Or, A Four Star Slap in the Face
I have a secret. After 193 rides and over three years of having a Lyft account (not that 2020 should count), I have a perfect 5.0 star rating. I dare say there are very few people in this world that have a similar record with that volume of rides. And while I am a model rider (ugh what a terrible way to start an article), this is a statistical anomaly that shouldn’t exist. One clash of personality, or one bad day—by me or the driver—and bam, a 4-star blemish on the face of my formerly flawless record. Or heaven help me, a 3-star. I dare not write further words for there are some evils that should not even be spoken lest I wake the demons of Ebert’s Inferno.
I hate the five star system. I kind of hate online aggregated reviews in general. First of all, there’s a lot of fake and/or seeded reviews. I have a really hard time trusting validity. Second, many people don’t understand the impact of their stars. In certain cases, a four star review can be devastating. It’s long been rumored that rideshare drivers will be booted from their platform (aka fired) for a sub-4.6 rating. These days, if you leave a four star rating you’re saying “this wasn’t good enough”.
Additionally, people have very different standards for what constitutes a good experience. What’s important to me for a hotel stay is definitely not in line with how other certain other people write about their hotel experiences. For me, what’s essential is a matter of preference, and those preferences change depending on my travel state: am I with family or by myself? Is it business or pleasure?
Context is everything. Like literally everything. And experiential impact is not linear. If I was to sit down and enjoy an unknown glass of wine by myself, I might like it. As I drank, if I ate food that was well-paired with that wine, I might like twice as much. If I personally knew the vintner, I might enjoy it more. If I imbibed with the company of my friend Tony, I might enjoy 10 times more. If Tony and I did this while talking with the actual winemaker, I might enjoy it even more. Or I might not, because that winemaker might be a horrible person. Of course, Tony and I would laugh about it later.
Same wine, unlimited possibilities for how I might experience it—let alone how someone else might experience it. (I really want to keep writing out more examples of what could happen with a glass of wine, but my editing team is going to cut me off. In fact, you’re probably reading a shorter version of what I actually wrote.)
Even meaning can be completely changed simply by switching contexts. Imagine Bach’s Chaconne being played by a violist live at a funeral and then juxtapose that with a vision of that soul-altering masterpiece as a soundtrack for a video of my Aussiedoodle (aptly-named Scruffy) trying to sneak a treat off the dining table. It would be funny.
So what do I do with aggregated reviews? I have to read them. It’s a tedious chore. I read some of the top 5 star reviews and some of the 1 star reviews, and even some of the in-between reviews. I inevitably ignore some of them due to thinking they’re fake or because I can tell that I am not the same kind of person as the reviewer. Then I Google the category/product to see what independent “experts” are saying (or since our Jason L. Baptiste podcast I’ll sometimes head straight to the Wirecutter to see if they have any comments). If I’m really looking for some unfiltered feedback, I’ll dare to search “*product name* reddit” to see what people are saying in the more ruthless parts of the web.
I prefer informed opinions, because they’re more likely to represent the experience that I will actually have with the product. More often than not, I still find myself tossing aside an expert review because the context by which they’re experiencing the product is not particularly relevant to mine.
I am a phenomenologist. I have to admit my curmudgeonly rantings about reviews were a portal towards phenomenology. For the uninitiated, while my Lyft rating is phenomen-al, phenomenal-ism isn’t the philosophy of what’s good, it’s a philosophy and study of phenomena—or put plainly, experiences as described in the first person. Reviews are a wonderful example of phenomenological data in our industry.
We’ve used phenomenological data for a long time in our industry. Focus groups, customer service feedback, reviews, social, and more.
What’s surprising is how leadership deprioritizes subjective, phenomenological data in favor of other types of data. Historical performance, dwell time, A/B and multivariate experiments, behavior analytics platforms of all kinds (GA being the most pervasive) drive our budgets and decisions. The decisions we make are based on aggregated data and we assume that because people did something they’ll repeat their behavior. Hypotheses in search of supportive data are, unto themselves, phenomenological. These are what I would call epistemological and logical strategies—the realm of the known and consequential.
I’m not disparaging using these techniques all together. Brands use them effectively every day to make smart decisions. The challenge with these types of investments is that it’s often difficult for us to quantify the “why” on successful decisions. We pick test A over test B because more people converted on test A. The “why” often ends up being a story that we infer.
We often resort to “hard” data because it’s difficult to quantify feedback and first person narrative.
