Revenue Predictability: What It Is and What It Is Not

Show me the money. Show! Me! The! Money!

Good old-fashioned romantic comedies have at times nuggets of wisdom for all of their cheesiness. Hopefully, you have learned to love and laugh at stale processes and business clichés. Keeping with our Rom-Com theme, look for subtle Rom-Com references sprinkled throughout this article and see if you can correctly catch all of them. The first is a freebie - Jerry Maguire.

Demand Gen Crashers

Imagine you are a Demand Gen team, and you have been pulled into another forecasting meeting where your CMO wants to know the plan for the next Fiscal Quarter, Fiscal Year, etc. You start discussing the top-level revenue commitments, and naturally, a missed goal means the bottom-up model your team built-in Excel needs updating and resharing. As you drill down, your team realizes there have been a few misses against month-over-month leading KPIs. What some folks do at this point is try to course-correct through inflated growth rates with the hope of catching up in the following review cycle. This type of course correction is a slippery slope. The magnitude of Leads, MQLs, SALs, SQLs, and OPPs, needed to make the revenue target becomes fiscally impractical as the conversion rates decrease further. For example, going from 100 opportunities created per month to 500 could mean a 7x to 10x increase in budget overnight!

Even with the most well-intentioned forecast models with some configurable conversation rates and a few crazy complex VLOOKUP or INDEX & MATCH formulas, you likely know in your heart (and hopefully from the data) that anything before opportunity creation is fuzzy and hard to predict. Talking to many Growth, Product, and Demand Gen teams, a common thread that has surfaced is never having the correct revenue data tied to the measurable marketing efforts accessible at the right time to act on it. Talk about a stale model that needs refreshing!

What if you had a better way to build a forecast model comprised of two major components?

Predictions on the revenue value for each lead by segment, the moment they come into your systems, not months later when a deal closes.
A way to improve your forecasts with expected vs. actual revenue predictions through an attainment calculation that you could trust.

Would you run experiments differently, would you ungate your content, and ultimately would you move away from a Lead and MQL generating focus in favor of revenue centricity?

SPOILER ALERT for many of our customers, the answer was an outstanding yes!

When we start talking about changing the Excel forecasting model, confusion sets in. These revenue forecasting models are typically built using concepts from various forecasting tools often expressed as a slurry of acronyms such as Total Addressable Market (TAM), Market Share, Share of Wallet (SOW), and many others that can often be confusing. Folks have misconceptions about how to use each of these forecasting tools and where to best leverage one tool over the other in pursuit of revenue predictability.

Since these terms are used interchangeably and have loose definitions across various organizations knowing what each forecasting tool is and is not is critical. Throughout this article, we’ll explore what each of the acronyms means, what tracking them does for your business, and introduce the concept of Predicted Lead Value (PLV) and the impact it has on your forecasting toolkit.

You, Me, & Total Addressable Market (TAM)

First, let’s start with the groundwork. Total Addressable Market or TAM is a standard calculation for estimating the sum of all the theoretical customers that could buy your product*. When companies think about their TAM, they are trying to forecast the revenue growth of a possible new industry, geography, or product offering. If the growth strategy is one where there is not a leading competitor, then companies would be gunning for a first-mover winner takes all majority advantage.

*Note There are many ways to filter down TAM to specific serviceable chunks such as Served Available Market (SAM) and Serviceable Obtainable Market (SOM), but for this article, we will not discuss them. Do note that they exist and can come up in forecasting models.

What is TAM good for?

TAM helps companies understand the potential for new product launches, segment account territories, convince investors to raise capital, inform sales routing, and steer revenue operations. Marketing and sales teams use TAM forecasts as building blocks for understanding the maximum potential of various campaigns and targeted playbooks. When used in your forecasting toolkit the core use case for TAM comes down to strategic hiring plans and understanding where companies should deploy resources.

What is TAM not good for?

Even though TAM provides a good framework for forecasting what revenue you could capture, its output does not tell you what revenue you will capture, hence its status as a theoretical forecasting method. Using TAM as a forecasting toolkit has its strength in macro modeling but is ineffective at forecasting incoming demand, and it is poorly suited to steer course corrections. Ultimately, the proximity to revenue is low, if you have read our blog on intent data then you can think of TAM as the farthest layer away from revenue. The closer you get to revenue the more likely your company is able to convert the account interest into something tangible.

Market Share Actually

Now that we have an understanding of the strengths and limits of TAM we need to unpack how Market Share ties into the forecasting toolkit and its relationship to TAM. Market Share represents the percentage of sales a particular company owns relative to the entire market. In other words, if you think of TAM as the whole pie, then Market Share is a slice of that pie. By growing your company’s Market Share, economies of scale start to amplify brands; brands typically see a natural reduction in customer acquisition costs, hiring costs, and supplier/integration costs.

What is Market Share good for?

Market Share gives companies a barometer of whether they need to focus on customer acquisition or customer retention. Generally, if Market Share is low (~3% to 5% for mature markets) or decreasing, the company should focus on customer acquisition. If high (~19% to 30% for mature markets) or increasing, the focus should be on customer retention and product innovation.

If a company is in a semi-established to a mature industry, a growing Market Share typically translates to more customers and lower churn rates.
If a company is in a new or growing industry, a growing Market Share typically translates to greater brand affinity and lower customer acquisition costs (CAC).

Market Share plays a critical role in yearly resource splits or reallocations based on strategies that swing between customer acquisition and customer retention within the Forecasting Toolkit.

What is Market Share not good for?

