Look at the Stars, Look How They Shine For You Image by DALL-E

Wading In

I recently joined a great Slack community called Eventgarde. Organized and facilitated by Timo Dechau, it’s a strong and growing group of people who nerd out about the type of analytics you can do with user event data. As a response to my intro Slack post in which I talked about my interest in “customer analytics,” Timo responded with “I love that you call it customer analytics. My vote would go that we just stop using any other terms and only use customer analytics since this is what it’s all about.” I agree, Timo! In an effort to popularize the term more widely, I’d like to wade into what “customer analytics” is, why it’s crucial, and how it can help us create more value from our everyday business data.

Defining the Customer

We all know what a customer is, right? But since this is half of the name “customer analytics,” let’s just say it: a customer is any person or entity who buys your products or services.

It’s super important to really understand who your customers are, because they’re essentially the backbone of your company’s financial health. The whole value of your business can be boiled down to what these folks bring in, minus what it costs you to provide the product or service—that’s what we call the contribution margin. By getting a clear picture of this, you can come up with smarter strategies to boost your customer value, which leads to better profits and sustainable growth over the long haul.

How is a Customer Different from a User?

A “user” is a person or other entity who could become a customer one day (perhaps in a free trial); is affiliated with a customer (an end-user of a B2B SaaS product); or part of the product (as in the case of Facebook, where users are integral to the product that advertisers buy). When I use the term “customer analytics,” I generally include both users and customers under that umbrella.

Users are not customers, but in my view of customer analytics–especially as it is applied to startups–users are still important for several reasons:

  • There’s a difference between paying for a product and using the product. I have a subscription to the New York Times app and read it every day. I also have a subscription to the Washington Post app and read it maybe once a week. I would say I’m much more of a churn risk–i.e. more likely to cancel my subscription–with the Post than the Times. The Times is a much stickier app for me; not only does it have news but it also has games. Measuring user engagement helps understand customer churn risk.

  • Customers and users can be different groups of people. In the case of, say, a B2B SaaS, the customer is a business using the product–or the people running the department that uses the software. Meanwhile, the users are the people who work in that department who use the software every day. Like the Times vs. Post example above, we need to assess engagement (frequency of use) of the users to assess churn risk of the customers. But with different groups of people, its more complicated to get there.

  • Many startups are pre-revenue, and the only way to assess whether anyone finds value from the product is to measure engagement in the form of non-monetary proxies, such as active days. If users are engaged and the product has become part of their life or job or habits, then monetization is possible, whether through usage fees, advertising, subscriptions, or other methods.*

  • Users are the would-be customers that don’t make it all the way through the acquisition and activation funnels. Understanding their origins, behavior, and points of churn can help improve the throughput of these funnels to grow the business more efficiently.

If find it helpful to use this analogy: customers are to revenue as users are to usage.

Vive La Difference

Not all customers behave in the same way. Some might stop by every day for small things, while others might only show up once a season but make big purchases. This mix is important because it helps you spot your “star” customers—the ones who engage most enthusiastically with your business. By diving into customer analytics, you can figure out these patterns and tailor your strategies to ramp up the value of your business even more. This means you can fine-tune your marketing to acquire more star customers and keep the ones you have happy and loyal.

Customer analytics does something really cool: it fills in the gaps left by traditional financial metrics like those you see in GAAP accounting. By looking from the bottom up at every single transaction with your customers, you get a crystal-clear view of what is driving your business. This isn’t just about tracking dollars and cents; it’s about uncovering the hidden patterns, preferences, and opportunities for growth that you’d probably miss with standard accounting. Three principals from PwC, writing in Harvard Business Review, call it “privileged insights:” unique and relevant information about customers that competitors don’t have access to. With this kind of detailed insight, you’re equipped to make smarter decisions that not only enhance day-to-day operations but also significantly increase the value of your business.

Indeed, research by McKinsey states, “…companies that use customer analytics comprehensively report outstripping their competition in terms of profit almost twice as often as companies that do not.” Collecting the right data and knowing how to use it can drive superior performance.

Customer Analytics Applied to Startups

In my time at TheVentureCity, we strove to understand both customers and users to develop the most complete, bottom-up picture possible of the company’s prospects. While we loved to get data on both customers and users, it wasn’t always possible. We used the best information we could get to make the most informed decision possible. Below, to simplify things, I will refer to customers as a general term to refer to both customers and users.

