Crafting Compelling Personas: Learnings from Verizon Connect in Atlanta
Introduction:
A few weeks into my tenure as a UX Researcher at Verizon Connect in Atlanta, I was given the task of leading research towards developing personas for our line of Hum products. At that time, I was very new to Verizon Connect. I did not yet have a holistic understanding of the range of Hum products, their particular development histories within the company, nor was I privy to the associated affective residues that such development processes inevitably generate within people who are involved in them. Indeed, my involvement in the persona project was going to lend a fresh perspective to all of these elements. As an ‘outsider’ in the company, I started with a few basic questions — what did Hum mean for the people who have been involved in its design and production? How did various people involved in the design, management, and production of Hum products imagine Hum users? How did the existing assumptions about Hum users compare with data about Hum users that had been generated by the research team over the years?
Archival research:
Considering I did not know much about Hum, I spent the first part of the persona project digging into the research archives and combing the data, analysis, and insights that had been generated by our brilliant research team over the last few years. These materials included a wide range of qualitative and quantitative studies that encompassed interview transcripts, audio, video, and analyses; field visit data; survey questionnaires, data, and analyses; market segmentation information; marketing data sheets; Kano studies; top task analysis, etc.). While this was a particularly tedious and slow process, each of these studies revealed, in a piecemeal fashion, specific insights about Hum products and users’ expectations and experiences of them. Much like an archival research project that historians routinely engage in, this archival research allowed me to collect, organize, and synthesize valuable bits of insights about Hum, the human motivations involved in its production, as well as its users across a valuable span of time. This process allowed me to re-purpose previous research findings (that were often only tangentially relevant) towards the persona project. I was also able to identify certain areas that may require more focused research efforts in the future. This was a crucial first step in the process of assembling the materials that I would eventually use to stitch together narratives about Hum personas.
Internal Ethnography:
While the materials gleaned from the research archives provided me with glimpses of insights about Hum products and its users, there was little material in the archives about how the people in the office were themselves thinking about Hum users. In order to contextualize what I had learned through the research archives, I teamed up with Shannon, the lead designer on this project, and began meeting with and conducting ‘informal interviews’ with a range of experienced people at the office. I specifically picked people who had been at the office for a long period of time, those who represented various lines of work on Hum (e.g. product management, marketing, customer service, research), and those who had witnessed the evolution of Hum products through various iterations, product cycles, and internal restructurings. These conversations were tremendously helpful for us to situate specific aspects of Hum products within the internal geographies of Verizon Connect, as well as Verizon Wireless more broadly. Considering the significant reconfigurations of people (and companies) working on Hum products over time, the internal terrain that supported the development of Hum products was quite complicated. And, did I mention, we were having these conversations amidst on-going lean transformations within Hum product teams? Despite such complexities, my ethnographic perch and conversations with experienced Hum team members provided me with a broad backdrop against which to make sense of the details I had gathered from the research archives.
A particularly telling aspect of these conversations was that our interlocutors inevitably held strong beliefs about who Hum product users were (or ought to be). People often shared that they felt that others’ views about Hum users were biased and misinformed, despite them sharing similar assumptions. To collect these insights, we were careful about framing these inquires as non-evaluative exercises. Contrary to my expectations of finding a wide range of assumptions about Hum users across the various teams and departments within Verizon Connect, we found that most people shared a range of assumptions about Hum users. Yet, there were distinct variations within these shared ideas across departments (product, customer service, and marketing) that drew upon the particular routines of each of their daily tasks — much like the ways in which rock shapes indicate dominant wind directions.
Assumptions in circulation:
I had thus far taken stock of existing research materials and, with Shannon’s help, we had mapped those insights onto the general internal, chronological geographies of Verizon Connect. Having spoken to key persons from product, marketing, and customer service, we had unearthed some general assumptions of Hum users across the office. To add depth to our understanding of these assumptions, Shannon and I designed a series of workshops with various product teams (which were, at the time, being reconfigured into lean product clans) urging as many people to participate as we could get from each clan.
