variety of needs or perhaps for dissimilar reasons. Some customers have very different timing on when they want to buy a product. How do you discover and understand the variety of your dedicated and potential buyers so that you can better design and market to their needs, thus maximizing profits? Trig has you covered with our custom research approach. Mapping the customer engagement journey can encompass both qualitative and quantitative research methods.
When Do You Use It?
Understanding the engagement journey technique we’ve devised is particularly helpful in strategy formulation for product innovation and marketing segmentation. Defining accurate personas of your customer base can and should take time to create, but the investment pays off with the knowledge of how best to customize your products and communications to meet diverse customer needs. We can identify low-cost options to introduce variability in the customer experience using software, assembly, or color changes. The strategy is commonly used by the large corporate innovation teams, but is available to savvy startups as well.
What Can It Look Like?
Engagement journey research is visually fascinating in graphic form due to the quantitative/qualitative research combination. Putting numbers to more abstract concepts like feelings and interest bridges a gap in understanding between company and customer or between one stakeholder and another. Below are example analyses designed to pinpoint engagement fluctuation levels from a variety of research segments.
The Experience Journey
Individual engagement can be visualized in bar charts, letting the viewer see how different customers may group by commonalities in their feelings throughout their experience with a product or service. How these plots layout this journey, and how customers are grouped qualitatively and quantitatively, can be tailored to the needs of an individual project and the scope of information gathering.
Above you can see side-by-side representations of engagement level bar charts. Different identified personas have unique highs and lows through the product journey. Certain similarities can be identified and utilized for program-level decisions, while persona-specific nuances can often drive low cost customization to the product.
Below is a zoomed in view of how different example phases can be identified and focused upon through targeted research inquiry. For example, one path of inquiry may be to focus on how a customer experiences a task they need to accomplish, step-by-step, and use this information to identify opportunities to address their needs and improve their experience. Another approach may be a fine-tuned investigation – such as inquiry to understand the engagement journey for long-term customers.
Persona Verification and Market Sizing
“How can I reliably map customers to personas at scale?” you may ask. Intense data collection from a variety of relevant market segments gives us a view into the many needs and values that accompany anyone who may be interested in your product. It’s easy to average all the findings into one single ideal customer to go after aggressively, but averages derived from a diversity of customers may reduce information to a wrong conclusion or get it right but lack important context.
We design survey instruments with nuanced questions derived from the qualitative research, then use a statistical approach to identify multiple unique personas. Breaking the information down into data-driven personas is helpful to identify unmet needs that competitors may miss due to only performing surface-level analysis. From those identified common personas we can use incidence rates to derive their respective market share. This can be a powerful check on validating or refuting the qualitative findings, where small sample sizes and individual respondent salience can skew the findings.
Cluster Analysis to Predict Persona Attributes
Cluster analysis is a statistical method to effectively group customers – based on their responses to qualitative research questions – into “family trees” where very similar respondents are statistically sorted into groups with highest similarities. Mixed data cluster analysis approaches can be done for complex data sets that may combine binary data (e.g., true or false), numeric data (e.g., number of tools used to do a job), and categorical data (e.g., motivations to buy a product).
In our analysis approach, we also continually challenge ourselves to make sure the statistical analysis is robust. For example, what if we weighted “true” responses as more important than “false” responses – would the clustering change? What if we used 5 factors instead of 6 to create the groupings? How does excluding an outlier change the results? Sensitivity analysis as part of the process builds confidence in the results.
Once we identify unique groupings of customers, the next step is to paint a more complete picture of what these persona archetypes look like. Maybe all your customers are in their 20s, however there may be very different needs and experience between a 25-year-old parent, a 21-year-old single student and a 28-year-old getting promoted at her dream job but wondering if she’s balancing work and home life properly.
Want Rich Data Visualization? We Have You Covered.
One concise way to view an extensive amount of statistical information is through a violin plot. This display type is ideal if you want to compare engagement levels, distribution width including the representation of outliers, median, and interquartile ranges. Violin plots give additional information beyond typical box-and-whisker plots, by including the kernel density estimation that indicates where a survey taker would most likely reside along the span of the range.