One of the great things about my job is that I get to talk to a lot of of different clients about a lot of different things. My remit is pretty broad which means that I get to look at a wide range of marketing challenges and advise clients on the best ways to tackle them. A lot of those challenges are related to the changing B2B technology buying cycle. For some clients it means changing the way they create demand and generate leads, for others it requires a change in the type of content they develop, while a different set of clients are making fundamental changes to how they sell. All of these are covered as part of Gartner’s ongoing Future of IT Sales special report.
One of the most profound impacts of the new buying cycle seems to be around leads and how they are scored. For the last five+ years, marketers have used a two-dimensional model based on demographics (title, company, size etc;) and behavior (what they downloaded, events they attended etc;). Some marketers switched from demographics to firmographics if their CRM Lead Management system allowed for account-level (rather than individual) scoring, but the model has pretty much stayed the same. Sure marketers made minor tweaks in terms of the value they ascribed to any particular attribute, but the basic premise didn’t really change.
We know that buyers are doing more research on their own, before talking to providers, which is partially responsible for the prevalence of content marketing. The upshot is that more content is getting consumed and that’s throwing off the lead scoring process for many providers as more individuals fit their definition of “qualified.” Marketers are falling into the MQL Trap and passing over more leads than sales can typically handle. So salespeople can’t prioritize their efforts on the best leads, raising the possibility they engage too early with prospects who aren’t ready to engage, or engage too late with the ones that should have been called immediately.
The good news is that this problem is solvable. Marketers have long known that the more they know about a prospect (especially at the company level), the more likely they can determine how likely they are to buy and what kind of fit they are for a particular set of solutions. This data was out there in social media and in public and proprietary databases. But it was costly to acquire and required data scientists to build models that ingested the data (along with the existing data they already had).
Both the cost and complexity were certainly deterrents. But now, there is an entire class of solutions available to help solve this issue. Predictive Lead Scoring applications utilize models that predict how likely a given company is to buy a particular solution. The providers in this space tap into a multitude of public and private data sources and split the costs among all of their customers. Then they add that big data to the data from CRM Lead Management and Sales Force Automation (SFA) systems (especially historical transaction and win/loss information) and create a predictive model that leads can be scored against.
This is the subject of my new research note, “Tech Go-to-Market: Using Big Data to Focus on the Right Prospects Can Improve Sales and Marketing Effectiveness” (Subscription Required). I talked to some technology providers that have been early adopters of these type of solutions. They’ve all been able to improve their sales and marketing execution as a result. Each company found several unexpected attributes that wound up being highly predictive. Many of the providers have replaced their lead scoring models, while others are keeping their existing models, but making sales aware of the additional information. A few companies are even leveraging the models higher in the funnel, to evaluate the quality of leads from events and content syndication programs.
Their are a lot of interesting ways this can be used. Product marketing can use information from the models to improve segmentation and targeting (and subsequent messaging) as well as sales enablement. Other marketers can improve their lead nurturing efforts by targeting specific content and individuals based on specific data attributes. And these models can certainly be run to determine whether an existing customer would be ripe for a cross-sell or upsell campaign.
In the research note, I mention the idea that this might be a “Moneyball” moment. If you search the Internet for references between technology marketing and the famous book (and later movie), you’ll see lot of them. But I would argue that we didn’t have this moment just because we started using data to make marketing decisions several years ago. To me, the real point of Moneyball is that Billy Beane bucked conventional wisdom by making decisions (based on data) that flew in the face of conventional wisdom and the way players had been evaluated.
I believe we could be at a similar point now. Providers must evolve to address the new buying cycle. Now that the data is more available and accessible, it’s making both marketers and salespeople re-think their approaches. We’re still early enough that driving this kind of change won’t be easy.
But remember, it didn’t take very long for other teams to copy Billy Beane. While Oakland could never quite get over the top, Beane’s protege, Theo Epstein, applied the Sabermetrics model for his Boston Red Sox team (and their much larger budget) and won a couple of World Series. Now nearly every sports team in every major sports league around the world uses predictive models to evaluate players. Maybe that’s what we will be saying in a few years about lead scoring.