FECS Part 1: Intro to FECS; Unpacking Growth

Intro to FECS: FECStra! FECStra! Read all about it!

For almost a year now, I’ve had the pleasure to work as a VC at Cue Ball Capital in Boston. It is a very unique fund in that it is evergreen — it works out of a long-term fixed capital pool. Rather than the standard 7–10 year lifespan most venture funds have, the fund has no time horizon so there is much less pressure on founders to exit in order to realize portfolio gains (or losses!). The fund looks at varying sizes of rounds (Seed to Series B / C) and across industries — in Cue Ball’s case, the verticals range from Enterprise SaaS to Media to Branded Consumer, with a special interest in Inclusionary Ventures.

One unique viewpoint we have at Cue Ball is one where we look at companies and helping portfolio teams through a method we call Front-End Customer Strategy, or FECS for short. It was developed at Thomson-Reuters by two of Cue Ball’s founders, Tony Tjan and Dick Harrington, and was used at Thomson to transform the company into the media information giant it is today. The original FECS article published in the Harvard Business Review can be found here, and is a great primer for what is to come.

The driving force for the FECS methodology is almost obvious: Know Your Customer. However, it’s always the simple things that are unexpectedly hard to think through; for example, what does it mean to know the customer?

Surprisingly (or not), this mantra breaks down into 6 fundamental categories of analyses:

1. Unpacking Growth
2. Customer Count + Revenue Contribution
3. Customer Retention
4. Inventory Turnover + Hero Product
5. Utility and Dependence
6. The 3 Minute Rule

In this first mini-series of articles, I will go through the most basic analyses in each of these categories and contextualize them with real-world applications of the data.

Before FECS:

After FECS:

First up: Unpacking Growth.

FECS Part 1: Unpacking Growth

The average American does not like lukewarm tea. Averages say otherwise.

We like numbers. Numbers are pretty great at telling stories or reporting facts. Sometimes we want to say “hey, I have 1 apple and 1 pen”:

No caption needed.

or “it takes 16.5 seconds for one swing on this pendulum”:

Focault’s Pendulum at the Panthéon, Paris

or “nice, the S&P 500 went up 12.5% from last year’s Halloween to this year’s Halloween”:

S&P 500 10/31/18–10/31/19.

Do we really know what’s going on behind these numbers?

We have three interesting degrees of craziness behind numbers:

• Our good friend Mr. Pikotaro is pretty up front with these numbers. One apple, one pen. Nothing else goes into it.

1 apple + 1 pen = 1 apple pen. Enough said.

• A pendulum gets a little more interesting as there are two fundamental numbers that determine a pendulum’s period, where l = length of the pendulum and g = gravity. As mere humans, because we can only really adjust the length of the pendulum, we can reasonably argue that there is a pretty clear relationship between length and period. We can calculate that the length of the pendulum is around 68 meters long.

Function for period of a pendulum.

• The S&P 500 has quite a few cogs in the machine… 505 to be exact. Each of the 505 common stocks has its own weighting in the S&P 500’s portfolio, differently affecting the overall value as each grows at its own rate. In layman’s terms, imagine a beanbag filled with 505 little beans, and for some incomprehensible reason each one grows / shrinks individually. Each bean might grow really big relative to itself, but it doesn’t make as massive of a difference in the bigger picture of the beanbag itself.
  A quick view into the top 5 stocks on in the S&P 500 (as of 11/8/2019):

Top 5 stocks in the S&P 500 as of 11/8/2019. Source: https://www.slickcharts.com/sp500

Let’s run with the S&P 500 example. What does a change in the S&P 500 tell us about any individual stock?

Absolutely nothing at all.

Well, maybe a little bit — it is a good directional indicator. But directional indicators often don’t tell us about what happens below the surface. Bear with me one moment and let me give another example, body weight. We like to talk about lowering our body weight, but as the numbers go down, does it mean anything? The underlying factor can be that yes, you did slim down, but it can also be that you just lost a ton of muscle mass while gaining fat. An extreme case but one I think is worthwhile in illustrating the deception in averages.

Unpacking averages is fundamental in understanding the key drivers of growth.

For a more elaborate example, let’s use Microsoft. Here is a chart showing revenues for Microsoft over 4 quarters, with growth period-to-period:

Microsoft revenues for 4-quarter periods with growth rates. Don’t mind the wonky start and end dates.

Let’s break these growth rates out for each of Microsoft’s segments: Productivity and Business Processes, Intelligent Cloud, and More Personal Computing:

Microsoft revenues for each of its segments.

Here, there is a pretty interesting story. The Productivity and Business Processes segment of the business follows the general pace of overall revenue growth, but the Intelligent Cloud segment has been growing rather intensely. It makes sense taking into account Microsoft’s big pushes into cloud computing and the fact that cloud is ubiquitous today, with everybody and their grandmother asking if software solutions are cloud-enabled. What’s also very interesting is that the More Personal Computing segment shows that while yes, Microsoft is found in many households today in personal computers and Xboxes among many other things, Microsoft is really at its core an enterprise business. This story isn’t one that is very obvious from the overall growth numbers, which tell the story of a company that 1. makes a ton of money and 2. is growing.

tl;dr: overall growth glosses over an interesting story. Unpacking averages is fundamental in understanding the key drivers of growth, here showing that Microsoft’s growth has come mostly from its Intelligent Cloud segment.

Just for fun, let’s apply this thinking of unpacking numbers to the pure revenue numbers this time. What do we see here?

Revenues broken out monthly for each segment.

Let’s look at Intelligent Cloud. In every year there is a marked revenue bump in the quarter ending June 30 — this is when Microsoft’s fiscal year ends, and most likely the time when salespeople try to close as many clients as possible. Productivity and Business Processes also has a very small (almost insignificant) bump that exhibits similar behavior. This end-of-fiscal-year-push is an indicator that this segment is also likely enterprise-focused rather than consumer-focused.

How do I say that? Let’s look at the consumer-focused segment, More Personal Computing. The quarter ending December 31 covers the holiday season, which is when people like to gift electronics to family and friends. Because this segment caters more to consumers, there is more sensitivity to this gifting behavior year over year. Comparing the other two segments to this one shows that there is no gifting behavior in the enterprise segments — after all, who gifts their IT admins with a subscription to Office 365 every year?

The moral of the story here is that numbers reveal more and more detail the further you peel back the layers to understand the inner workings, much like ogres:

Hint: They all have layers.

And once more for the people in the back:

Unpacking averages is fundamental in understanding the key drivers of growth.

Stay tuned for Part 2: Customer Count + Revenue Contribution.