Market Share gives you an understanding of where to reallocate resources, but it does not tell you how to reallocate resources. If, for example, your company needs to focus on customer acquisition, Market Share calculations will not tell you which campaigns to run, nor will they tell you which types of people to target based on their worth to your business. Market Share is good at informing strategy but lacks its ability to direct an execution plan.

Forgetting Share of Wallet (SOW)

SOW is uniquely different from TAM and Market Share type calculations. SOW is a revenue forecasting metric that indicates how much (on average) a customer spends on a company’s product or service as compared to how much they spend on competing products or services.

This forecasting method is associated with the B2C type of company. This method was popularized and mainly used by B2C companies for the better part of a decade. However, in recent years, this type of forecasting has made its way to the B2B side of the house, and rightfully so. When companies leverage this forecasting tool, they shift focus away from customer acquisition (typically more lead generation focused) and instead focus on customer expansion/cross-sell (revenue centricity).

What is SOW good for?

SOW is good for understanding how to price your product correctly and provides a framework on how to generate more revenue from your existing customers. In a land and expand motion, SOW gives you a comparison between the cost of acquisition against the size of wallet rather than the current spend enabling proper resource and time allocation to ease a customer into spending large multiples with your business. This forecasting tool works best when your business offerings have add-on features that provide extra value or complementary products extending value across teams or use cases. Driving a greater SOW ultimately leads to added revenue, decreased churn, and increased brand affinity.

What is SOW not good for?

SOW helps frame the upsell approach when all of your customers get the same value from your product; it does not work well when your customers are diverse and get varied value from your product. For example, when one customer thinks of you as a strategic partner, they are more likely to perceive your products as having a massive Return on Investment (ROI). In contrast, other customers could see your business as akin to a task taker and likely perceive your products as having moderate to low ROI. Distinguishing between these customers can make SOW calculations cumbersome and ineffective for some businesses.

Help me help you with Predicted Lead Value (PLV)

At this point, we have gone over TAM, Market Share, and SOW. PLV might be a new term that is less familiar to you but rest assured of its importance. To explain PLV and its impact, we need to step back and understand levels of aggregation and proximity to revenue. Whereas PLV lives at the individual level, TAM, Market Share, and SOW typically live at the company or account level. In short, revenue is brought down from the account level and onto the individual level through a mix of average conversion rates and average deal sizes for a particular segment.

At the end of this article, we also introduce the concept of Predicted Account Value (PAV) which functionally does the same things as PLV but at the account level. The PAV output* is a mix of average conversion rates and average deal sizes for a particular segment and is typically used for cross-sell estimates by segment.

*Note: PAV is not currently supported as a standard configuration within the MadKudu platform, but in our deep dive article on this topic we will put forth ways in which you can calculate PAV within your systems.

What is PLV good for?

Knowing the PLV of every individual in your system means you have a real-time vantage point into the revenue potential of all of your incoming demand. Imagine knowing how much a trial or product sign-up is worth, regardless of the acquisition channel (ex: CPL for a product sign-up from Facebook versus LinkedIn might favor Facebook, but PLV could tell a different story around the CAC and predicted revenue). Ultimately PLV enables customers to segment different company resources by expected revenue throughput and for most of our customers, having this information allows them to rapidly course-correct well before sales start an arduous deal process.

What is PLV not good for?

PLV is not the same thing as a customer's size of wallet. It will not be able to provide your business with an estimation of how much a customer can spend with your business. Although PLV is a powerful tool that provides a real-time assessment of an individual's value, it does not provide a dynamic assessment based on your company's changing GTM strategy. Since your business processes are continuously improving and new initiatives are not equally represented in historical conversions, some forecasting built using PLV outputs will need monitoring and iteration. For many of our customers, balancing expected versus actualized attainment is delicate coordination between many teams and various stakeholders, which can be challenging if feedback culture is not developed.

You had me at Forecasting

Now that we have gone through your forecasting arsenal, it is up to you to update your tracking (where needed) and be more data-driven, knowing that you have a firm grasp of various acronyms such as TAM, Market Share, SOW, and PLV. To credibly predict revenue, your organization will need to steer clear of methods not close to revenue, such as TAM, and leverage methods close to revenue, such as PLV. Revenue predictability by nature has to be less theoretical and has to be rooted in your GTM context - utilizing your revenue data. Revenue predictability cannot be based on external evaluations of your market.

We are not discrediting different forecasting tools and their ability to help answer additional questions; no one tool solves every problem. However, for revenue-centric Marketers, “maybes” do not inspire confidence.

There are no silver bullets for forecasting and predicting revenue with 100% certainty. Forecasting models can get better and closer but never capture all of it. As your business grows and innovates, your forecast model has to be flexible enough to adapt to the changes but be grounded enough to give you the fiscal and personal confidence to bet your career on it.

See the table below (Exhibit 1) and framing visuals (Exhibit 2) for a handy quick guide on all of these terms.

For those following along, here are the Rom-Com movies in order of reference: Jerry Maguire, Wedding Crashers, You, Me, and Dupree, Love Actually, Forgetting Sara Marshall, Jerry Maguire (x2).

Exhibit 1: Quick Guide

Exhibit 2: Forecasting Radar Chart

‍

The radar chart above highlights how your forecasting toolkit moves from a high-level account forecast not close to revenue (TAM) to an aggregated individual forecast very close to revenue (PLV & PAV). The greater the difference between the red arrows, the more theoretical the forecast model outputs. The larger the blue arrows, the more practical the forecast model outputs.

‍