Using Web Analytics to Identify Critical Behaviors in Converting Prospects Into Leads

Google and CEB helped change the way many website owners think about prospects and leads in 2013 when they published their research showing that amongst 1,500 business people involved in B2B purchases, the buyers did 57 percent of their research before reaching out to communicate with prospective vendors. This research helped demonstrate how it is as important to understand how prospective customers are interacting with our websites before we know who they are as it is to have a Web-informed sales process. But it can be difficult to perform a thorough analysis on prospect behavior on a website simply because we are not sure how to identify these anonymous users.

Web analytics provide amazing insights into how general visitors behave on your website. Out-of-the-box reports are great at showing how visitors choose to navigate across a site, the types of content they consume, and the tools they interact with. With segmentation of visitors that have filled out lead forms, started applications, or started cart transactions, we can start to see the differences between prospects, leads, and customers when it comes to engagement with a site.

However, in creating these important segments it is much easier to analyze lead behaviors and customer behaviors, as these visitors have essentially raised their hand and said “here I am” by taking steps such as filling out forms and making purchases. Prospects, on the other hand, are a little more difficult to identify. Often times when analysts are examining prospect activities such as content preferences, frequency of visits, and number of visitors, the prospect segment is essentially made up of “the others.” The prospect group is typically made up of the users that do not fit into other groups such as career seekers, past leads, current customers, and investors. Creating a segment with such a vague and encompassing definition may decrease the usefulness of any analysis as a large percentage of users within the segment will never convert to a lead regardless of how persuasive your offering or how persuasive your site is simply because that is not what they are there to do. This leaves website owners with a difficult and important question: How do we more effectively analyze the behaviors of prospects?

The first step in tracking prospects is knowing who is not a prospect. It’s important to persistently track website leads and current/past customers when they perform actions such as fill out lead forms, make purchases, or login to customer areas. In Google Analytics we do this with Custom Dimensions and in Adobe Analytics we do this with eVars. Aside from leads and customers, there are a few other segments of visitors that are not likely to ever be viable prospects. I filter out career seekers and investors by their content views. A user that views four pages in the job section of a website and never views a product page is not likely to be a real prospect. I also filter out competitors by their Service Provider names. In Google Analytics it’s great to create a separate View for prospects which completely filters out known leads, customers, career seekers, and competitors.

Once the known leads and customers are segmented out, we begin to create behavioral filters for our prospects. Below is an example of a prospect segment I use often.

segment-prospects

In this segment, you can see I have created the following filters.

  1. I’ve used the persistent tracking to remove all known leads and customers.
  2. I’m assuming that anyone that is a serious prospect will be looking at a minimum of one product page.
  3. While not all prospects will look at organizational information, I find that in lead scoring users that do view organizational information typically turn out to be the highest quality leads.
  4. For this example, I am looking at users that visited the site between two and seven times. The number of visits will vary significantly based on the price and complexity of the product or service. For many products, prospect research can be performed in a single visit and many leads are converted on their first visit. In this example, I am assuming that the product is complex and requires more deliberation before a prospect decides on specific vendors. A good place to look for how much research users perform before filling out lead forms is to look at the average number of visits and the ratio of visit counts for users that have converted to qualified leads in the past.
  5. For this example, I am also filtering out all of the typical ISPs. While many qualified prospects can be doing product research from home, I want to see specific companies visiting the website.
  6. I find that qualified prospects doing multiple visits will typically perform all of the visits in a short time period such as a week or month. This will vary significantly by the length of the sales cycle.

Once our behavioral segment is in place I will look through the Service Provider report to ensure that I can see organizations that I know to be viable prospects. For example, if the company sells construction equipment I will look for construction companies in the Service Provider report or if the company sells software for financial institutions I will look for banks, credit unions, and brokerages. Below is an example Service Provider report of a company that sells data storage systems specifically for large financial institutions.

service-provider-report

Once I see multiple good prospects in the Service Provider report I will start to look at the behavioral data to find attributes of the sessions. The following reports provide great data on how these prospects learn about products and research potential vendors.

  • Session/User Attributes: Understanding data points such as product views per session, time on site, and device preferences starts to give us an idea for how prospects are looking at the site as a whole.
  • Service Provider: By grouping multiple users from the same company we can start to see information such as the number influencers and decision makers in a sales cycle. Looking at company visitation overtime we can start to see the frequency of product research and from that infer how close they are to a conversion event.
  • Click Paths: Reviewing the sequence of the page views tells a lot about how prospects research vendors. The depth of a click-path and how decentralized the click-paths are can provide great insights on how much information prospects need before reaching out to a vendor.
  • Content: Viewing page popularity, time on pages, and exit points give us an idea of which types of content are necessary verses nice-to-have for prospects comparing vendors. Grouping content into categories such as product specifications, white papers, case studies, and industry research allows us to know what types of content are most critical in vendor research.

While it is very difficult to know who is visiting a company website to perform vendor and product research before they are known leads, Web analytics tools are getting better at enabling website owners to understand how this large anonymous segment of users consumes content and utilizes website tools. With all of this information in hand we start to better understand how our prospects research potential vendors before we know who they are. We can start to see which types of behaviors indicate that a prospect is likely to become a qualified lead verses and an abandoned anonymous user. With accurate data on the number of prospects visiting a site we start to understand how much of the site needs to be focused on vendor research and product education verses lead nurturing and sales assistance. While most B2B websites are focused heavily on lead generation, organizations that understand prospective customers need to perform due diligence research before converting to a lead will ultimately be better at educating their prospects and turning them into high-quality customers.

Related reading

tencent_emily-ma_featured-image
kenneth_ning_emarsys_featured-image
bounce-370x229
site search hp
<