Has Your Hot Prospect Well Run Dry?
- Marketers should consider several use cases when determining if predictive prospect sourcing makes sense for their organization
- Predictive marketing applications solve three primary problems
- There are shared capabilities that are required to be successful across all predictive applications
Editor’s note: This is the first blog post in a series on predictive marketing applications.
Predictive analytics is quickly evolving to become the equivalent of a marketer’s Swiss Army knife. Tried and true applications include prospect sourcing and scoring, while lesser-known applications such as smart segmentation, predictive persona building and tactic matching are increasingly being explored and tested.
When assessing which “blade” in that Swiss Army knife to use, marketers must fully understand the problem that needs to be solved through analytics. Predictive marketing applications solve three primary problems:
- Too few prospects – how to source?
- Too many prospects – how to prioritize?
- Conversion challenges – how to engage and convert?
Prospect sourcing addresses the first problem. Despite advances in inbound marketing, most B2B marketers struggle with sourcing an adequate volume of demand entering the Demand Waterfall®. As a result, most B2B organizations still must execute extensive outbound marketing to engage prospects. Our research shows that outbound tactics can be prohibitively expensive (e.g. average cost per lead by lead level) if they are not planned and executed effectively. Marketers are also tasked with enabling their sales counterparts to manage prospects from a variety of sources and with varying degrees of familiarity and engagement.
Marketers should consider several use cases when determining if predictive prospect sourcing makes sense for their organization:
- What it is: Sourcing high-propensity prospects is often referred to as lookalike modeling, which builds an ideal prospect template based on historical marketing and sales performance. The template is then used to match and source new accounts and contacts.
- How to use: Identify viable accounts and contacts for targeted outbound marketing efforts. If you don’t have a strong outbound lead development function, consider digital marketing tactics to predictively source prospects to drive inquiries for followup.
- What it is: Support account-based marketing strategy by maximizing the investment in developing high-propensity targets within key accounts.
- How to use: Care must be taken when delivering predictively sourced accounts and contacts to sales. Even though these prospects may be more likely than most to eventually buy, they are not yet inquiries and may not be familiar with the seller organization and offering. Having and enforcing service-level agreements for volume and quality of followup by lead development is critical to maximizing the investment in predictively sourced prospects.
- What it is: Prospect sourcing can be used to deliver high-propensity cold prospects directly to sales reps or territories together with pre-packaged plays (e.g. competitive replacements, discounts) intended to drive rapid entry into the pipeline.
- How to use: When coverage models are not adequate to guarantee complete territory coverage – or to address market triggers (e.g. M&A) – provide sales reps with high-propensity targets to jump-start new sales reps in territories.
When determining if your organization is ready for predictive prospect sourcing, consider these shared capabilities that are required to be successful across all predictive applications:
- Data. Predictive models require data in two categories: predictor and outcome variables. You’ll need at least a thousand records populated with each.
- Data science. All the data in the world will not contribute to a good predictive model if it can’t be matched to prospect data. The best predictive models draw data from a huge variety of internal (MAP, SFA, Web site) and external (Web searches, social media conversations, syndicated content usage) sources.
- Mathematical/statistical models. The third component of predictive analytics is the math itself. While the mathematical modeling is sophisticated and requires years of intensive training to master, the techniques themselves are generally highly flexible and may be applied to a huge variety of predictor and outcome variables.
Most B2B organizations either don’t have – or aren’t in the position to apply – the necessary resources for building their own predictive application. Fortunately, there are nearly a dozen predictive marketing applications standing by – so, if you’re seeking a predictive prospect sourcing solution to replenish your dry prospect well, we can help you review your options!