Customer Analytics For Dummies (2015)

---

It’s important to create a persona that represents the primary segment of customers for a product. The advantages of thoroughly researched and well-developed personas are:

· As stand-ins for real users, they guide decisions about design and functionality.

What is Marcus going to use this product for? If Marcus’s needs and skill level are already defined, it is simple to figure out how to serve him better. Personas help prevent programmers or designers from creating products that would be perfect for themselves or from making guesses about what customers want. A persona helps keep the end user involved in the product development.

· They allow you to concentrate on designing for a manageable target who represents a larger group.

Focusing on one customer segment and fully satisfying it is much more efficient to increase sales and profits. Trying to satisfy all your customers at once does not work.

· They give a common and consistent goal to the design and marketing teams.

Marcus, with his particular motivations and level of proficiency with the product, needs to be on everyone’s mind. If everybody works to satisfy him, not only the design of the product but also the message sent to customers will be consistent and more convincing.

· Prioritizing design elements and resolving design disagreements can be done in an economical way.

Personas are great communication tools. In explaining decisions to colleagues or people higher in the company hierarchy, use of personas helps refocus the conversation on the target customer.

· Identifying opportunities and product gaps to drive strategy becomes easier.

Having a precise idea of the target customer, actually trying to see things from his point of view, is an excellent starting point for brainstorming. Novel feature or product ideas can emerge from this single customer-oriented thinking.

· Designs can be continually evaluated and validated based on personas to reduce the frequency of usability testing.

If the whole programming team strives to make a product designed for Marcus, there should be fewer surprises when conducting real-life testing.

Working with personas

When you embark on the process of creating customer personas, multiple personas typically emerge as you identify various customer segments and use cases. The number and type of personas reflect the diversity of your customer base.

For example, automotive websites like Cars.com, Kelley Blue Book, and Edmunds.com have different types of customers that research information on cars. Some customers are auto enthusiasts, others are parents with kids, and still other significant segments of the customer base are young professionals intimidated by the car-buying process.

One such car company created seven personas from the data it collected from interviews and surveys on how customers researched automotive information online. The company broke the seven personas into four primary and three secondary personas. Breaking the personas into primary and secondary groups helps website developers decide on the prominence of automotive information. The primary customer segments are embodied in primary personas that want many details about car specifications. The designers then cater to this primary class of users by providing multiple views on engine, body, performance, and feature details. However, they can’t ignore the secondary personas who represent another significant portion of website visitors — those who are intimidated or less interested in the minutiae of car specs.

In building the personas, the key characteristics identified in surveys and interviews that defined the persona were the type of car (new versus used) and the emphasis on price versus style.

One primary persona, Bill, is 34, single, and possesses strong opinions about car brands and style. He’s less interested in other people’s ratings and loves to research cars whenever he has time. His defining quote is: “I have to have a new car every year.” An abbreviated version of the Bill persona is shown in Figure 5-1.

In contrast, a secondary persona is represented by Sam (see Figure 5-2). Sam doesn’t like researching cars, and this is the first one she’s buying on her own. She’s just out of college and is looking for a good deal on a used car. She wants something reliable but fun. Her defining quote is: “Buying cars is intimidating but I have friends I can ask.”

Having the relevant information about primary personas like Bill means the designers should be sure car enthusiasts can get to all the technical specs they want on the website. The designers balance the Bills with the Sams, who will visit the website to look for basic information like features and reviews. Sam isn’t interested in technical jargon, so the designers have summary information and graphics to denote high and low gas mileage vehicles and price, and display the rates prominently on the car detail page.

Figure 5-1: An example of data included in a persona for an automotive website for a car enthusiast, a primary persona named Bill.

Figure 5-2: An example of data included in a persona for an automotive website for a first-time car buyer named Sam.

It’s common to use stock photography for personas (as in this chapter). While it’s fine to use actual customer pictures, it can be difficult based on customer privacy concerns. Give the persona a face, but don’t be too concerned if it’s a model rather than a real customer.

Getting More Personal with Customer Data

A few simple steps are required when you need to build an effective persona.

Step 1: Collecting the appropriate data

The most important rule in building a persona is that the information you use for it should not be made up or based on opinion. Base your persona on actual data. Collecting, aggregating and analyzing customer data from customer databases and social media helps provide input into a data-driven persona process.

That said, although you want customers to like your product, the best data to start with is not obtained by asking your customers what they like or don’t like.

Personas concentrate on what a user does, what frustrates him, and what makes life a bit easier. A good persona is a narrative that describes a person’s typical day and experiences, as well as skills, attitude, background, environment, and goals. Personas identify the motivations, expectations, aspirations, and behaviors common to a large segment of customers.

Start with demographic data from a customer segmentation analysis: gender, age, industry, how frequently they use the product, and so forth. This tells you a lot about who your customers are; however, it rarely includes information about motivations and goals, which are essential in building a persona.

