The User Is Always Right Personas

Making Personas Work for Your Site Steve Mulder

… and so that’s the plan for redesigning the web site. We believe this strategy will increase revenue enough to pay for itself within six months. Any questions?

What happened to feature x? In our experience, feature x isn’t worth the investment. our customers don’t want it.

Hmm. That’s not what we discussed in the kick-off meeting.

Our contextual analysis revealed that even in critical use cases, feature x was not part of the typical customer’s mental model. The presence of feature x even produced cognitive dissonance among our most loyal customers base, where the opportunity score was below 0.7. the roi simply isn’t there. I still think customers would use feature x. I’d use it.

Yes, but are all of our customers just like you?

No, but I know what my customers want, and this is important.

But our users don’t need it. that’s what the analysis shows.

forget everything you know about users. we need to be innovative!

But innovation begins with first understanding what our customers really need.

are you saying I don’t understand my customers?!

No, but our analysis…

Feature x stays in scope. Okay, are we done?

Business results depend on satisfying users You are not your user Learning about users requires direct contact Knowledge about users must be actionable Decisions should be based on users

Personas bring focus

Personas build empathy

Personas encourage consensus

Personas create efficiency

Personas lead to better decisions Personas for Design Information architecture, interaction design, visual design, content development, user testing Personas for Marketing Framework for marketing campaigns, branding, messaging, market research Personas for Strategy Framework for business decisions, offerings, channel usage, features

Creating Personas

The Landscape of User Research and Testing User Interviews QUALITATIVE (INSIGHTS) GOALS & ATTITUDES (ASPIRATIONAL) BEHAVIORS (ACTUAL) User Surveys Usability Testing Site Traffic/ Log File Analysis Eye Tracking Field Studies (Contextual Inquiry) Shadow Shopping (Shop-Along) Intercepts Customer Support Data Card Sorting Focus Groups Diary/Journal Studies Participatory Design User Advisory Panel Automated Usability Testing User Reports Collages QUANTITATIVE (VALIDATION) A/B Testing

Qualitative Personas

User Interviews

Cross-section of online customers, offline customers, non-customers, other audiences

15 interviews is a good starting point

Informal, loosely structured conversations

User Interviews: Topics

History with the company

Intro to company, usage/purchase history, early impressions

Domain experience and knowledge

Domain expertise, competitors, share of wallet

Goals and behaviors

Needs/triggers for usage, typical process , channel usage, feature and content usage, gaps, wish list

Attitudes and motivators

Description of experience, likes/dislikes, influencers, psychological drivers

Opportunities

Reaction to new ideas, features, content, improvements

Observation of actual behavior (field studies, usability tests)

Segmentation Goals Behaviors Attitudes

Segmentation: Marketing vs. Personas

Marketing

Sell to people

Personas Understand how people will actually use the site Age Income Gender Other demographics Goals Behaviors Attitudes

Segmentation by Goals

What are different people trying to do on the site?

Hierarchy of Goals

Other goals:

Buy commercial property

Find an apartment

Sell a house

Be happy Be independent Buy a house Understand process Learn about points Motivator Motivator Goal Need Task

Get a mortgage loan

Get insurance

Find a moving company

Segmentation by Behaviors and Attitudes

How do users differ based on what they do or how they think?

Behaviors:

Frequency of real estate activity

Frequency of visits to the site

Channel usage for various needs

Use of competitors

Attitudes:

Knowledge about real estate

Motivators affecting users’ likelihood to buy or sell

Perception of the company/brand

Etc.

Explore different combinations

Frequency of real estate activity Knowledge about real estate The risk-taker who thinks he knows more than he actually does The novice who needs a lot of guidance The pro who wants to use site tools and doesn’t need help The smart one who wants validation of what she already knows

FedEx segmentation

Level of preparation Desired level of personal interaction Me

Segmentation: The Tests

Your segments should…

Explain key differences you’ve observed among users

Be different enough from each other

Feel like real people

Be described quickly

Cover all users

Clearly affect decision making

Qualitative Personas with Quantitative Validation

Surveys

Surveys: Who

Cross-section of users

Goal: 100+ completions per segment

17,000 emails @ 3% completion rate = 500 completions

Incentives and reminders can boost completion rate

Surveys: What

Questions to gather data on segmentation attributes (dependent variables)

Goal for visiting site (buy home, find apartment, etc.)

