Infusing Meaning to Find New Paths

The teams I have studied range from seed-stage, where there’s usually just an idea, to early-stage, functioning startups, to more successful ones that are moving into growth-stage. The majority of them come from the perspective of having an initial idea or a domain or market in which they want to solve a problem. These typically fall into the quadrant that is more focused on diffusion and incremental innovation—the space where Lean is dominant. A smaller subset have created a new technology, often in school, and want to find a market for it. These tend to fall into the vision-driven startup quadrant, in which there is a more radical innovation, but diffusion tends to be limited. Both of these groups face challenges to realizing disruptive innovation with their products, but, as discussed, these categories are fluid, and with the right tools and methods, startups can focus on moving toward more radical innovation… or toward more widespread diffusion (Figure 2). And both of these can be done through developing a broader understanding of embedded meanings and focusing on how new meanings can be developed.

Developing Meaningful Solutions – While all startups initially begin with, at minimum, a general product idea, most have done little, if any, research into the context of the potential user. Some founders come from backgrounds in which they have market knowledge or domain expertise, giving them some level of insight. But in most startups, the conception of research is barely on their radar, and that which is, is rooted in Lean methodology. That is, it is focused on validating the idea with early adopters, developing an MVP, and testing it, not learning about user needs as a starting point or for greater context. “Social proof” –what other people think is correct– is a major goal, rather than really understanding the user. This is an issue rooted in the larger, Lean structure as well as the funding mechanisms. There is a lack of focus in Lean on doing any sort of in depth research. And where some level of research activity is found, it is often boiled down to understanding whether and how early adopters use a product. These studies are typically conducted through reductionist tools, such as Crazyegg heat maps; Mixpanel, and Heap for cohort analysis; tools for funnel analytics; and A/B testing platforms, like Optimizely and Google Analytics. These are often supplemented by some interviews with actual users (early adopters) or cold calling potential would-be users. Beyond those forms, doing research with users and interpreting meaning from users’ perspectives is something generally left out of the picture. And while some of these products diffuse and have solid metrics, there’s often no understanding of why.

One such example of this is Molome, a Thailand-based team I followed at JFDI, an accelerator in Singapore. Molome’s product was initially a sticker app for photos that had grown viral globally, in the days before apps with similar capabilities, like Snapchat. They had incredible signup numbers—in the millions—and were incredibly successful not only in Thailand, but also in Indonesia and in Brazil. But they didn’t know why. The Lean process led them down a path of experimentation… but it never helped them gain any enlightened understanding of what was meaningful about the product to users. They ultimately shifted their product to a meme-generator and folded soon after.

On the flip side this conundrum of having successful diffusion and not knowing why is the me-too product conundrum. For every successful idea, there are scores of copycat products. And, often those copycats are taking an idea that has been successful in one place, like Silicon Valley, and trying to replicate in another region. Aventones, a Chilean startup I followed at NXTP, an accelerator in Buenos Aires, is a good example. Aventones was an online car-pooling system much in the same vein as the original concepts from Uber and Lyft’s predecessor, Zimride. While following the model of some quite successful, disruptive startups, they did not find success themselves. What seems to be lacking was an understanding of how the introduction of this model might present a meaningful change in Chileans lives, where public transportation is safe, efficient, and relatively cheap. And, in following from that, Lean methods didn’t help illuminate how the model might best be adapted to be meaningfully different there.

It seems that particularly in seed-stage startups, which are seeking product-market fit, elements of an ethnographic approach could enable a greater localized understanding of the context or problem space in which a product is being created. As a methodology with a very holistic approach, ethnography seems poised to help new founders gain a deeper view into the context surrounding the potential or nascent product. It would also open up the rigid experimental processes of Lean to more interpretation. Ethnography is an iterative and reflective process that focuses on open-ended questions, as opposed to the fairly rigid, controlled experimental process of Lean. At the front end, it would help in exploring problem spaces, before making hypotheses about them. And as a product develops, ethnographic data provides a deeper level of data on which to base decisions. This need not eclipse the types of data provided by Lean methods. Ethnography embraces both the qualitative and quantitative forms of data. It adds a richer layer of perspective through which to analyze and interpret the quantitative findings. Observing people in all the richness of the context in which they are embedded would not only serve to more thoroughly understand how a product idea might solve any sort of problem, but moreover, what problems there are, how they really matter to people, and, importantly how meaningful the solution might be in people’s lives and interactions. As an approach it serves to interrogate meaning socially, and how that might inform decisions

Mapping Meaning and Interpreting Markets – For other early-stage startups, a major hurdle is that they have created a technology that solves a problem or presents an opportunity, but do not have a clear market for it, so it cannot diffuse widely. Or, they believe they have a market for it, but that market doesn’t embrace it as their vision anticipated. In these instances, the larger issue is less oriented around how meaning might be radically different as a product diffuses in different social settings, and more oriented to understanding how a product that is already radically different might create new meaning for people.

Solapa4, an Argentine team I followed at NXTP, experienced this problem. They had created an algorithm to combine geographic information system (GIS) data sources from satellites. This ostensibly provided a fuller, richer set of data on locations for decision-making in agribusiness, the field in which they felt this would have the most impact. They chose to target crop insurance providers, as they felt their product provided meaningful data to them. It turns out that it is not nearly so meaningful based on the systems of assessing risk in such firms. They have since pivoted and expanded outside of Argentina, but are still in search of making a more meaningful impact in the agricultural domain.

