B2B Intent Data: Limitations & How to Use it

For the past decade, we've seen B2B marketers buy into intent data tools as the "holy grail" of being able to identify in-market buyers. The very software category name "intent data" is exciting as it suggests a panacea whereby we simply sign-up and provision one of these so-called intent platforms and voila -- identifiable leads!

However, the reality is more murky in terms of data quality and actionability. This article is not to harshly criticize the intent data category, but to offer B2B marketers with some practical advice on what they should really expect from intent data sets and how to rightly use them.

First, let's correct a misconception up front: accounts that demonstrate increased online search and content consumption are in-market to buy. Not true -- this only signals increased interest in a topic. Further qualification is required via live conversation or an accurate predictive model to determine in-market buyer status. Only accounts with verified in-market intent are Leads, all others should be classified as Learners

Second, let's be aware of the limitations of intent data:

• Intent data suffers from reverse IP lookup inaccuracy (made worse by work-from-anywhere in the post-COVID era), so care must be applied when tailoring communications
• Intent data may be able to identify accounts, but you still need a way to de-anonymize individual users to make the data actionable for Sales follow-up

For additional reading on limitations and challenges of intent data, I suggest these 3 articles:

https://www.forrester.com/blogs/the-10-biggest-intent-data-mistakes-for-b2b-marketing-and-sales/

https://www.gartner.com/en/digital-markets/insights/how-to-purchase-intent-data

https://martech.org/the-false-allure-of-b2b-intent-data/

How to think about Intent Data

For that past decade, pioneering vendors like 6sense, Bombora, Mintigo, and Demandbase have been competing to establish themselves as having the most accurate “intent data” for marketers to use to identify in-market buyers for subsequent ad targeting and anonymous user matching [see 2015 MarTech article]. However, when we look closer at the data we find that it is not a leading indicator of buying intent after all. Instead, it’s a collection of website visits and content consumption that signal topical interests, but there is no direct causal link that connects these interests to actual in-market buying intent without live verification or applying additional predictive modeling.

For instance, millions of business professionals perform online searches and consume content (podcasts, webinars, guides, newsletter) on a wide variety of topics to simply learn. Learning is not the same thing as buying – so classifying everyone with a surge in topical search as a buyer is logically flawed. This is why MA and CRM platforms inside many B2B companies are filled with tens of thousands of learners who are misclassified as leads.?

This should really be renamed to “activity data” – but that would have not been attractive to VCs back in 2014 when many of these companies were funded. So what we have today is a category (intent data) that goes unquestioned by many B2B marketers.

How to use Intent Data

The most frequent use of intent data is to try and spot potential in-market accounts so that the information can be passed to Sales for outreach. However, as I've noted above, this is built on a bit of flawed logic that these accounts are actually in an active buyer journey -- thus, more qualification is needed first.

For this reason, it's important for Sales and Marketing to agree on how they will use intent data so that they can orchestrate their efforts for maximum efficiency and effectiveness.

I'm not a fan of Marketing using intent data to "identify leads to pass to Sales" because of the uncertainty in the account actually being sales-ready. This only creates unnecessary work for SDRs to filter these "false positive" Leads and route them back to Marketing for reclassification as Learners. For that reason, I prefer to allow Sales to select the accounts that they believe are in-market (signals + predictive model) for outreach.

The following is my recommended approach for orchestrating intent data use:

1. Sales Prospecting: The very nature of outbound prospecting is that sales reps are going to call on a lot of ICP targets that are unfortunately not in-market to buy at that moment of time. For this reason, intent data plus a predictive model can be useful to Sales teams to identify potential accounts for outbound calling and improve their chances of timing their outbound prospecting correctly. Accounts demonstrating increased levels of online search and content consumption aren't guaranteed to be in-market -- but it does offer Sales a better starting point for focusing their prospecting focus.
2. Demand Generation: Because Sales already has a difficult job of trying to find in-market ICPs and time their communications at the right point in the buying journey, Marketing should NOT make the problem worse by flooding Sales with unconfirmed leads. Instead, these “intent signals” should be used by B2B Marketing teams to classify a non-customer account as either a Lead or a Learner. Leads are pretty obvious: they should be "hand raisers" that want to talk to Sales and the only way Marketing is going to know this is if they actually qualify them (live conversation, email, direct message, or online form submission). Learners are a group of ICP-fit accounts that are not yet ready to buy so they require patient nurturing to build a know/like/trust relationship by inviting them into (non-selling) conversations to share insights, connect them to peers with similar interests, and facilitate engagement that is centered on being an information “concierge." Through this engagement, the B2B Marketing teams can uncover whether the account is actively in-market, or simply learning about topics to increase their internal knowledge base.
3. Customer Marketing: Monitoring intent data against a list of your installed base of customers can help identify those that may be in-market to buy additional services (upsell, cross-sell), or they may be evaluating competitors for the products or services they already buy from you. In both cases, you want to protect and grow your existing customer base and intent data with predictive model support can be useful.

A Word of Caution

With vendors being excluded from a majority of the buyer journey, many have turned to using intent data a new way of crashing the party. Instead of earning buyer trust, vendors believe they can simply purchase third-party data “signals” that indicate a buyer is in-market so the vendor can intercept (interrupt) the first two-thirds of the buying process when buyers are performing self-directed research and conferring with peers.

Interestingly, research by SaaS ABM provider 6sense shows this may actually backfire as hijacking the early portion of the buyer journey actually leads to a lower deal close rate.

As with any tool, there is no silver bullet. Data inaccuracy and gaps (account IP identification), inferential intent assumptions, and trying to map accounts to individual user identity remain a challenge.

Be sensible in how you approach intent data for what it is: a data input that has its limitations.

Good hunting. ????