By analyzing what each and every customer does in the product and then aggregating the data, we were able to paint a picture for founders and our investment teams that showed how “happy” those customers are. Happy customers find clear value in the product and are willing to pay for it. They demonstrate this to us by…

  • Using the product with a relatively high level of frequency

  • Using the product repeatedly over an extended period of time

  • Spending increasing amounts of money via expansion of use or up-sells/cross-sells

If a company can efficiently find and grow a base of happy customers over time through reliable acquisition channels, it is a sign that they have achieved product-market fit (PMF), a critical inflection point for any startup. Measuring PMF is a subset of customer analytics that deserves a whole post of its own (h/t to Jonathan Hsu at Tribe Capital; if you really want to dive in, his follow-up post on unit economics takes it to the next level). This approach primarily helps us determine if a startup is ready to scale up or if the team needs to do more work before scaling. Premature scaling can put the entire company at risk. Measuring PMF is particularly important during the investment due diligence stage, forming the rationale behind the Growth Scanner product we built at TheVentureCity. Tribe Capital recently spun out Termina.ai in an effort to make this technology usable by other asset managers directly.

Diving Deeper

Customer analytics applies at any point in a company’s lifecycle, well beyond the question of PMF. As a company matures, it collects more and more customer data. The more data over a longer period of time, the better the customer analytics. Our data team used the following customer analytics frameworks to dive deep into companies:

  • Customer lifetime value (CLV): What is the worth of a customer over its active “lifetime,” and how does this compare to the cost required to acquire that customer (unit economics)? What is the probability that each customer is still active and what can they be expected to be worth in the future? For more information, see my CLV blog series, Part 1.

  • Recency-frequency-monetary (RFM) analysis: The three underlying attributes that build predictive CLV models can be used to segment a customer base with relatively little math. While RFM cannot be used to calculate the worth of each customer, it can put them into different groups. For more information, see Part 3 of my CLV series.

  • Behavioral segmentation: Are there customers who perform better than others (on a CLV basis or otherwise)? If so, who are they, how do they use the product? Once we have identified who they are and how they behave, the question becomes how can we acquire more of them?

  • Aha! moments: How can the behavior of a customer during the onboarding process predict their retention as a customer in the future?

  • Revenue distribution: Is revenue spread across many individual customers or concentrated in just a few? A typical distribution roughly follows the 80-20 Pareto rule, with 20% of customers generating 80% of the value. Highly concentrated revenue represents a risk to any startup.

We used these frameworks and other methods, including exploratory customer base audits, to help both founding teams and our internal investment teams understand customer behavior. This understanding provides a solid foundation for mapping out how to grow the business.

Putting It Into Practice

Customer analytics plays a crucial role in companies by providing valuable insights into customer behavior, preferences, and trends. This information enables businesses to make informed decisions, improve customer experiences, and drive growth. Typical applications fall into the following categories, which have some overlap:

  • Understanding customers’ needs, wants, and pain points

  • Personalization and customization for tailored experiences

  • Customer segmentation and targeting based on behavior, demographics, and other factors

  • Customer lifetime value and retention to maximize value of customer relationships

  • Identifying opportunities for growth into untapped markets, new segments, and emerging trends

  • Measuring and improving customer experience

  • Data-driven decision-making

  • Continuous improvement in all products, services, and customer experiences

Costs and Risks

Companies have to weigh the potential benefits described above with the costs and risks related to operationalizing a customer analytics initiative.

There are foundations on which to build any data initiative that, if they fall short, will severely hamper your efforts. Make sure…

  • You have good data. That means it’s accurate, complete, reliable, and timely. Buying or building a customer data platform (CDP) is a critical investment that deserves its own post. Collecting clean, high-quality data over time costs money.

  • The data is secure and complies with regulations such as GDPR and HIPAA. Complying, and remaining in compliance, with these standards costs money.

  • The people throughout your organization understand and support what you’re doing. This means both executive sponsorship (top-down) and organizational buy-in (bottom-up). Many people may resist change, lack data literacy, or express skepticism about data-backed decisions. Such resistance can sink data projects. Allocating time, energy, and communication overhead to aligning and training the organization costs money.