These workshops were designed around a series of ‘active imagination’ exercises in which we asked the participants to close their eyes and imagine Hum users in as much vivid detail as possible. These ‘active imagination’ exercises were meant to provoke lucid motion pictures of Hum users in highly detailed environments that our participants could observe through their minds’ eyes. After brief movements of exploring these details through their imaginations, we asked the participants to record relevant details in colorful post-its. These post-its (which have come to symbolically represent UX research), with specific details from our participants’ imaginations of Hum users, were then collected and organized under specific prompts — What do they look like? Where is the person? What are their homes and families like? We implored our participants to not spare any details, and, as a result, we were able to collect a wealth of information that revealed specific patterns of assumptions about Hum users that were in active circulation within and across Hum clans. We conducted a total of 6 such workshops — one with each product clan. Once again, while there were obvious variations in details, we were struck by the strong patterns of shared assumptions about Hum users across the product teams. Shannon led the ‘cleaning’ of this data and Seth, our generous data scientist, helped us synthesize these findings into attractive word-clouds. We presented this to the office during our last Friday Research Readout session.
Social and political philosophy has long held that the people’s assumptions about the world (whether they are consciously or unconsciously held) yield productive outcomes, in that these assumptions manifest themselves through concrete, often measurable, outcomes. Similarly, we could extend this reasoning to understand that assumptions about Hum users, whether they are consciously or unconsciously held, manifest themselves through particular behavioral, or task-oriented functions however technical (developers) or managerial (leadership) one’s responsibilities may be. This understanding reinforces the significance of personas for products like Hum. Personas provide concrete ways to mediate and moderate these internalized narratives about users by a) recognizing the existence of these narratives, b) acknowledging the power of narratives (assumptions, stories) about users to shape product design and details, and c) harnessing the power of narratives/ story-telling by providing publicly circulated, research driven, and socially compelling convergent points to align various nodes of existing assumptions, stories, and narratives about product users that are already in circulation. In a way, personas may be thought of as tools that can help reign in and re-shape disparate assumptions about users into shared collective narratives about users to effect better alignment within and across teams, and to ensure that our users (who are specifically represented by personas) are consistently foregrounded and referred to during each phase of design and production.
Research documents
Following an innovative suggestion by Tamara Adlin and John Pruitt (who incidentally works with my brother at Dell in Austin) in their widely acclaimed book (The Essential Persona Lifecycle), I began to create research documents for each of the personas that will potentially be ‘birthed’ during the appropriate phase of our project. These research documents contain rich biographical details of each persona. Most of these biographical details have been drawn from and referenced to existing (or ongoing) research projects. To aid the flow of the biographical accounts and to ‘breathe life’ into the respective ‘skeleton personas’, I fused additional details (some of which are drawn from the accounts of interview participants, while others are fictional) to the data-driven biographical outlines. Drawing inspiration from Alan Cooper’s now-classic work, The Inmates Are Running the Asylum, this gluing together of data-driven nuggets with compelling life-stories of people has allowed for these biographical outlines to remain precise without necessarily being dogmatic about accuracy at all points. This has allowed for the personas to come to life.
My objective was two-fold: on the one hand, I wanted to create skeleton personas that were rigorously linked to and validated by research. On the other hand, I wanted the persona biographical outlines tell compelling stories about people and their lives. This necessitated a balanced fusion of fictional story-telling with validated research. Neither of these components would be effective as persona building materials by themselves. On the one hand, persona outlines that are too fictitious may be questioned, discredited, and rejected by others (why is your story more compelling than mine?). On the other hand, considering our circumstances did not allow for primary ethnographic research with actual users, persona profiles drawn from research insights obtained from discrete research tasks (asks) would, at best, conjure dismembered skeleton figures, hanging by their parts — far from the wholesome, compelling personalities that are required for effective personas.
