3 Easy Steps to Identifying Demand Units within the New Demand Unit Waterfall Using MintigoAI

Almost a decade after the launch of the foundational framework for managing B2B marketing generated leads – the famous SiriusDecisions Demand Waterfall®Terry Flaherty and Kerry Cunningham, senior research directors at SiriusDecisions, launched the Demand Unit Waterfall last year at SiriusDecisions Summit in Las Vegas. The new Demand Unit Waterfall was a much awaited and welcome change in the brave new world of B2B marketing; a world that has seen continuous change, with over 6,500 marketing technology offerings, a shift in focus to Account Based Marketing (ABM), and the emergence of AI-powered marketing and sales applications.

The primary building block of the Demand Unit Waterfall is the ‘Demand Unit’ itself, defined by SiriusDecisions as “Buying Centers with needs that your solution fits & a Buying Group organized to address those needs.” In our recent webinar, Kerry and Vicki Brown (also of SiriusDecisions) walked us through their framework for identifying Demand Units using AI.

In this blog, I’ll recap how the MintigoAI platform can help you to identify the Demand Units for all of your products and solutions in three easy steps. As shown in Figure 1, we can breakdown the definition of a Demand Unit into three components:

  1. Target Accounts with fit for your products
  2. A business need that initiates a buying process
  3. People that are part of buying committees
Figure 1 – Three Components of the Demand Unit

Let’s start by digging into step 1. In this step, you will identify all accounts that are most likely to buy one or more of your products. We all know how sales and marketing have been selecting account lists for decades; usually a spreadsheet from your CRM is pulled by sales operations, vetted by managers and AEs, and finally a list of accounts is prepared and handed over to Marketing to find leads (or names). We typically apply our understanding of “who” buys from us, something that is often defined by firmographic and demographic information, and then we apply tribal knowledge to come up with the target universe. This methodology for building your target universe is like watching pixelated, standard definition 480p TV in this day and age; this means your target universe lacks accuracy and resolution. The impact is poor performing marketing and sales campaigns. Your campaigns will never achieve their expected ROI.

The precision and reach of your target universe can be significantly increased by using  artificial intelligence and machine learning (commonly referred to as AI/ML). Minigo’s AI platform uses its own insight data covering millions of businesses globally and insights from customer-provided data to build predictive models for all of your products using AI/ML. It ranks, scores, and appends every account/lead both from your sources (CRM/MAP) as well as its own universe of companies. The net result is that you now have access to the best accounts, those that exhibit a fit for your products. Figure 2 illustrates a typical output, a list of accounts ranked and scored by product. Each account is enhanced with data elements that make up the score/rank. This enables you to easily create a list of the best accounts for a given product, discover product bundles, or find cross-sell opportunities within your existing customer base.

Figure 2 – A sample output of score ranked and scored across multiple product.

Now that we have identified target accounts with fit in step 1, we need to identify accounts that show intent for your products – step 2. Intent is a hot topic and you can learn more about the basic concepts here. There are two types of intent; 3rd party intent and 1st party intent. Both types are used to identify the buying stage of a prospect. Mintigo’s solution makes it easy to apply intent data and identify the accounts that are either in early stages (mostly 3rd party intent signals) or that have moved into the consideration stage (mostly 1st party intent). Either way, you will always know where your prospects are in the buying process. Knowing the fit and buying stages  makes it easy for you to customize the engagement model. For example, sales should immediately follow-up with a high fit prospect in the consideration stage, i.e. an account with the  highest probability of buying your products and that is showing positive intent (both 3rd & 1st party). On the other hand, you may want to nurture prospects that are high fit but are not showing positive intent.

Now that we have the target accounts with fit and intent for your product, we need to find out if we can identify locations where the need exists. Let’s say IBM is your target account and is showing intent for your product. IBM is a very large global company with multiple LOBs and decision makers. It will help an account executive (AE) enormously if we can narrow down the location that is displaying buying intent behavior. Most intent providers, including Mintigo, will provide you intent location – i.e. the location where the intent signal comes from. Let’s say we see intent signals coming from Austin, TX and New York City, NY. Now the AE can focus on IBM locations in TX and NY. While this is a good approach, you need to be aware of several potential pitfalls of this location identification approach. Firstly, a prospect may be travelling or working in a remote office. The LOB or the functional area may be based in San Jose, CA but now you think it’s Austin, TX. Secondly, most B2B buying decisions are made by committees consisting of various stakeholders. These personnel may be geographically distributed, so relying on intent location to identify the source of a “functional area with need” should be done with caution as it may result in frustration and erosion of trust between marketing and sales.

Finally, we need to identify the buying teams, or groups of people involved in a buying decision, within the accounts that have both fit and need. In step 3, you will use lead based predictive modeling to identify the best leads/contacts within your target universe of accounts from steps 1 & 2. Lead based models use several persona attributes to identify the best leads by product. Mintigo’s platform uses AI/ML to verify and correct contact/lead information, automatically finding features such as quality of data, department and ranks, etc. that may impact your models. In the end, you are able to find the best leads/contacts from the best prospective accounts by product.

By following the 3 steps above – and using Mintigo’s AI platform – you will be able to identify Demand Units with good precision and efficiency. And by refocusing your efforts on finding these Demand Units, you will find that your B2B sales and marketing campaigns are much more successful.

Atul Kumar


Atul leads our innovative product organization at Mintigo. Atul spent several years conceptualizing and delivering data-driven marketing solutions at Oracle, Veritas, Symantec and ReachForce. His innovative work in the area of lead management and lead scoring has been widely published, and laid the foundation of current lead scoring models and predictive marketing methodologies. He co-founded SetLogik (acquired by ReachForce), a modern cloud data management platform. Atul holds a Ph.D. in Materials Science from the University of Cambridge, U.K.