The area of business valuations has always been a challenging field, especially given the increasing digitization and complexity of the global economy. The emergence of the COVID-19 pandemic has created more uncertainty, and further stratified the economy away from asset-heavy, centralized entities towards asset-lite, decentralized entities of the future.
Tech heavyweights dominate the S&P 500, replacing the likes of the banks and industrials at a lightning pace over the last decade. The business models and valuation metrics of these tech giants are fairly well-established, ranging from hardware & software services (Microsoft/Apple) to advertising (Facebook/Alphabet) to a bit of everything at Amazon (eCommerce, Cloud, etc). Companies like Shopify (eCommerce), Salesforce (CRM software) and a host of others have scalable, asset-lite models and sit a level below the giants; therefore, not only do these types of companies dominate today, but they will likely do so for the foreseeable future.
But once you get into the mid-market (below the heavyweights, but above the startups), valuation starts to become a very difficult exercise. Many of the early-stage Facebooks and Shopify’s will be:
- Losing large amounts of money and be nowhere close to becoming profitable (#BurningCapital)
- Require several cash injections over a period of several years in order to become profitable (#InvestorDilution)
- Making risky strategic maneuvers – such as acquisitions, entering new markets, new product development – that can amplify cash burn during that time period (#Uncertainty)
For this reason, most valuation models (IRR, NPV, etc) that rely on predictable cashflows over the mid to long-term get thrown out the window for these types of companies. They are typically bankrolled by VCs (Venture Capitalists) whose risk tolerance and investment profile skew towards a winner-take-all, all-or-nothing approach. Even when they start trading publicly, they are prone to failure as we have seen with companies like Blue Apron and a host of others; let alone the WeWorks who were on the verge of going public at egregious valuations. Many times VCs and private equity just want to dump on retail in order to exit their positions; that’s why we need a methodology.
Customer-Based Corporate Valuation (CBCV)
While researching this subject last year, I stumbled upon Theta Equity Partners website and their published paper on CBCV.
The premise is that by digging deeper into the customer data behind publicly-traded (in this case), non-contractual tech companies, you can come up with valuation metrics that give you a more-accurate valuation picture compared to just looking at the P&L.
As an example. You may typically analyze the financials and look at their Balance Sheet and P&L (EBITDA), looking at margins, profitability, etc. and then calculating a set of ratios (Price/Sales, etc) to determine valuation. This is typically an aggregation and synthesis of accounting data. In contrast, the CBCV methodology looks at data such as ARPU (Average Revenue Per User), CAC (Customer Acquisition Cost) and Churn (% of loyal customers) in order to create a picture of expected revenue growth/profitability relative to the ability to acquire new customers. This is an aggregation and synthesis of customer data.
When you get into the details of it, the contrast is striking. Whereas a traditional approach can be extrapolated on a linear dimension (as you would expect for companies in the traditional, asset-heavy economy), this approach allows for extrapolation on a non-linear dimension (in line with a more digital, asset-light economy). The CBCV approach is how, for example, you may be able to spot the next Shopify, a company that is not yet profitable (yet is trading at a valuation of more than $100Billion).
The following will be a condensed summary of the key points in the paper (34 pages and formula heavy), followed by a short discussion afterwards.
Summary of CBCV Paper
For many firms, customer equity represents the majority of shareholder value of the firm, enabling an explicit link between customer behaviors (i.e., acquisition, retention, and spend) and the overall financial valuation of the firm.
In contrast to contractual businesses (ie. SAAS, etc) there are three main challenges of performing CBCV in non-contractual business settings:
- Customer Churn is observable for contractual firms but unobservable for non-contractual firms
- Repeat spending and purchasing behaviors are unpredictable and therefore difficult to model
- Accessing and aggregating granular, customer-level data is difficult, especially when looking at public companies
It is important, however, to still model for a) churn b) repeat spending and c) customer-level data that may not be fully broken down, yet still can be extrapolated from. The premise of this type of modeling is that if a non-contractual business can expand its market presence and drive up ARPU, then it will be valuable in the future. The ratio of ARPU to CAC (ie margin/unit economics) can only be guestimated if both churn and repeat spending behaviors are modeled in. In a non-contractual business, there is no standardized way to model this, as customers spending patterns will be based on an array of factors. Yet the best businesses in this space are able to create demonstrable loyalty in their existing customer bases (low churn), and acquire new customers at a reasonable CAC (high margin).