Drafting these research documents required me to draw deeply on the insights I had collected and organized from the research archives, details I had picked up from conversations with key persons across the office, relevant shared assumptions about users we had gathered from cross-clan workshops, and my incessant questions to our research team members about particular aspects of previous research projects. Given that there were on-going product initiatives that were working on pressing timeframes, we also needed to synchronize our efforts and findings to such parallel projects. Indeed, these research documents are an amalgam of fact and fiction, short term product imperatives and longer-term company visions, and sprinkles of imaginative details that have been negotiated and re-cast through additional review workshops with product managers across product clans.
These research documents are repositories of information and insights for designers to refer to as they run design drafts in synchronous processes. Indeed, Shannon, our lead designer on this project, has been able to refer to these research documents as stand-alone sources of information as she crafts together iterations of various formats of design drafts through which the stories contained within the research documents will be re-told.
Persona crafting as diplomatic missions
Leading the research on this persona project has exposed me to a very wide range of people across the entire width of the office in a very short time. On the one hand, this exposure has been enormously helpful in aiding my expedited orientation in my new workplace. On the other hand, understanding the range of interests, assumptions, and motivations each person carries has presented a daunting challenge — how to negotiate across differences and to find paths of least resistance in a fast-paced, rapidly shifting office environment? Indeed, from the very beginning, Cheryl (my research manager) and I have shared a vision of the persona project — it is as much of a diplomatic mission as it may actually be a ‘research’ project.
Given there were on-going needs to define users and to create working personas for pressing product development initiatives, Shannon and I discovered there was a parallel persona project that was already in flight when we started — albeit narrower in scope than what we had set out to accomplish. Yet, given that people had already begun attaching names to photographs and attributing general characteristics to these parallel personas, we were compelled to take stock of these and craft strategies to meld our project to this initiative without appearing over-bearing or evaluative. Towards this, we scheduled one-on-one discussions with key members of this initiative. In these discussions, we offered our help (e.g. to validate their working personas with research) while at the same time imploring them to lend their expertise, experience, and leadership roles to help us in our project. These conversations provided us in-roads to pick up on the work that had already been done by the members of this initiative. We dove-tailed out efforts by prioritizing the personas they had started to build, providing research validation (through the research documents) for the skeletons they had assembled, and by fine-tuning and detailing compelling narratives while retaining as much of the previous work as possible.
As a new researcher in a big company tasked with such a high visibility project, I have been particularly keen to find diplomatic solutions to some nagging problems. As, Elizabeth Goodman, Mike Kuniavsky, and Andrea Moed in their seminal work “Observing the User Experience: a practitioner’s guide to user research” have rightly declared, personas are controversial by nature. This goes to the earlier point about individually held entrenched assumptions about users that personas can seem to challenge or contradict. Thus, I have been particularly cognizant about framing persona workshops and its constituent activities as essentially non-evaluative (your understanding is not better or worse than anyone else’s) but generative (our assumptions put together generative a useful broader picture) exercises. Yet, given my commitment to draw as much as possible from the tremendous insights generated by our research team and captured in the research archives, the research documents for corresponding personas may also be tactically maneuvered as assertive documents.
In a recent conversation, our UX Research Director Donal (in Dublin, Ireland) suggested precisely this — that personas ought to be able to flow through the cracks and into everyday discourse seamlessly (the diplomatic mission), yet, they must also be assertive if needed and provide concrete and compelling guide-posts for converging stories about users to align with broader product, business, and company goals (the research project).
Where are we now?
Having completed research documents for all of the possible personas, Shannon and I are now in the process of reviewing these outlines (in textured detail) with product managers across all teams. We have designed our workshops to elicit comments and suggestions (“can be improved” & “suggestions to improve”) for each of the persona research documents. My aim from these workshops is to generate a sense of ownership amongst the product managers. The workshops require them to spend time working through the materials, sculpting the research documents, and, where needed, customizing fictitious details within the research documents with narrative embellishments of their own. While it has been a challenge to find time to schedule these workshops (given people’s already busy schedules), we have been open to scheduling one-on-one sessions to ensure that each product manager and design lead is able to provide feedback on the research documents and draft designs that we can then incorporate into the personas that will hopefully be rolled out at the end of the month.