Think about it. Brands provide imperfect communications about imperfectly designed and imperfectly manufactured products and services, which are imperfectly interpreted and imperfectly interacted with by imperfect customers who imperfectly communicate back their imperfect experiences through very imperfect means and through imperfect modes such as words, behavior, and (if they are so lucky to have means to collect them) pitch, tone, and body language. All the while in different contexts over time which have boundless opportunity to impact experience.
This makes interpreting and employing first-party data challenging, which forces us to try to simplify it by employing epistemological and logical analysis to our phenomenological data. Stars, word clouds, common requests, user profiles etc—these are ways that we “measure” the narratives we ingest. Again, these things can be helpful, but what we’re really doing is finding ways to convert experiences into quantifiable knowledge, instead of gathering true phenomenological insight.
Quantifiable knowledge is useful for many of the same reasons that Google Analytics is useful. You can aggregate it and make logical decisions based on what the data is telling you.
Phenomenological insights are deeply personal and peculiar revelations that require empathy to interpret to be useful. They also require discerning factors to be meaningful to a brand.
Can we claim to serve the “experience economy” and not be serious about judging success based on individual experience?
The phenomenological data we collect needs to have a way for us to recognize its relevance. Think back to my thought process regarding reviews—I have to skip a lot of them in order to find reviews that are actually helpful. You’ll need a way to sort first person feedback to understand the individual’s circumstances and context, their relationship to the category and your brand, personality, and other factors.
That’s a lot of data to ascertain in a broad or open context. It doesn’t lend itself very well to open reviews, social, or broad customer service feedback (let alone how these channels attract certain personalities). On the flip side, the traditional gated data collection process of focus groups is so far removed from the context of real life it’s hard to confirm broad results.
You’re going to need deep relationships to garner that kind of openly provided narrative and detail. Building relationships requires time and trust. This may require carving out new roles or hiring new talent. Community managers, psychologists, people who are people people. People you can trust to be able to understand what experiences are important for your brand to take action on. They’ll need to understand your brand as well as they understand your customers. They may need to create guidelines or have markers to help delineate when experiences should be taken seriously. They’ll need discernment, wisdom, and depending on the scale of your brand, you’ll need a group who can understand diverse perspectives.
You’ll also need experiential markers of what constitutes success. Questions such as “what is the intended experience of the product and are people experiencing that intent?” and “how do we capture and realize unintended positive experiences?” are good questions to ask when setting these markers. You’ll need to give these markers business legitimacy by including them in KPIs (so that you invest in them).
Two examples of brands that have done this well come to mind: Arfa and Lively.
Arfa’s entire business model is based around creating personal care brands for underserved communities through intense intimate conversations—usually in person meetups at coffee shops or homes—to determine wants and needs of diverse people groups with similar skin care concerns. It’s an empathetic model where truth, vulnerability, acknowledgement, and trust are all cornerstones in bringing people together to share their ideas and experiences with each other to help create real products that solve real problems.
Arfa has started with two brands: Hiki sweat products and State Of, symptom-specific skincare created with perimenopausal and menopausal women. HIKI are in bright, colorful packaging, making being sweaty less embarrassing. The header in their website says the products were “Created, tested and developed by and for people courageous enough to tell us about their butt sweat.” It takes a lot to get people to talk about their butt sweat to strangers, but co-founder and CEO, Ariel Wengroff, says now when they meet virtually during the pandemic, they’ll start off meetups by encouraging everyone to share how their mood is that day on a scale of 1-10 to break the ice, and continue to encourage that same vulnerability. It’s a much more informed and well-rounded way to build “data.”
In a similar manner, Lively is a lingerie brand that is approaching “sexy” differently than it has previously been done in the industry. In their words, the brand is “inspired by women with wild hearts and boss brains.” By directly reaching out to targeted customers and women, they built a community that was interested in sharing their experiences. When we spoke with founder Michelle Cordeiro Grant at Shoptalk 2019, we were impressed by this community-driven approach. Lively built a community by listening to what their customers were telling them and expanding their product line based on direct feedback with their sights set on acquiring customers away from Victoria’s Secret.
By giving legitimacy to narrative, creating an environment to collect the phenomenological data, and being thoughtful about which data is impactful for your business, you’ll be able to avoid the pitfall of investing in previously successful initiatives without understanding why they were successful. With the pace of change in our world, repeating the same initiative is less likely to achieve the same results. Understanding context and “why” through contextual phenomena will open your playbook up to empathetic initiatives that meet your customers where they are at.
If you’ve made it this far and you’re looking for actionable next-steps—might I suggest leaving a rating of Future Commerce on Apple Podcasts? No less than five stars, please.