Contact Management Optimization--Direct Marketing, Mail Order, Direct Mail, Television and Infomercial Marketing, E-commerce News

Direct Marketing, Mail Order, and E-commerce News from the National Mail Order Association

====================================================
Newsletter Sponsor

Contact me…Please!
By Steven Permuy

Periodically, I get together with a couple of good friends I knew from high school and go into a recording studio across from the Count Basie Theater in Red Bank, NJ (near the famed Jersey shore) to play songs that were relevant in our teens and then hang back afterwards to watch the engineer work his mixing and mastering magic behind the console. Now, the console features an elaborate equalizer that is used to control, individually, a number of different frequency bands in the recording system, which is used to fine-tune the sound (not altogether different from what you’d find in a home stereo system).

The foregoing reminds me a lot of what Contact Optimization can do for the marketer. That is, to make the best decisions regarding what eligible customers get access to your campaigns through the measurement, arrangement and rearrangement of various promotional, product and individual inputs—getting closer to that fabled market of one.

What’s the mantra?

In direct response, the mantra of ‘testing and learning’ coincides with the precepts of optimization, which seeks to make a system, process or decision as fully functional or effective as possible. It’s a discipline that is exercised broadly in fields afar of direct marketing as Realpolitik foreign policy and treatment planning in medicine, for example, but is no less importantly applied than in the marketing sciences.

Optimization is about testing…testing what? How about testing treatments comprising various product and channel permutations, for example, where each permutation (alternative offer) provides a certain return?

Figure 1

Going For The One

Too often, in the process of ‘hitting numbers,’ are marketers not making the best choice for the individual—i.e., ranking campaign selections in a priority fashion (by deciles)—which culminates in opportunity costs (money left on the table). This is where Contact optimization comes into play because you want to shift from selecting an individual solely on the basis of meeting product campaign quotas, towards selecting the ‘best offer’ based on the ‘best economic contribution’ resident in each individual.

Contact optimization compares a range of offers for each individual supported by a mathematical equation (e.g., the propensity of an individual to take action from a communication, multiplied by the value of that action, less the cost of that communication) that is overlaid with real world business rules or ‘constraints’ (e.g., budget limits and segment exclusions) and “what-if” campaign scenarios that are exercised outside of, but informs, the individual selection; furthermore, it is a statistical methodology for optimizing marketing contribution across silos (multiple channels/campaigns) that applies and integrates those interactions, putting them on a common basis (e.g., ROI analysis across the enterprise).

Figure 2

Grist for the mill (the secret sauce—shhh)

Providing a solution to an optimization problem without constraints (a condition an optimization problem must satisfy) would be rather simple and unbalanced. The measures to be optimized (such as contribution or sales) would be calculated without regard to any other parameter such as budgets, eligibility of the individual, contact frequency, to name a few.

Constraints create equilibrium and focus on the campaign planning process by eliminating a portion of the potential communications that will not yield desired results. So, once you have propensity models for most campaigns built off a cleansed and consolidated customer database (single-view-of-customer), identify and apply a series of business rules or constraints that are used to define alternate scenarios, which are then used to guide individual selection.

DM and Brand (in the same boat)

In direct marketing, large volume mailers are looking to maximize the return of investments that are stultified by inelastic First Class postage, pressure to reduce wasted use of paper in the midst of going “green,” and alternate channel incursion. Utilizing contact optimization will not only help mailers qualify individuals with a greater potential to respond, it will also maximize postal savings and, via an integrated marketing platform, provide an economically favorable way to increase speed to market of mailing campaigns.

You start by taking a propensity and deliverability-scored customer or prospect database in order to identify individuals at valid and mailable addresses that are most likely to respond; you then apply decisioning that produces segment and channel exclusion rules (e.g., do not mail; e-mail only), budget limits, and campaign scenario testing; and finally you specify how many ZIP+4 records should be provided to achieve automation discounts. Part of the process entails utilizing a platform that supports the propensity scoring and segmentation, which then allows you to go back and tag those individuals in the database with the particular treatment code the contact optimization "recommended."

Figure 3

In brand advertising and multi-channel marketing, media mix models, which involve a very specific and high-value skill set to develop, are used to allocate monies across different campaigns and channels. However, contact optimization can make the media mix model more valuable by providing implications that otherwise would not have surfaced with media mix alone.

This is because media mix provides optimization on a limited set of dimensions (i.e., media, promotion, trade) and guides the allocation of an important but limited number of marketing variables. They can generate valid predictions but are not transparent in identifying the specific nature of the interrelations between the variables on which the predictions are based. Therefore, media mix models provide little or no guidance about how changing the “structure” of marketing activities would impact results. And contact optimization goes outside of those boundaries by addressing marketing complexities that media mix analysis can only cover slightly. It can go beyond the process of what has worked in the past and allows you to try something very different.

Figure 4

Contact optimization creates scenarios that can incorporate media-mix dimensions (media/trade/promotions) for input. If media were a dimension of the scenario then you would need to know the input of media against sales. Contact optimization would gather these and a myriad of other inputs to parameterize a model used to create scenarios.

Contact optimization can demonstrate a way to optimize CRM programs by deciding where to put most of the resources.

Analytic Alchemy

Marketing is increasingly being driven by data analysis and measured results. Both direct marketing programs and general advertising are held to quantitative standards of accountability and use measurement as a method of continuous improvement.

Contact optimization can contribute intelligent decisioning that utilizes analytics to identify highly qualified individuals based on robust, prospect or customer-level data, incorporating custom models and business rules to reduce product offer contention, develop effective contact management strategies, deploy timely offer-testing and measure the effectiveness of each communication.

Broadly-speaking, optimization is about finding the best alignment of data points (i.e., product, individual and offer variables). It is a deterministic approach toward achieving the ‘best outcome’ (such as lowering the cost of acquisition while increasing the conversion rate of qualified leads) given some list of requirements or constraints that a solution to an optimization problem—e.g., to obtain a better return from a $50,000 direct marketing spend across three channels competing for the same budget—must satisfy.
###

Author: Steven Permuy. SPermuy@alterian.com