7 Data-Driven Reasons Why B2B Demand Gen Fails

White Paper 7 Data-Driven Reasons Why B2B Demand Gen Fails STEVE BIAFORE Chief Science Officer, LeadCrunch www.leadcrunch.ai

7 Data-Driven Reasons Why B2B Demand Gen Fails Figure 1. B2B Transaction Probability Matrix(Approximately 50 trillion entries) In this matrix [see video], sellers are listed along the y-axis and buyers along the x-axis.. Naturally, sellers would want to speak to buyers with the highest probability of transacting with them. Estimating these transactions probabilities requires processing an enormous amount of data. Once we have probability estimates sellers can ask high value questions like, “If you analyze my row, a vector with 10 million values, which companies are most likely to buy from us?” An accurate matrix can potentially drive conversion rates from 1% to 3% to 5% and beyond. Now that’s math worth doing. www.leadcrunch.ai 3 www.leadcrunch.com

3 Figure 1. B2B Transaction Probability Matrix(Approximately 50 trillion entries) In this matrix [see video ], sellers are listed along the y-axis and buyers along the x-axis.. Naturally, sellers would want to speak to buyers with the highest probability of transacting with them. Estimating these transactions probabilities requires processing an enormous amount of data. Once we have probability estimates sellers can ask high value questions like, “If you analyze my row, a vector with 10 million values, which companies are most likely to buy from us?” An accurate matrix can potentially drive conversion rates from 1% to 3% to 5% and beyond. Now that’s math worth doing. 7 Data-Driven Reasons Why B2B Demand Gen Fails

7 Data-Driven Reasons Why B2B Demand Gen Fails 2. THE DATA ABOUT THESE ENORMOUS NUMBERS IS FLAWED To deal with big numbers like the above, you need Big Data. Big Data helps you paint a mathematical picture of what your B2B environment looks like. Unfortunately, most of the widely available B2B data elements are deeply flawed. The three most prevalent data elements used to build B2B lead lists are industry codes, company headcount, and top line revenue. These three elements are flawed in not only the way they are designed and collected but also in the way that they are used. The top 3 US-based industry coding schemes are SIC codes, NAICS codes, and “home-grown” taxonomies, such as the one used by Linkedin. Each of these has serious design limitations. First and worst, the logic used to assign codes lacks rigor and consistency. Second, these coding schemes lack a reliable way to measure the relative similarity or difference between industry codes. Third, there are no incentives for companies to verify the accuracy of their own data and no central system within which to operate. Therefore, industry codes are frequently vague or simply incorrect. Headcount and revenue data suffer similar woes. First, category bins are created with arbitrary ranges that limit precision. Second, as we saw with industry codes, there is no system to verify data although the government does provide some degree of pressure to induce accuracy. Therefore, headcount and revenue outside of publicly traded companies are generally inaccurate or outdated. Despite these limitations, B2B companies still rely heavily, if not exclusively, on these three data elements to extract data for the top of their sales funnel. The three most prevalent data elements used to build B2B lead lists are: ● industry codes ● company headcount ● top line revenue These three elements are flawed in not only the way they are designed and collected but also in how they are used. www.leadcrunch.ai 4 www.leadcrunch.com

7 Data-Driven Reasons Why B2B Demand Gen Fails 3. THESE NUMBERS ARE ALWAYS CHANGING An often overlooked but critical aspect of data is time. Time raises essential questions such as, “When will this company be ready to buy a new accounting system?” or “When will this company be ready to expand into international For many B2B sellers, markets?” Velocity, a related concept, is also important. Some companies move the prospect’s direction quickly while others move slowly. and velocity matter more than its current For many B2B sellers, the buyer’s direction and velocity matter more than its state. current state. But even the simplest form of temporal information, trend data, is difficult to find. And if the underlying metrics relating to a company’s current state are inaccurate, computing accurate trend data is even more difficult. Temporal data is critical in B2B sales and marketing because it can unearth “sales triggers” or signals that a company is ready to buy a particular product or service. A sales trigger for one B2B seller might be a sharp increase in headcount, for instance. Conversely, a sales trigger for another B2B seller might be a decrease in headcount. To obtain any sense of trend you need at least two data values at different points in time. Remember when we spoke about 50 trillion points of data a minute ago? If you need an entire snapshot of a company at two different time points, you just doubled the amount of data you need to process. You can see just how much data we need to wrangle in order to gain even a basic picture of how companies change over time. 4. THESE NUMBERS MEAN SOMETHING DIFFERENT TO EVERY SELLER The quality of a B2B lead list is not just a function of the data that describes the buyer—it is also a function of the data that describes the seller. What a seller sells and how they sell it plays a key role in determining who is likely to buy, when they are likely to buy, and what message will move them to buy. This means that two different sellers of the same exact product may get different buyer target recommendations. In fact, we’ve actually seen this phenomenon at LeadCrunch with our own customers. And now if you factor in at least two different time points for the seller, we double our data again. www.leadcrunch.ai 5 www.leadcrunch.com