Another team I followed, Scrollback, at JFDI, also faced these challenges. They developed a browser-based Javascript chat tool to replace Internet Relay Chat (IRC). Scrollback is backward-compatible with the open protocol IRC of the past 20 years, which has not seen any major technology innovation through that period. The Scrollback team started with the technology and initially tried to bring on customers from universities, admission offices, online forums, other communities that presently used IRC. But this did not work— they were already used to what they were using. They failed to show how this might be meaningfully different for their users.

In both of these instances, another factor to consider is the current shift from technology-driven innovation into realms where innovation is more technology-aware. That is to say, the current startup environment is much more oriented toward taking technological advances that have already been developed, and applying them in a new domain. And selecting the right domain, and the right market is key.

Founders who have created a technological innovation without a clear user base in mind need to conduct more in-depth contextual research to identify potential product-market fit. One method of doing this is targeting different markets in the fashion of Lean, and seeing what gets traction. But immediate traction doesn’t give insight into how and why something is used– why it would be a good solution for this market. Doing this type of research has implications for the product aims and the user groups that are –or could be focused on, which could be particularly valuable for smaller, marginalized, or otherwise neglected groups. For both founders who don’t know their potential market, and even those who do believe they know their market, there is value in an ethnographic approach, which emphasizes both an etic and an emic approach—giving them a broader perspective, and also getting an situated understanding from the user perspective. For founders who don’t know their potential market, but have done some basic research into it, an emic perspective would help reveal the perspective from the point of view of the user.

Once a startup experiences some success and moves into a growth stage, they are focused on expanding and adapting an established startup product to different markets. Here, ethnographic approaches can help in adapting successful solutions or new technologies to other markets. As noted, some founders come in with local knowledge –often deep, detailed knowledge, and a perspective that drove them to solve a problem. They may have a strong emic perspective in that instance. But then, to grow, they need to gain an emic perspective in a different culture and understand how the society and culture are changing as well. Doing deeper research into other potential markets, while taking an etic perspective to the product vision overall would help to project a more meaningful analysis of the space and allow the founders to step back and reflect on their initial market goals.

Limitations of the Current Model

Resources, Access, and Skills – The potential for ethnographic research to aid in startup innovation is great, but the challenge is how startups access such resources or develop necessary skills. Startups are particularly cash-strapped and scrappy by nature, so they are unlikely to have resources to bring in dedicated researchers or consult research contractors before they are well into the growth stage. Through various networks and programs like accelerators, startups are able to learn many new skills and develop and expand their competencies, so there is some potential for gaining some skills in ethnographic research approaches. But, in reality, the networks of mentors and advisors such programs curate typically come from business-oriented or technology-oriented backgrounds, not the realms of design, UX, or any sort of research, for that matter. Thus, they have little access to the knowledge or resources to help them expand these skills. And those resources and mentors they do have access to generally encourage the current model and Lean methodologies, which reinforces using the aforementioned tools, which are representative of the sort of deskilling of labor in research and UX work, much like Lombardi (2009) described in relation to ethnographic work in the private sector. While there is an opportunity to extend the ethnographic community into the realm of startups, it seems that the current model of funding and prioritization of diffusion are still major barriers.

Moving Beyond the Folk Model – The potential opportunities for ethnographic methods in startups is tremendous. Ethnographic research would indeed be transformative for many startups in terms of developing more meaningful, innovative products. But aside from being unlikely to make inroads given the current approaches and dominant model, it also doesn’t push the needle very far beyond the dominant folk model that exists in corporate ethnography. Ethnographic research in corporations functions much the same way as outlined here—finding a new area in which to develop a product, exploring new markets, providing richer data and a more dynamic understanding of the context, and interpreting meaning. Ethnographic research within startups would be necessarily different from corporate ethnography in many ways, and it would certainly have different sorts of impacts terms of helping startups find and develop new, innovative paths to take. Yet, it still is not too far removed from the folk model. By contrast, a much more compelling, innovative path for ethnographic research to explore – and one which would also have a tremendous impact on startups—would seem to be in forging new paths in venture capital.

PATHMAKING: ETHNOGRAPHIC POTENTIAL IN VENTURE CAPITAL

The current state of innovation coming out of Silicon Valley and other tech hubs seems lacking to many. Many of the products and services emerging at the moment seem to be frivolous, indulgent or navel-gazing; they reflect a focus on Silicon Valley. Recent commentary on the topic suggests the goals of these “innovations” are basically to provide for themselves everything that their mothers no longer do” (Arieff 2016). Not only are women and minorities underrepresented (Whitney & Ames 2014), there is an “echo chamber” that creates a closed system in terms of what is considered novel, useful, or innovative. In short, the products aren’t very meaningful for many people. The reasons are myriad. There’s a focus on potential for diffusion above all else, leading to large bets placed on what the next “unicorn” will be. And those bets tend to come in waves that follow the latest trends projected from Silicon Valley thought leaders, not any sort of meaningful, systematic analyses.

While ethnography holds great potential in helping startups find new paths, as we’ve just discussed, the much more fruitful potential seems to be in guiding funders down new paths. After all, they play an outsized role in deciding what succeeds or fails. Venture capital-backed startups in the tech sector have a much higher survival rate than comparable companies (Parhankangas 2012). If ethnographic research has the power to drive truly transformational outcomes in the startup sphere, it is by helping identify new horizons and forging new paths for venture capital, who in turn shape the startups themselves.