Most recently, the person project has entered the “hyping up” phase where Shannon and I have been taking every opportunity to talk about personas. This hype will be useful to provoke curiosity, to amplify anticipation, and to circulate conversations about personas as ‘common talk’ before the finalized versions are actually ‘birthed’ and put into general circulation.
Some learnings and challenges
The first thing I have learnt is that building personas takes time. Although we did not conduct any external primary research for this persona project, it has taken us about five weeks to get to where we are at the moment. Considering this project has involved working across various clans, and with key persons from various departments in a fast-paced office environment, it has been particularly challenging to get suitable times on people’s calendars for workshops, review sessions, or one-on-one discussions. Moreover, given the wide-ranging lean transformations that have been on-going since the beginning of this project, Shannon and I have had to work on rapidly shifting grounds, always juggling multiple projects, and working with often brimming calendars of our own. Luckily, we had the foresight to provide ourselves a 4-week cushion when we initially drafted our project plans.
Persona building requires effective coordination and communication across multiple channels. As mentioned earlier, our person project has encompassed stakeholders from wide ranging departments at Verizon Connect. This has meant that we have had to coordinate schedules for discussions, workshops, and follow-ups with many people across the office. Given the multiple fronts of work (both on the persona project and other ongoing projects), it has been challenging at times to keep track of all the moving parts, and at times even to ensure full coordination between Shannon and I. Indeed, to make sure we are on the same page as the project climaxes, we have had to set up regular weekly meetings (Monday and Friday) to give us a chance to exchange notes, to catch up on our individual efforts, and to make plans for the coming weeks. I have also begun to put workshops on peoples’ calendars well in advance to make sure we can get their time, and to make accommodations for alternate workshop sessions or one-on-one reviews with key stakeholders.
I look forward to the coming weeks as we build on what we have done thus far and to get the personas (research materials and designs) ready for final reviews and circulation. We recently designed a workshop (which Shannon led) to generate ideas from our UX group about effective (and fun) strategies for the ‘birthing’ process. I look forward to sharing these experiences and additional learnings along the way with you in a subsequent post. In the meanwhile, I will leave you with one of my favorite analogies for understanding personas:
“Think of a town meeting. Each neighborhood might send a single representative who stands in for everyone else in the neighborhood, even though that one person cannot accurately communicate the particular demographics, attitudes, needs, and desires of every one of his or her neighbors. Instead, the representative communicates the essence of all his or her neighbors’ needs… personas will represent …data in the same way that a single neighbor can represent an entire neighborhood” (from “The Essential persona Lifecycle”, by Tamara Adlin and John Pruitt, page 64).
References:
“The Persona Lifecycle: Keeping People in Mind Throughout Product Design”, by John Pruitt and Tamara Adlin. Morgan and Kaufmann Publishers, 2006.
“The Essential Persona Lifecycle: Your Guide to Building and Using Personas”, by Tamara Adlin and John Pruitt. Morgan and Kaufmann Publishers, 2010.
“The Inmates Are Running the Asylum: Why High-tech Products Drive Us Crazy and How to Restore the Sanity”, by Alan Cooper. Sams Publishing, 2004.
“Observing the User Experience: A Practitioner’s Guide to User Research”, by Elizabeth Goodman, Mike Kuniasvsky, and Adrea Moed. Morgan and Kaufmann Publishers, 2012.
— — — — — -
Sujit is an anthropologist and a Senior UX Researcher at Verizon Connect in Atlanta. Sujit is an alumni of Emory University where he completed his PhD in Anthropology. You can reach him at sujit.shresth@gmail.com
