ABM will help focus your company’s efforts; but even so, B2B marketers still struggle with the challenge of gaining—and maintaining—intelligence on their buyers at scale. The more contacts you accrue the harder it becomes to focus on targeting the right leads, or even identifying them at all to begin with.
This problem grows exponentially with your company. At first, it’s possible to manually keep tabs on a small pool of contacts. What information you don’t know about them can largely be gleaned by scouring the Web.
But have you ever tried Facebook stalking a mid-market business’s entire database (never mind an enterprise company)?
As your company grows, CRM and marketing automation become a crucial part of mapping accounts and managing campaigns. When that happens, the sheer numbers involved render manual methods of gathering information absurdly inefficient. There‘s just so much information to process: from first-party data (whether from inbound campaigns or outbound prospecting), to unstructured, anonymized data like online behavior and social media profiles.
Predictive modeling: A solution — but be careful!
Predictive analytics has helped countless B2B companies overcome these challenges. In fact, many ABM leaders like Engagio’s Jon Miller argue that ABM just can’t be done without predictive analytics.
An effective predictive model will deliver higher ROI and greater efficiency, saving dozens of hours and generating more, better quality leads. Predictive models give marketing reps a far better idea of which leads are sales qualified, and which would be better off being sent for further nurturing — leaving sales to focus on only the most promising leads.
But there are still three underlying issues which require more robust Artificial Intelligence-powered solutions than basic machine learning predictive analytics.
1. Maximizing the quality of the underlying data in your database
What if that formerly influential VP-level contact you’ve been targeting just got fired, or moved to a different company? How much time will you have wasted on a dead-end lead?
It’s hard enough keeping a clean database as it is, what with all the incomplete or false information from inbound marketing campaigns corrupting your system. Now throw into the mix the fact that B2B professionals are such slippery fish, and it’s clear just how out of date your databases likely are.
Think about how many B2B professionals on a quarterly basis get promoted, fired, move companies, switch roles, retire, or for another reason are no longer in the same place within the decision-making hierarchy they once were. They just can’t sit still.
Which begs the question: How can you enrich your data in real time to keep it relevant? Most predictive analytics vendors don’t have a solution to this problem – they simply build models based on your (gradually deteriorating) data to provide (gradually deteriorating) predictive models.
But AI hold so much more potential than building basic predictive scoring models. For example, AI can enhance your data management, by scouring multiple first and third party data sources to build a clear picture of your audiences – and ensure any changes in their status are reflected in the data in real time. (At Leadspace, we call this Audience Data Management.)
This is a really big one: Traditional predictive analytics systems take a black box approach — they tell you what to do, but not how they came to those conclusions. But such an approach isn’t entirely reliable, and not only for the reasons already mentioned above.
We all use GPS systems like Waze to help us get from A to B. But we also, you know, look at the road from time to time (well, most of us anyway).
The point is, no amount of technological ingenuity can replace good old-fashioned human knowledge, experience and intuition. It’s not enough to know who is theoretically a good lead based on previous behavioral metrics — you need to know why, and how best to convert them.
Once a lead makes it to a sales rep, they need all the information about that person they can get to know which leads to contact, and why. Without that, they’ll just be shooting from the hip — albeit (if your predictive model is accurate) to reasonably qualified leads.
This problem becomes all the more acute if you’re pursuing ABM, where customized pitches are a requirement, and where knowing the right prospects is double challenging: 1) You need to find the right account and then 2) the right influencers (leads) within those accounts.
Again, applying AI to your data in a holistic fashion — that is, not simply to build one-time predictive models — can help you discover new “hot” prospects who resemble your Ideal Customer Profile within your named accounts.
3. Understanding accounts, as well as the individual leads within them
The term “account-based marketing” can be a little misleading. While ABM views accounts as markets of one, in order to actually engage with them you obviously need to be aware of the key influencers within the company. That means to say: people.
That’s why lead-to-account matching is a key ingredient of ABM. You could have a fantastic list of prime accounts, but be pitching to the wrong people inside those accounts. Or your database could be filled with great leads, but you can’t match them to the right accounts. Either way your ABM strategy is unlikely to get very far.
Traditional predictive analytics vendors don’t provide lead-to-account matching capabilities. Instead, they tend to focus on scoring either leads or accounts, and leave it at that. But in today’s B2B marketing environment of increasingly sophisticated approaches to demand gen, such an “either or” approach has become obsolete.
Yet again, applying AI to your company data can match leads to accounts to a degree of accuracy — not to mention speed — that no marketer could achieve alone.
So it’s clear that, despite its usefulness, traditional predictive analytics has some clear limitations.
For more useful tips on pursuing account-based marketing, download our free ebook — The Modern Marketer’s Guide to ABM:
Top image courtesy of iStock