The concept of campaign optimization isn't new. Companies have been attempting some level of optimization since the days of mass marketing. We've seen the evolution of mass marketing to segmentation and then to user-driven statistical modeling.
Using advanced, multialgorithmic, data-driven modeling or data mining, marketers can use everything they know about their customers and prospects to more intelligently tailor their marketing campaigns to the individual. Campaign optimization means using customer data to build better, more targeted and more profitable marketing campaigns.
So what's the problem? The problem is that the campaign management process and the campaign optimization process have historically been separate and mostly manual processes — not integrated and not automated. The typical organizational scenario is this: Marketing has the domain-specific knowledge, and its job is to create campaigns — including creative, copy, offer, channel and target audience. Historically, it used manual tools such as spreadsheets and project management applications to manage campaigns. The analysts and statisticians do all the number crunching using complex, highly technical statistical tools and then make recommendations to the marketing department about who to target or what to sell.
Unfortunately these two groups have different skill-sets, knowledge areas and perspectives (and sometimes even different parts of the organization). The result is that marketers don't always understand how to fully use the information they receive from the analysts, and the analysts don't always understand the nature of the information that the marketers really need.
Why is this untenable now? The speed of business has changed. We are running on Internet time now. In the Silicon Valley, people are fond of saying, “If you stop for lunch, you are lunch!”
There are new customer access points (i.e. the Internet) that means there are more ways than ever to reach customers. There are data mining tools that can be used by both the business user and the statistician, and that can quickly and easily build lots of models.
The result of this is that there are more models, more customer contact points — more campaigns. Creating, coordinating, optimizing, deploying and managing all those campaigns and all those touch points in real-time is extremely difficult to do manually without an automated tool designed specifically for that purpose.
What's required in the next generation of campaign management tools? All the functions involved in campaign creation, optimization, deployment and management — including analysts, statisticians, marketers, managers and IT — must be able to use the tool. It must fully use and integrate the results of data mining technology to optimize campaigns.
It must work directly with the company's data in whatever form that data is in and must be able to tell managers what the financial outcome of a previous campaign was, or of a planned campaign is projected to be. Finally, it must offer a variety of ways to plan, view and store multiple overlapping campaigns.
What could you do with such a next generation tool? Rather than contacting every prospect from the best performing list (based on a test mailing), with integrated campaign optimization you can select the best prospects from all available lists combined. Rather than sending the best performing offer to every customer, send each customer the offer they are most likely to respond to.
Some specific applications of such capabilities are:
* More profitably attack customer attrition by sending out different retention offers to different customers based on modeling results. By not offering the same incentive to everybody, you save money by reducing the amount you give to those who would respond to lower incentives.
* By integrating modeling directly into your campaign you can refine your acquisition and lead generation strategies by developing multiple models each time you begin a campaign – no more using the same model over and over again. The models are built based on the freshest, most accurate data and then used to optimize across all prospect lists.
* Immediately reply to customer inquiries by setting up campaigns that run automatically based on triggers such as Web site hits or banner inquiries.
* Build a multiphase campaign so you contact customers at different times, through multiple channels with a coordinated consistent message.
Marketing is the last functional area of business to be automated and computerized. The next generation of campaign management tools is just starting to come out. By giving marketers an automated tool to track increasingly complex marketing campaign strategies and to optimize campaigns down to the individual level, campaign management tools are taking marketers one step closer to the goal of right target, with the right message, at the right time, through the right channel, with the right offer and at the right price.
Kerry Reilly is vice president of database marketing consulting at Unica Technologies, Lincoln, MA.