Our proposed model reflects important empirical realities associated with noncontractual customer behavior, including latent attrition, repeat purchasing which may vary across customers and over time, and time-varying spend-per-purchase patterns. Their model uses DCF (Discounted Cashflows) to arrive at their valuation.
QRev is the component of the model where customer data is factored in.
Our valuation goal, then, is to specify processes for the acquisition of new customers, how many repeat orders these customers place after they have been acquired, and how much they will spend on each of those orders.
Typically, contractual firms will disclose the # of customers added each quarter, and the # of customers at the end of each quarter. As non-contractual firms can’t do this, publicly-traded companies may break down their numbers around ‘active customers’ and a number of varying other customer metrics that relate to the business. For the purpose of this paper, they compared Overstock versus Wayfair.
QADD (Customers Acquired in the Q), QAU (Quarterly Active Customers), and AAU (Annual Active Customers) are coupled with revenue data (ie. QREV quarterly revenue and AREV annual revenue) to start painting a picture of valuation.
They then developed a way to model ‘the acquisition process’ including:
- Identifying a ‘pool of prospects’ who may be acquired in the future
- The conversion of those prospects into ‘intenders’ who will one day become customers
- Duration of conversion from ‘prospect’ to ‘intender’ to ‘customer’
They use population and workforce growth rates to estimate market size (depending on the country). The main point is to be able to create a model around the ‘acquisition process,’ estimating market size using statistical models is relevant but won’t be expanded on in this summary.
Once you have the specific data points on ARPU, etc, the model starts to take shape around CLV (Customer Lifetime Value). Obviously these modelers incorporate advanced statistical models and valuation software, but the point here is just to understand the basic concepts and extrapolate.
Accounting for the time value of money properly, summing these quantities provides us with the expected CLV of newly acquired customers. Successful businesses are able to acquire many high CLV customers.
Wayfair customers generate more profits than Overstock customers after acquisition – the net present value of future profits after acquisition are $59 and $47 per customer at Wayfair and Overstock, respectively. However, Wayfair spends far more than Overstock to acquire new customers. Wayfair’s CAC is $69, nearly double Overstock’s $38.
Wayfair was expected to grow exponentially, but there were doubts about its ability to become profitable.
In 2020 – one of the most volatile years in market history – Wayfair has traded in a range between $21.70 and $181.39, demonstrating the fluidity required in these types of processes. One would have to deep dive into the Financials and Quarterly reports to understand why – sufficient to say that the CBCV Valuation model needs to be contextualized with the broader macro environment and a series of other factors.
Application in the Present Day
Theta Equity continues to analyze new companies and publish research around the CBCV valuation theory.
They even have a nifty simulator that you can play around with for Farfetch’s valuation.
Obviously, this is very advanced modeling to determine the valuation for investment purposes.
But for non-public companies, there are many relevant points here in terms of growth, valuation, and future financing:
- High Growth Rates (whether monthly, quarterly or annually) are not steadfast indicators of future performance alone. A view on unit economics (at least ARPU vs. CAC) is also required to guestimate future valuation
- Estimating Churn (ie. Loyalty) is difficult for non-contractual firms, but worthwhile. If you can paint a picture of repeat buyers, even conceptually, it helps to drive-up valuation
- The COVID19 crisis has obviously exacerbated the need for financing to cover cashflow gaps for many firms. Putting a value around customer ‘goodwill’ for private companies could help receive more favorable financing (whether through equity, debt, or buyout). The better the unit economics/CLV ratios are, the more profitable the business will be when things bounce back
Overall, the CBCV valuation model is a relatively novel and innovative way to value publicly-traded technology companies; however, there are ways that this methodology can be applied to use cases beyond investing in public securities. It may be useful for forecasting, strategy, and primarily valuation; but ultimately it is a new mindset for analyzing businesses and their growth profile. And it may help identify the next Shopify, Farfetch and others before they even become public.