Instrumenting Your Volume Market Engine

Everyone understands the concepts of a market funnel – leads are loaded into the top from various sources; they go through some level of qualification and scoring before being passed to sales; and the sales team then develops them further into pipeline opportunities and deals.  In traditional enterprise markets, demand creation and low-touch selling are subordinate to the rolodex-driven, four-legged sales process.  In volume markets, frictionless marketing and sales are the process.  Where enterprise sales is an artful pursuit, volume marketing and selling is all about math.

Don't get me wrong.  There are aspects of volume engines that require tremendous artistry.  Crafting a concise, hard hitting call-to-action or designing an eye-popping landing page demands the best creative resources on the planet.  But the core of a volume market engine is its ability to generate, develop and process huge numbers of leads with extreme efficiency.

Volume Market Funnel

I prefer to visualize volume market engines as a series of closely-connected funnels because that image immediately suggests some of the engines' key instrumentation metrics.  At the highest level, we need to understand the relationships between raw leads, qualified leads (MQLs), sales pipeline and bookings.  But even that's not enough.  What we really want to know is how many raw leads need to be loaded into the top (or left side, in the figure above) of the engine to produce $X in bookings at the end.

The best way to design and instrument a volume engine is to start at the end and work backwards.  If we leave renewals aside and focus purely on new organic bookings, important questions come screaming forward:

  1. What is the conversion ratio of sales pipeline to bookings?  Part of the answer depends on how diligently the pipe is scrubbed at the beginning of each quarter.  A carefully scrubbed pipe may consistently yield a 2:1 conversion ratio, where the conversion ratio of a loosely-managed pipe may be 5:1.
  2. What is the average selling price per new deal?  What percentage of new deals can be up-sold within the reporting period (e.g., calendar quarter) and at what ratio to the ASP?
  3. Based on 1 and 2 above, how many MQLs are needed to build a defined sales pipe, and how much time does the sales team need to develop those leads?  The first part of the question is very tricky because not all leads perform equally.  For example, leads from a "Please Have Sales  Contact Me" webform will usually convert at a much higher rate than leads that have been developed through a slow-drip nurturing process.  The engine needs to be intelligent enough to accommodate these differences and allow us to set our knobs and dials appropriately.
  4. Based on 3 above, how many raw leads are required to produce the targeted number of MQLs, and how much time does the marketing team need to nurture those leads?  Again, it's very important to instrument based on lead sources because leads will flow through the engine at different velocities based, in part, on their origins.
  5. What sized marketing budget is needed (or available) to produce the required lead flow, and how can that budget be optimally deployed?  This is the part of the model that needs to incorporate data like average cost per lead (CPL) for each lead source and historical lead source performance.

Let's look at a configuration dashboard for a hypothetical (and somewhat simplistic) volume engine.  In this example, we want to model the number of raw leads we need to generate in the current quarter to deliver $1M in bookings two quarters in the future.  Volume Engine ExampleWe start by inputting the model drivers and our future bookings target.  We then iteratively cycle across the Leads cells with the goal of balancing lead targets in a way that meets our bookings goal.  In the example, the model is still short of the goal by $3,700.

If we've built-out the rest of the engine properly, the lead targets defined in this dashboard will flow down to lower layers, where they can be further allocated to specific programs, campaigns and other sources.  As we run those programs and track their results, it's quite easy to see how all layers of the engine are performing.

An interesting insight in our example is that we need to deliver 5,100 raw leads (website registrations, contacts from external media programs, respondents to outbound marketing campaigns, etc.) in the current quarter to to drive our future $1M bookings target.  I love seeing numbers like this because they create a reality and urgency that's often difficult to convey without quantification.  Everyone's headset seems to change when the numbers are big and derived from known conversion metrics.

Instrumenting and operating your market engine in a systematic way requires thought and effort, no question about it.  What's more, you'll undoubtedly refine the underlying model and its levers as you gain more market experience.  That said, these are not optional investments.  Poor execution means you will be running the business on guesswork; good execution means you will create what's arguably the most valuable asset in your company's IP portfolio.


  1. bobpotter50

    Fred, I remember reading this when you first wrote it and then I read it again today. This is good stuff and clearly you are years ahead of most tech companies in moving away from 4 legged sales calls. Your advice has really helped me think through our own sales model.

    • fredh

      Thank you, Bob. The enterprise sales model is perfectly legit for some vendors and markets. But if you’re using open source, freemium or some other strategy to build a large, organic community, you’re going to need a volume engine to progressively engage with that base. I also believe that many companies in the enterprise space will gravitate toward volume models to smooth their quarterly bookings (strategic deals can be pretty lumpy) and improve lead flow.

  2. Brendan


    Can you explain – or do you have a post explaining – what allows a company to move from enterprise to volume? Is it price point, the nature of the product?

    My theory is even a company with a heavy touch sales cycle (which is what I assume enterprise model means) needs to accurately forecast using metrics and work hard at generating the best inputs to that sales cycle.



    • fredh


      You’re absolutely right that, regardless of your market model (enterprise, volume, hybrid), it’s critical to understand the numbers behind that model. In particular, you need to identify the stages in your cycle – end to end – and track conversion ratios at each stage. This is just good marketing and sales discipline.

      To answer your question, volume models generally tend to have certain things in common – commodity products, lower initial average selling prices, large populations of anonymous community members that need to be nurtured using low-touch automation methods (open source products typically have all of these characteristics, which is why I do so much work with open source vendors).

      I’ve found that the best way to construct a model is to start at the end and work backward. Doing so allows you to start with a concrete bookings goal and keeps you focused on that goal as you iterate through the modeling process. It will force you to recognize your good, bad and ugly conversion ratios, and it will make you think hard about value exchanges you can offer at each gate to improve those ratios. That process will ultimately help you decide which demand gen investments you can’t live without on the front-end of the cycle, what goals you should have on the leads that result from those investments, and how to measure your performance against those goals.

      I hope I’ve managed to answer your questions. Thanks for taking the time to ask.


Comments are closed.