In business, a moat is an advantage that goes beyond survival - it establishes a company in such a way that competitors can’t easily copy the company’s success.
Many companies have moats, but not every moat is the same. Facebook has a network moat - everyone’s friends and family are already there, and as a result, it’s very hard for a newcomer to replicate that user base. Amazon Web Services has a switching cost moat - companies looking to migrate out of AWS often have to pay tens of millions of dollars. Microsoft has a corned resource moat - arguably, they have the best enterprise distribution engine.
The list goes on, but the point is clear. Companies with staying power have moats. With AI and machine learning driving ever-shortening feedback loops, we’re starting to see a new moat emerge: collection and analysis of customer feedback.
It feels self-evident that customer feedback is important. Y Combinator’s mantra is “build something people want” and Amazon prides itself as a customer-centric company.
In building a company or launching a product, you face two risks - market risk and execution risk. Market risk is when people don’t need or want the product. Execution risk is when competitors can execute better and take market share.
Airbnb skewed toward market risk. No one knew if there would be a market for renting a spare room in someone’s house. On the other end, Zoom was more execution risk. When they started, video conferencing was already a well-understood space with multiple alternatives.
Customer feedback tackles both risks.
In a new market, customer feedback is often the only way that you can gain signal on their product. Airbnb had their breakthrough moment when they realized that poor photography was a barrier to bookings. To solve this, Joe Gebbia, cofounder of Airbnb, went to NYC and started taking photos for the hosts.
We would go in, we take photos. And then we’d show them on the back of the camera and say ‘hey, what do you think?’ and they’re like ‘oh my God, my apartment looks so good! do you want to stay for some tea of coffee?’ And, so, I would sit down on the couch, and our earliest customers, much like yours, where zealots. These were the people willing to take a risk, try a new weird crazy product, be kind of outcast amongst their friends, which meant that they had a lot of knowledge. They had a lot of insights into this kind of activity because they were already doing it in other web sites.
- Joe Gebbia, cofounder of Airbnb
Fast-forward a decade and today, Airbnb is an integral part of millions of travel plans.
In execution risk, discovering and solving customer pain points is your best competitive advantage. Zoom’s founding vision came from customer pain points in using WebEx - poor video/audio quality and a choppy mobile experience. As Zoom took off, customer feedback showcased that customers wanted more than just a good conferencing solution, they also cared about privacy and security.
Anyway, I think to build a better solution, you’ve got to spend more time with your customers. Ultimately, it was innovation. Innovation is really about you want to be the first company to understand customer pain points. And, also, take actions quickly, and to be the first [vendor] to build a solution. If you keep doing that, sooner or later, you are going to win. So, that’s our approach. That’s the reason other competitors lost.
- Eric Yuan, CEO of Zoom
When you leverage user feedback to gain insights and make improvements, your products fit users better. Your new features are more on target. And your priorities are closer to what your customers need.
These small advantages accelerate progress and, over time, they accumulate and create an obstacle against other competitors in the market. They become a competitive moat.
Even though it feels obvious that companies should incorporate customer feedback into how they design their product, how and how well they do it is subject to tremendous variance.
The distance between vision and implementation is a function of getting the correct context out of conversations with customers. That context isn’t always obvious and often requires significant investments of time, energy, and money to discover.
The largest difficulty with customer feedback isn’t actually obtaining it - customers will suggest improvements, ask for help, send in complaints, and praise new features by themselves. But customer feedback is noisy, and it’s difficult to find a useful signal in that noise.
Part of this problem is that feedback comes in dozens of different forms. A couple of years ago, companies sent out annual satisfaction surveys, keeping feedback in a neat and tidy box. Today, any SaaS company has FullStory sessions, support tickets, emails, dedicated Slack channels, Tweets, and sales call data. These are all distinct types of data that don’t play well together; they carry different tones, come from different kinds of customers, and arrive in different formats.
On the positive side, you now have an almost endless stream of customer data to mine for insight. On the negative side, it takes exponentially more energy to make sense of all the information.
When a company is small, a founder can look at all the incoming feedback themselves and get a good mental image of customer priorities. But that doesn’t scale - eventually, responsibilities are delegated, and no member has direct access to more than a small fraction of customer feedback. When a company’s largest prospect delivers feedback that they won’t commit before they get a niche feature they need, it’s hard for the company as a whole to judge whether it’s worth the engineering time to build it.
Customer data also means different things to different parts of the company. The sales team might place additional emphasis on customer requests for new features, the support team on ensuring existing features don’t break, and the product team on balancing tradeoffs across different priorities from leadership and customers.
All this adds up to a time and energy investment that most companies aren’t able to make. It’s not that companies don’t understand the value of feedback. It’s that most companies struggle to sort through the noise and extract the context they need to fully leverage customer feedback. Yet, by walking away from customer feedback, they leave valuable insight and product direction on the table.
In an ideal world, the feedback you receive should be fed into your company’s roadmap, messaging, growth plans, and culture. Part of what made Amazon so successful was their laser focus on customer experience; they sought out a wide variety of types of feedback and then relentlessly focused on improving the customer experience to drive the rest of their company flywheel.
The landscape and scope of customer experience has evolved in the three decades since Amazon was founded. Today, we’re at a point where analyzing the sheer bulk of customer feedback is no longer a job humans can handle unassisted.
The solution is AI. More specifically, large language models (LLMs). Before, companies were getting overwhelmed with the volume and diversity of incoming feedback. Now, LLMs can synthesize distinct pieces of feedback and find trends. They can help you identify new and growing problems that were previously lost in the noise. Most importantly, they can package all of this in a way that humans can read and understand at a glance, so anyone in your company can access the context they need.
That’s why we started Enterpret. We set out with the mission of building analytics for customer feedback - bringing Amazon’s flywheel to startups and Fortune 500 companies alike.
To do this, we build a custom ML model for each of our customers that ingests data directly from various data silos such as Zendesk tickets, Twitter tweets, and even Slack/Discord messages. The model automatically categorizes feedback into various themes and reasons. Product and engineering teams can then query on the model to understand sentiment, find insights, and ultimately drive product outcomes.
There’s a growing distinction between companies that can properly leverage customer feedback and ones that can’t. More feedback, properly analyzed by large language models, has led to actionable insights, better products, and ultimately more revenue.
We’ve worked with Notion to empower “a holistic view from our social media coverage, to our support tickets, to every single interaction that we're plugging into it.” That then led to additional insights for the product team, including answers to product questions such as “Hey, what are the top requests related to search? When do users report issues with real time collaboration?”
For more information on our partnership with Notion, see our case study on How Notion is supercharging its product feedback loop using Enterpret.
In a world where users are becoming more well-informed and enterprises are grappling with the disruption of new technology, it becomes increasingly important for you to be aligned with the customer. Doing that means a distinct advantage in distilling market trends, locking in customers, and achieving higher margins. In other words, collection and analysis of customer feedback has become a moat.