To understand what your customers are trying to do, use the following methods.

Conducting a new survey

Surveys are one of the most cost-effective ways of gathering data from lots of customers. You can survey existing customers from customer lists or prospective customers by using panel agencies that recruit qualified customers or prospective customers. In addition to basic demographics, ask questions about motivations, goals, and challenges.

See Chapter 13 for ideas on conducting a top-tasks analysis that gets to the primary things customers really want to do with your product.

Interviewing customers

Interview customers to get a better understanding of motivations, goals, and challenges. You can conduct the interviews in person or over the phone.

Don’t cold call customers. When collecting customer information, be sure a customer opts-in to being contacted. Allow customers to opt-out of being contacted again if they wish.

Avoid focus groups if possible. Unless your product or service is used by groups of people at the same time, conduct one-on-one interviews. Group sessions can easily be influenced by personalities and are often best for generating new ideas to test later. In the interviews, have customers talk about their problems, frustrations, and goals.

Observing

Watch users as they use your product, a competitor’s product, or just try and solve problems on their own. This is often referred to as ethnographic research or field studies. In observing, you can see things that are hard for customers to articulate.

For example, while I worked at Intuit, we watched customers as they entered their information into point-of-sale cash registers, then exported their sales at the end of the day into QuickBooks. The process of exporting and importing was both time consuming and error-prone. This frustrating process inspired the Intuit design teams to come up with ideas about how to make balancing the books easier.

Usability testing

Have users attempt common tasks with your product or service. Don’t ask users what they like or dislike; instead, watch them interact with it and look for pain points and opportunities to improve your offering. (See Chapter 14.)

Secondary research

If interviews cannot be conducted or surveys commissioned, partial second-hand information can be obtained from within your company: in sales, marketing, product, customer support, and tech support. While it’s not ideal to rely on secondhand information, personas from secondhand data are usually better than not building personas at all.

Use a mix of methods, both surveys and interviews, and minimize opinions and guesses. You’ll often notice a behavior in an interview or field study and then you can look to understand its prevalence by conducting a follow-up survey to a larger set of customers.

Avoid asking for opinions and concentrate on motivators. Many times, users do not know exactly what they need, but they do think about what they want.

While there is an art and science to writing good survey and interview questions, in general you don’t want to lead customers down a predetermined path. Avoid questions that suggest an answer or have simple yes or no questions. For example, “Would you like this product to be easier to use?” is not the best question because in most cases customers will say yes. A better question would be, “If you could fix only one thing about the product, what would it be?”

Like building customer segments, just because you have data (like detailed demographics) doesn’t mean it will be relevant to constructing a persona.

Step 2: Dividing data

After gathering data about your customers, look for themes and patterns that are relevant and based on behaviors and goals. Categorize open-ended comments or actions observed in field studies with other similar comments and actions to identify patterns. In particular, look for patterns around:

· What they expect to accomplish with your product

· How they go about achieving their goals

· What their technical background or product proficiency is

At this point, it is useful to brainstorm:

· Do you need more information?

· Can you group the data in other ways?

For example, we were conducting research on the major email providers such as Gmail, Yahoo!, and Hotmail. After surveying 1,000 users, we found repetitive comments about how Yahoo! changed its interface too often. This would be a good candidate for understanding how often Yahoo! actually changes its interface and what type of customer is averse to the changes.

Step 3: Identifying and refining personas

From the most important groups identified in Step 2, define an initial set of primary personas as described in the earlier automotive website example. These will constitute your “set of characters” the product or marketing teams should focus on.

For each persona, create a realistic portrait.

· Create an identity with details that make the persona more real:

· Name

· Picture

· Personal history

· Create a portrait with data that connects a persona to a customer segment:

· Goals

· Habits

· Expectations

· Skills

· Attitudes

· Motivations

· Environment

Some personas in your set of characters are not among your target customers. In fact, some of them could be examples of whom you are not designing or marketing for.

Don’t overdo the persona process. Although personas provide many cost- and time-saving benefits, there is a diminishing return if you spend too much time and effort developing them. Don’t get too bogged down in documenting every detail or spend too much time creating bloated documents. If development teams and marketers are faced with too much information, it will get ignored.

Creating a good persona requires composing both qualitative and quantitative research, but it should be presented in no more than about one or two pages. It is not a job description. Stay away from minutiae like tasks, duties, and responsibilities. Concentrate on skills, attitudes, motivations, environment, and goals.

Follow these guidelines as you develop a persona:

· Use multiple sources of data: Segmentation, interviews, surveys, and observations.

· Get buy-in: Interview stakeholders from sales, product development, product support, and marketing to look for agreement and insight.

· Don’t base it on one customer and his idiosyncrasies.

· Distinguish types of data:

· Data that connects a persona to a customer segment (for example, proficiency level with your software)

· Data that only makes the persona more real to you (for example, his name is Andy MacPherson)