Driver persona

Why even bother ourselves with personas?

Since the first introduction meetings and kick-offs till now, there has been one solid question that almost all stakeholders have been asking us; “Who are our users?”

We all are concerned that Snapp has more than 2.5 million drivers from the smallest villages and cities to the most populated ones, And the question “Who are our drivers?” contains numerous other questions such as;

• Who am I designing for?

• What kind of features and benefits do our users need?

• What do they expect from our product?

• What are the pain points of our product?

• How could we satisfy them with our services?

To answer the questions above, massive mixed-method research was needed to dive deep into our users' attitudes and behaviors. Therefore, we decided to run a user persona project to find out our fictional leading role users.

Introduction & Problem statement

Why PERSONA?

Promoting a product without knowing who your target audience is or what your target audience wants is an impossible task. You’ll just be making decisions based on what you think they want. That’s not sustainable over the life of any brand.

User personas are now widely recognized as an integral part of the user experience. Everyone involved in each stage of the development process should be invested in the personas, from stakeholders and designers to developers.

However, creating user personas can be time-consuming and requires some good old-fashioned rolling up your sleeves and getting your hands dirty with empirical data. Choosing a design that not only complements our hard work but helps us get the most out of it is no trivial task either.

Here in Snapp!, after lots of ups & downs, we decided to create our driver persona to serve the vast majority of our users (drivers). We all are concerned that Snapp has more than 2.5 million drivers from the smallest villages and cities to the most populated ones. Hence, creating a persona is not a workspace decoration – it’s something that will be used in every step of the design and development process. One way of keeping everyone on the same page is to involve the end-users of our persona in its very creation and to continue to request their feedback throughout the persona design process. That’s why creating user personas is vital for our company.

To explore the character behind each persona, we need to dig deep into drivers' mindsets, beliefs, culture, personality, and behavior. And one of the most beneficial research methods leading to our exploration is “individual interviewing” and "thematic analysis" to analyze the qualitative data.

Methodology

What are user personas, and how do we find out about them?

User personas are archetypical users to represent more extensive groups of users with typical behavioral and attitudinal traits, needs, goals, pain points, etc. For example, hundreds of kings fighting for their people with no greed for titles who would sacrifice their lives for justice could be in the category of “Jon Snow” user (or better to say, king!) persona.

To explore the character behind each persona, we need to dig deep into users' mindsets, beliefs, culture, personalities, and behavior. And one of the most beneficial research methods leading to our exploration is “individual interviewing” and "thematic analysis" to analyze the qualitative data.

Research process

Segmentation → We segmented the users based on RFM, which will be explained in the following, and collected phone numbers afterward.

Recruiting → With the help of our fantastic outbound team in the call center, we recruited users for the interview.

Interviewing → For each segment, we started with interviewing six users who had to be extended in case of not find patterns among users’ attitudes.

Analyzing → After each segment was interviewed, we ran an affinity diagram workshop to find user speech clues and patterns.

Delivering → Concluding the available data into a particular user persona.

Segmentation

Due to the number of our driver users and available data, dividing the users based on RFM (Recency, Frequency, and Monetary) seemed the most accurate way to have all our users segmented. Another leading factor in segmenting process was the city, which we chose to have users of large cities (Tehran, Karaj, Mashhad, Isfahan, and Shiraz) for this phase of the project. At last, we had seven user groups:

• Champion

• Loyal Customer

• Potential Loyalist

• Customer Needing Attention

• Hibernating

• About to Sleep

• At-Risk

→ To know more about RFM customer segmentation: https://www.putler.com/rfm-analysis/

To find out user attitudinal patterns, we started to have at least six interviews for each segment, three from Tehran and Karaj and three from Isfahan, Shiraz, and Mashhad. And in case of not find valuable patterns and themes, we would increase the number of interviewees.

Recruiting & interview

The critical role to discover users' perspectives, feelings, needs, frustrations, and more is the questions we ask them. In this project, we iterated the question from time to time to have more concentrated answers.

Question categories:

• Demographics

• Personality

• Occupation

• Habits & interests

• Online Behavior

• Goals & motivations

• Obstacles & frustrations

• Snapp! Behavior
  

→ To see the full questions of our interviews: Persona Questions.

For each interview, the outbound team of the call center recruited two users for us always to have a backup interviewee if the other user canceled the session. And each interviewee’s Snapp! The account was charged 150,000 Toman as an incentive.

Data analysis

Familiarize with the Data

The first step is all about getting to know the data. We Go through all the data we have collected and take notes on everything that happens or is said. Also, audio or video recordings, then we perform some transcription, which will allow us to work with and incorporate this form of data into our analysis.

Generate Initial Codes

A code is a brief description or title for what is being said or done, so, each time you note something interesting in our data, we write down a code—e.g., “uses android phone” or “family person.” Like keywords or tags, each code makes it easy to sift through the data later.

We use colored stickers to indicate which code each piece of data refers to because we’ll need to collate all the data into coded groups once the coding is complete. Code for as many potentially exciting themes as possible and keep a little of the data surrounding each code to ensure we don’t lose too much context. We used a digital spreadsheet (invisionapp) to keep track of your data, codes, and themes.

Search for Themes

In this step, we start to sort our codes into themes. Whereas codes identify interesting information in our data, themes are broader and involve active interpretation of the codes and the data.

As we search for themes, you’ll find that iteration is key—we likely want to move codes back and forth to form different themes until we find the optimal groupings. Some themes might be sub-themes to others. After several groups, other codes might seem redundant, and we place them in a temporary mixed theme.

Review the Themes

Review and refine the themes that we identified during step 3. We read through all the extracts related to the codes to explore if they support the themes and contradictions and see if they overlap. We split the theme into several themes or moved some of the codes/extracts into an existing theme where they fit better. We keep doing this until we feel we have a set of themes that are coherent and distinctive.

Define and Name Themes

During this step, name and describe each of the themes identified previously. We tried to name themes descriptive and (if possible) engaging.

Final Data | User Personas

PS: this is just a draft of the first segment and since the persona project is a huge and on going project i will update the case study after finishing each segment

Post Measurements

Follow-up Survey

Since the achieved data is based on six users for each segment, we have to validate our data by a massive survey to see whether the attitudes and needs follow a typical pattern or not.

→ With the help of our market research team, we are going to survey to dig deeper into users’ needs, expectations, motivations, and goals.

→ I will find out drivers' personality information by the Supplementary survey and fill the personality section in the persona. For now, I left this part blank.