7 Data-Driven Reasons Why B2B Demand Gen Fails 5. THE ENVIRONMENT MATTERS TOO Not only do we need to analyze buyers and sellers, but also the environment in which they operate. There are many environmental forces that impact B2B transactions including the economy, technology disruptors, competitive forces, and government regulations. These environmental factors are not trivial and can have a profound impact on which buyers are most relevant to a seller at any given time. For example, an economic shock can raise the profile of some companies and put others companies out of business entirely. Understanding the environment requires more data still. 6. THE DATA ELEMENTS HAVE INTERACTIONS IN A MOVING TIME WINDOW Think for a minute about all the different data elements we have discussed. Now think of the incredible amount of data needed to track them as they change and interact. Sellers change, new products are introduced, competitors buy one another, new fiscal policy is enacted, and currency fluctuations open and close opportunities daily. The combinations of seller products with different buyer contacts is enormous. Even during the time we evaluate these two factors, five other factors might change, each of which have differing effects on buyer relevance to the seller. Data is required to see and understand these changes. As you can see, the our mountain of numbers keeps growing. If we cannot get the 7. THERE IS PULL TOWARD BIAS AND STATUS QUO data we truly need, In the face of overwhelming uncertainty, we humans tend to cling to what we know. we fall back on the If we cannot find a way to measure what truly matters, we just measure what we can data we have, whether it matters or not. If we cannot get the data we truly need, we fall back on the data we have, whether or not it accurately captures reality. whether or not it accurately captures In summary, there are vast seas of uncertainty created by the countless, unreliable, reality. ever-changing numbers that describe your markets. Sellers, sailing these high seas, are often tempted by the siren call of small islands of good data. Unfortunately, heading to these islands comes at a serious price—bias. Sure the data is good but what is missing from this data? Not surprisingly, most sellers have already docked on these islands instead of assembling their own high quality data. Therefore, the entire B2B seller market all lives on the same islands of biased data. What was true for ancient explorers is true for B2B sellers. If you want to discover great treasure, you have to go where others won’t go. Great explorers use the little data they have to sail their ships into uncharted territory where they create their own data. This data is hard won but it ends up yielding real treasure in today’s economy—proprietary data assets. www.leadcrunch.ai 6 www.leadcrunch.com

7 Data-Driven Reasons Why B2B Demand Gen Fails HOW TO CHANGE YOUR B2B DEMAND GEN GAME By now it should be clear that demand gen programs suffer for good reasons. They lack access to the breadth of data needed to develop quality leads as well as the machine capability to mine this data. LeadCrunch’s goal is to help you fix these problems so you can create effective lead lists and close more deals. Following is an outline of how to increase your close rate by tackling the relevant Create better data math. 1. CREATE BETTER DATA. To start, we need to make your data more Clearly define your complete and accurate by getting a better picture of the millions of numbers game companies in your market that are “data opaque,” meaning that they are either poorly profiled or not profiled within most paid databases. Use tools that let you efficiently and intelligently Steps to create the data you need: “play numbers games.” a. Determine which companies in your market are data opaque. Data opaque means that there is no easy way to get a clear picture of these target companies. It won’t be easy for you to do, but it won’t be easy for your competitors either. b. Piece together what you can. There are thousands of fragmented data sources that contain bits of information about the companies you care about. Your job is to assemble this puzzle. New innovations in data science help us to do this with greater speed, precision, and completeness. c. Create new data using data attractors. Data attractors are essentially a form of barter. The attractor gives away something valuable in exchange for data. One of the biggest examples of data attractors is found in the online search business. Google provides accurate search results (something valuable) in exchange for information about what you are seeking (your search data). Google then monetizes your search data by matching you to relevant ads. B2B marketers have also used attractors by offering free white-papers (something valuable) in exchange for an email address (data). www.leadcrunch.ai 7 www.leadcrunch.com

7 Data-Driven Reasons Why B2B Demand Gen Fails 2. Clearly define your numbers game. You need to set up your B2B matrix and fill in the values. You then need to define the types of questions you want to ask and how the matrix will answer each one. Filling in matrix values is no small task. It requires us to estimate the probability of a transaction between any pair of companies. That requires deep, detailed, and accurate data about both the buyer and seller. It also requires knowledge about different types of transactions, how they happen, what prevents them from happening, when they happen, and when they don’t happen. Machine learning is the right tool for this big, complex job. 3. Use tools that let you efficiently and intelligently “play numbers games.” Every seller is different and every marketing campaign is unique. You need a model that creates the right custom B2B matrix required to answer your specific questions. Then, as you run campaigns, you need to feed the results back into your automated learning engine. PUTTING IT ALL TOGETHER But you need to use data science to avoid the minefield that accompanies the task of generating high quality leads. Using data science, you can pinpoint high quality sales leads—the kind that you can close deals quickly and efficiently. And doing that can boost your company’s revenues dramatically. About the Author: Steve Biafore is the Chief Science Officer at LeadCrunch. Steve has 20+ years experience in predictive analytics and data science and has authored numerous analytics patents. Steve has built multiple analytics businesses in the financial, retail and healthcare industries. He was part of the $820 million acquisition of HNC by FICO where he was a co-inventor the Falcon™ fraud detection system, one of the first successful real-time neural network products. It is still the dominant solution in the financial industry for fraud detection. ♛ Contact us [email protected] * 888-708-6649 * www.leadcrunch.ai 101 W. Broadway #200, San Diego, CA 92101 www.leadcrunch.ai 8 www.leadcrunch.com