  • Your team has the appropriate technical skills and works together effectively to make the project successful. Despite the claims of software providers, even no-code solutions require some level of technical expertise. The more systems you have to collect data from, the more customer ID’s, data dictionaries, and points of failure you introduce to the mix. Hiring specialized technical resources costs money.

Each of the foundations listed above requires both money and good execution to realize value. Therefore, each represents both a cost and a risk. How much cost and how much risk depends on the size and maturity of the organization and the complexity of the data.

There is another risk on the benefits side of the ledger: does the customer analytics project have enough upside to be worth it? In the world of startups, one of the things that was constantly on my mind was “potential energy.” How much value could your data provide, if you know how to extract it? One of my must-read data thought leaders, Benn Stancil, describes potential energy here:

The data of a mid-sized B2B SaaS product simply doesn’t have the potential energy of Google’s search histories, or of an Amazon’s browsing logs. If the latter examples are the new oil, the former is a new peat bog. No matter how good the tools are that clean and analyze it, how skilled the engineers are who are working on it, or how mature the culture is around it, it’ll never burn as hot or as bright.

A customer base audit performed prior to the project using the best data available can help answer such questions as:

  • Who are the customers, and how are the segmented?

  • How quickly are customers acquired and how well do they retain?

  • How many customers do we have in each segment and what are they worth to us?

  • Where are the most obvious “low-hanging fruit” opportunities for creating value?

  • What are the opportunities that will pay off over a longer time horizon?

  • If we invest in customer analytics, what is a reasonable range of estimates for the incremental value we can deliver to the business?

  • Is an investment in customer analytics likely to pay off?

Answering these questions can help you assess how much potential energy your data has prior to spending much money to extract value from it.

Examples

At TheVentureCity, we could amortize the cost of our data team and infrastructure across an entire portfolio of companies for whom we performed customer analytics. Therefore, the threshold for value extracted from a customer base audits or other deep dive we did was lower. While some of our startups grew to be quite large with high potential energy in the data, many smaller ones had lower potential. We did our best to deliver value to as many of our portfolio companies as possible.

Retail Fintech

We partnered with a fintech startup and discovered a unique customer segment that conducted transactions at an unusually high frequency. These weren’t just regular consumers; they were entrepreneurs leveraging our client’s product as a business opportunity. By recognizing this, we helped the startup launch a tailored affiliate program that catered specifically to these entrepreneurs. This program not only facilitated their operations but also empowered them to expand their customer networks more effectively. The result was a significant increase in user engagement and a higher rate of customer acquisition, turning a routine observation into a strategic business advantage that drove growth for both the company and its entrepreneurial customers.

Retail Cross-Selling

An analysis for a fast-growing retailer revealed that customers purchasing a diverse range of products were twice as valuable as those sticking to a single product type. This insight was a game-changer. It led us to develop targeted marketing strategies aimed at promoting cross-selling, which not only boosted sales but also enhanced customer participation in the loyalty program, thereby increasing customer lifetime value and loyalty.

Video Streaming Cohorts

During the pandemic, a cohort analysis for a video streaming company highlighted a surge in engagement among new subscribers, with retention rates surpassing historical averages. This crucial insight informed our investment strategy, prompting us to model future scenarios that accounted for fluctuating engagement levels. Our proactive adjustments ensured the company could ride out the volatile era and maintain robust growth.

Travel App Geolocations

For a travel app client, we conducted an advanced analysis of user geolocations and engagement patterns. This not only mapped out where users were most active but also identified potential hotspots for content development. By focusing our marketing efforts and content creation on these high-engagement areas, we significantly increased organic user activity and enhanced the overall user experience, leading to higher satisfaction and retention rates.

Special Thanks

To Jonathan Hsu at Tribe Capital for his thought leadership over the years and assistance on this article.

* “It’s interesting to actually consider that customers being distinct from users is a bit of a ZIRP [zero interest rates policy] phenomenon. They can only be separate if a customer is willing to subsidize the usage by some other party at some level. Whereas when interest rates are high, the customers want the company to build directly for them. For example, in a ZIRP world, Facebook can build primarily for their users and build only crappy products for their advertising customers. As interest rates go up, Facebook has to focus on ad products and reduces investment in the end-user experience, etc.,” says Hsu.