Originally posted on VentureBeat
With all the talk about predictive-driven sales and marketing, a new question is emerging – which data is most valuable? Many B2B businesses are achieving unprecedented customer insight by leveraging all kinds of external demographic and firmographic data to see if a company is a good fit for their product. Some are pairing that with signals from their marketing automation systems and web analytics to predict whether a prospect might be ready to buy soon.
Now, another breed of “intent data” has emerged. External data providers like Bombora, The Big Willow, IDG and TechTarget are aggregating information about web visitors on B2B publisher networks to help businesses figure out when certain prospects might be in the market for their product. This kind of insight presents an exciting new frontier for data-driven marketing.
Intent Data Defined
As with any new data source, it’s helpful to have a clear understanding of what it includes and how it can be applied. Intent data generally falls in one of two main categories, which each best serves a different purpose:
Internal Intent Data (also referred to as first-party data) is the activity a company captures on its own website or through application logs. This kind of information usually contains highly predictive buying signals because the content is so relevant to the purchase decision – i.e. exactly what pages a prospect touched, which links they clicked on, and how long they spent on each page.
External Intent Data (also referred to as third-party data) is collected by publisher networks either at the IP level, or through user registration and shared cookies. These sites track the articles a user reads, content they download, their site searches, and potentially even comments they leave. For example, data might show that people from the ibm.com domain are viewing more articles than normal about “help desk software,” which could provide a hint about IBM’s software needs.
Sound a bit creepy? To some people, it is. There’s a reason that Google doesn’t open up its search database and LinkedIn doesn’t sell its social graph — that kind of personal data is super sensitive and core to their businesses. But many publishers are pushing the boundaries of privacy and finding new ways to monetize their traffic. Several have loosened their terms of service to gain the leeway to track individual actions and tell outsiders what an IP address or even a registered user is doing. This is something consumers should keep in mind if they don’t want their behavior to be monitored across the web.
Although it’s early days for third-party intent data, it seems obvious that the behavior bread crumbs people leave across the web must be valuable for marketers, especially if they indicate when a prospect’s interest is surging. But how do you turn this insight into impact? Some companies use intent data to generate lists of net-new leads that seem interested in a specific product category. This lets sales development reps target folks who are warmer than a cold list.
Another use case we’ve seen is leveraging intent data to glean insight into the existing prospects in a CRM database. If you find new clues about these accounts, you can potentially append behavioral score categories to applicable records. As a result, companies might be able to better prioritize sales outreach or personalize messages in order to boost click-throughs — for example by finding companies that are reading lots of articles about cloud computing and sending them a relevant email campaign on that subject.
Pitfalls to Avoid
As with any new type of data, there are always shortcomings to consider, which vary a bit for each of the above-mentioned use cases. Our company regularly tests new data sources, and we’ve taken a close look at third-party intent data to evaluate the benefits it can deliver for marketing.
For net-new leads, it’s important to consider that most external intent data is aggregated at the domain level. Reps won’t know who read the articles, and will have to spend time identifying the right contacts at each account in order to effectively prospect these leads. They’ll often have cold conversations and end up on wild goose chases, because a spike in articles consumed by a domain doesn’t necessarily mean that the relevant contact is interested or that the company is actively evaluating and ready to buy.
In the second use case, a business wants to know more about prospects that are already in its database, but might not be actively engaging on the web site. Questions to think about here are coverage and accuracy. What percentage of sales records have a match in the most recent intent database? When we ran our tests, 86% of the companies in our sample set had no third-party intent data associated with them. Companies should also look for activity around the subjects that are relevant to them (e.g. “people reading articles about help desk software”). However, when we filtered down to even the most popular topics, match rates often dropped below 2.4%. Generally speaking, when evaluating data, the goal is to find insight that impacts a larger percentage of the pipeline so it can drive a material impact.
The next question to ask when a signal is present is whether it’s accurate. Is a surging account highly correlated with a prospect being in-market? We had an intent data vendor match records for a diverse set of our customers, and then ran historical backtesting. What we found was that the intent flag was no better at predicting opportunities than random chance. We’ll continue to test intent, but in the meantime we recommend focusing on use cases where the risk of being wrong is low. For example, marketing teams might use third-party intent data to personalize emails and advertising campaigns, which could increase click-through rates (a great way to generate more first-party intent data).
Keys to Success
If you’re considering external intent data, the most common mistake you can make is to jump in without defining clear use cases or your ideal result. While everyone knows intuitively that there are prospects they aren’t seeing and would like more insight, if you want to drive real impact, it is important to start with your criteria for measuring success and work backwards. For example, with the net-new use case, we’re optimistic that there could be value in intent data – it’s just a question of how many leads you can generate, how much sales effort it takes, and ultimately whether it’s worth the cost per good lead. That type of framework provides an objective way to compare intent data with other lead sources or list buys.
Conversation like this help to push the constraints of today’s data and systems, and expose the most promising opportunities to drive top-line growth. As David Raab puts it, “marketers and the technologists who support them need both the ideal vision of how things would work in a world of perfect data (which isn’t the same as a perfect world!) and the realistic understanding of what’s likely to be practical within their planning horizon.”