A complete predictive analytics solution combines a sound understanding of your target market and multiple sources of intent data and real-time engagement data to accurately predict and target new accounts. Target market data includes current customer intelligence and lookalike modeling, plus firmographic data derived from organization characteristics and technographic data that looks at organizations’ current solutions to glean information about purchase behavior. Real-time engagement data comes from responses to various sales and marketing tactics, including direct mail, display advertisements, inside and field sales outreach, and email campaigns to help round the solution out.

Intent data can build on target market intelligence with first-party and third-party data by helping uncover the content research and engagement trends for solutions in your stack. This type of data includes first-party data such as website traffic monitoring that companies can already access internally, and can be an invaluable advantage for a predictive solution. True intent data incorporates third-party data such as intelligence from the B2B web, making it even more powerful as a contributor to a predictive strategy.

Combined, this internal and external intent data provides a framework from which sales and marketing teams can begin to characterize the accounts that make up their current and prospective customers. Intent data forms part of a solid groundwork from which a predictive customer acquisition strategy can build if it has broad coverage of the target market. Real-time engagement provides the final piece to a true predictive solution.

"A proper predictive analytics and data-mining project can involve many people and many weeks. There are also many potential errors to avoid. A "big picture" perspective is necessary to keep the project on track. This course provides that perspective through the lens of a veteran practitioner who has completed dozens of real-world projects. Keith McCormick is an independent data miner and author who specializes in predictive models and segmentation analysis, including classification trees, cluster analysis, and association rules. Here he shares his knowledge with you. Walk through each step of a typical project, from defining the problem and gathering the data and resources, to putting the solution into practice. Keith also provides an overview of CRISP-DM (the de facto data-mining methodology) and the nine laws of data mining, which will keep you focused on strategy and business value.

Topics include:

• What makes a successful predictive analytics project?
• Defining the problem
• Selecting the data
• Acquiring resources: team, budget, and SMEs
• Dealing with missing data
• Finding the solution
• Putting the solution to work
• Overview of CRISP-DM"