What’s the Most Powerful Use of Big Data for Lead Generation Marketing in 2013?

What are the most powerful ways B2B lead generation marketers will be using big data in 2013?

  1. Some may say attribution – understanding which channels, content, and messaging is working best.
  2. Others may claim personalization and engagement scoring – delivering the right message to the right person at the right time.
  3. A few may argue sales efficiency and effectiveness – giving the sales department more relevant information at the right time, helping to close more business faster.
  4. But ultimately, I believe the most compelling case can be made for closing the loop – from marketing lead to sale and ultimately maximizing true ROI.

To be fair, I introduced these use cases in my last post, “What’s the Big Deal With Big Data.” And I do believe closing the loop is the most compelling, so let’s take a closer look here.

lightbulb-dataTypically in a high-consideration (B2B), multi-touch sales process, most marketers are only tracking web activity to lead. This only tells a part of the story. In reality, marketing campaigns are still interacting with people into the late stages of the buying process. Connect your data from lead to sale; and directly tie marketing activity to revenue and you get the full picture.

As digital marketing techniques and technologies have evolved, front-end metrics from web analytics tools often fall short in delivering enough high-value insight to drive business decisions. Deriving high-value business insights from disconnected datasets and offline sales processes often requires time-consuming, complex, and expensive software implementations. It doesn’t have to be long, complicated, or expensive, but you do need help – either internally uniting from IT and database experts, or outside help from specialists with experience doing just this.

Let’s explore why and how this will create enormous impact within your organization by sharing highlights from a recent client case study. First, why do it? What business questions will you be able to answer after closing the loop?

  1. What opportunities exist to boost sales conversions vs. just increasing traffic or lead conversions?
  2. How do content, channels, and user experience affect lead quality and conversion-to-sale?
  3. How do buyers behave differently from non-buyers across campaigns, channels, and your website?
  4. Where should I invest my marketing and media dollars to produce the largest ROI, not just more traffic or conversions?

Let’s say these are your key requirements: Must be able to track a visitor through their customer journey, record their website activity via web analytics, and then continue to track that activity through lead capture and sale.

In this case study, these were the key challenges: Connect lead management, digital marketing programs, CMS, CRM, and point of sale, all of which currently operate as disparate systems. Also, a limited budget that ruled out a vendor-based single data warehouse solution.

So what were the key steps to implementation? Let’s get ‘er done. To start, rather than utilize a single data warehouse system as an integration point and repository of data, we took a hybrid approach by correlating data across multiple systems. By tagging the user with a unique identifier, and ensuring that key was retained across all the systems, we connected web activity to offline activity by aggregating this pre-correlated data in an SQL database. It was simpler, more elegant, less time-consuming, and less costly – although this approach will not deliver depth to answer infinite questions of your data – something a tool like Adobe Insight can do.


Customer intelligence solution: this low-cost, low-complexity approach tags the user with a unique identifier that follows the user through all systems and aggregates the data in an SQL database.

Implementation required a few key steps:

  1. Write JavaScript code to set a cookie from your website and pass it as a custom variable to your analytics platform, in this case Google Analytics. This ensures the unique customer key will be generated on the customer’s first visit, and then passed with every click, event, and goal completion that Google Analytics captures. It also ensures that the customer key will be consistently retained in the user’s browser across sessions.
  2. Pass the customer key to the client’s lead management system. Most have web service API for lead form submissions.
  3. Modify your reports and data extracts to automatically load a SQL database with web analytics and lead data. Since leads are eventually converted to sales in the lead management system, you now had all the data needed to perform a closed-loop analysis.
  4. Push some of the key findings from this data into your real-time dashboard system. This unlocks key sales and marketing data from complex and hard-to-read spreadsheets and made it easy to view by anyone, anytime. By delivering this data passively, we were able to focus on our next and most important step, mining the data for insights.

This solution had low implementation costs and low complexity. No expensive servers or configurations were needed and no modifications to infrastructure necessary.

What key insights did this deliver? A short summary:

  1. Sales conversion rate was much higher than they had ever anticipated. This meant that lead-to-sale nurturing was more important than initially thought.
  2. Almost 50 percent increase in email engagement from visitors who ultimately became buyers = nurturing is important.
  3. Purchasing affiliate leads were rubbish. While low-cost, these leads converted to sale at a very low rate, and thus produced little ROI.
  4. Best media dollars spent – PPC. By setting an average order value (AOV) for products sold, we were able to immediately see sales generated from PPC advertising, less the ad spend and agency fees = 100 percent return.
  5. Search was clearly the highest performing channel, with 65 percent of buyers having touched organic or paid search at some point in the customer journey.

Best use of big data in 2013? You bet your data it is!

Related reading