Frequency of site visits for each possible user goal

Importance of each possible goal to the user

Knowledge of real estate

Previous real estate experiences

User’s self-perception of real estate expertise

Questions to test the segmentation against (independent variables)

Other behaviors (site/channel usage, feature usage, etc.)

Other attitudes (toward company, about self, etc.)

New features and content to test

Surveys: What

Recommended order of questions:

Current goals, usage, and behavior, including channel usage

History with the site and company

User of or importance of existing features and content

Satisfaction with existing features and content

Importance of new features and content

Psychographic questions

Demographic questions

Site Traffic Analysis

Entry pages

Referrers

Exit pages

Common paths

Feature usage

Search terms

Conversion rate

Duration

Frequency

CRM Data

Annual spending

Frequency of purchase

Products or services purchased

Channel usage

Customer support usage

Length of time as a customer

Quantitative Nirvana: Complete User Portrait

Survey data

What the user does

Site traffic analysis

CRM data

Self-reported survey data

What the user says What the user is worth

CRM data

Self-reported survey data

Testing the Segmentation

Cross-tabs (pivot tables) in Excel

Quantitative Personas

Quantitative Segmentation

Making Personas Real

Key Differentiators

Name

Photo

www.sxc.hu www.morguefile.com www.istockphoto.com

Personal Information

Domain-Specific Information

Computer and Internet Usage

Profile

Quote

Business Objectives

Prioritization

Primary

Secondary

Unimportant

Excluded

Scenarios Francis and Michael have agreed that she’ll take charge of learning more about the home-buying process. She goes online, does a Google search for “Atlanta real estate,” and follows a link to the site’s home page. She sees that she can search for houses from the home page, so just for fun, she does a quick Atlanta search to see what kinds of houses show up. There are lots of houses in many different neighborhoods, and she easily narrows her results down to the area where she and Michael live, using a map. There are still many results, and she’s not quite sure which search options to use to narrow the search further. Then she notices a link for first-time home buyers and follows, it hoping for basic how-to information. The link takes Francis to a step-by-step tutorial that explains the whole process, and she immediately feels like she’s found the right site from which to begin her house search. She carefully reads some articles for first-time home buyers, taking notes as she reads. She bookmarks other articles she wants to go back and read later. She also comes across the site’s calculator and starts trying different combinations of numbers to find out what she and Michael can afford. She particularly likes the glossary of terms so that she can finally figure out what “points” are and learn more about different types of mortgages. After an hour and a half of reading, her brain is full, and she shuts her computer down for the day, feeling like she got an excellent start. The next day, she comes back to the site to look up information specific to Atlanta neighborhoods and finds lots of information on each. She’s able to focus on five neighborhoods that look particularly good. The fun begins that night, when she takes Michael through all that she has learned, and they set up a regular schedule for looking at online house listings.

Using Personas

Persona Document

Persona Cards

Lifesize Cutouts

Persona Cubicle

Other Ideas for Keeping Personas Alive

Posters

Tchotchkes

Day-in-the-life photos, audio diaries, etc.

Role-playing

Quizzes

Staple to documents

Email addresses

Business Strategy

Features and Functionality

Features and Functionality: Brainstorming

Francis’ Goals

Learn about the home-buying process

Find out what she can afford

Discover what areas of Atlanta are desirable

Find a house that matches her criteria

Business Objectives Visit the site often Register for email alerts and newsletters Subscribe to premium services Recommend the site to others Possible Features and Content Introductory how-to articles Glossary of jargon Videos and podcasts Success stories Q&A with experts Message board Blogs Rent vs. own calculator Local experts to call/meet Checklists to print Interactive quizzes Common mistakes Animated cartoons Book recommendations

Features and Functionality: Prioritizing

Structure

Structure: Task Analysis

Search for houses from home page

Enter location by city and state, or zip code

Enter price range

Enter number of bedrooms and/or bathrooms

Submit search

View and narrow results

Browse first page of results: photo, price, address, basic stats, and description

Click to show results on map

Click map to narrow results to one neighborhood

Browse new results

Read tutorial

Click link for first-time home buyers

Read landing page for learning area

Click teaser for step-by-step tutorial

And so on…

Structure: Use Cases

Search for Houses

Level: User goal

Primary actor: Francis the First-Time Home Buyer

Preconditions:

Francis arrives at home page or House Search landing page

Main success scenario:

Francis enters city and state or zip code.

System uses auto-completion to show options below search field as Francis types. System shows up to 10 matching city names in order of number of homes for sale in each city. Francis can click a city name to select that city.