Finding the next best offer for customers is only the tip of the iceberg when it comes to maximizing ROI from campaigns. Companies need to look into using the right channel, the right tone of communications, and even the right time to contact customers to achieve this ultimate objective.
You can download PDF version of this whitepaper here.
For many years, a majority of companies approached campaign development subjectively, letting creativity rather than facts be the main driver in designing campaigns. As more and more companies have been deploying campaign management solutions and have started tracking how their customers are responding to each and every offer, science is moving into the driver’s seat.
Today, customer analytics models can not only provide identification of the best proposition for each prospect, but also define how these prospects would react to if the proposition is communicated in different ways. Companies now customize their communication for each customer through analyzing and understanding the different preferences of their prospects. For example, leading automotive manufacturers such as Lexus and BMW customize both the messages and the images they use in promoting certain cars or campaigns to different prospects, based on their demographics and past behavior – and it completely makes sense, as the tone of voice or the message a student would react to is completed different than that of a busy executive, even when the offer is the same for both.
In order to maximize the return on campaign investments using customer analytics, companies need to analyze and optimize six main components:
Making the Right Offer
- Targeting the right customer – Using predictive models for estimating individual customers’ propensity to uptake any offer, companies can focus their efforts on the right list of prospects, saving communications spending and avoiding overstimulation of the prospects they have.
- …with the right proposition – Testing alternative proposition campaign concepts and offerings (e.g. which product to cross-sell) via market research or pilot campaigns, companies can identify the list of next best offers for different customer segments, maximizing the value generated from each customer.
- …at the right price – Using price/discount elasticity models, companies can also identify the ideal level of price/discount applied for each campaign (e.g. percent of discount to be offered), where the total value of demand reaches its maximum level.
Using the Right Communication
- Using the right channel – Campaign channel optimization balances the likelihood of prospects’ response to campaigns with the cost of using each channel. Using response rate models, companies can optimize the mix of channels used for each offer.
- …at the right time – Different customer segments present different levels of interest and responsiveness during different times (e.g. during work-hours vs. weekends), and after certain events (e.g. after they’ve inquired about a product). Using best-time-to-contact models, companies can contact every prospect when he/she would be most open to an offer.
- …with the right message – Last, but not the least, different customers have different interests, hence respond to different words, images and tone-of-voice. Companies can test messages to find out the ideal way of communicating an offer to each prospect segment.
Parallel to these six components, companies need to take into account the communication constraints and customer privacy, incorporating do-not-call lists and using the right frequency of communications with each customer for best results.
Whether a campaign succeeds or not is heavily determined through the way it is delivered. A company might have the best value proposition, but if it is not delivered at the right time, with the right message, to the right person, it would never catch the attention of its targets, hence would be doomed to failure.
A series of e-mail marketing activities in Dell demonstrates the impact of optimizing campaign communications, with a staggering increase of 710% in their campaign response rates through the simple testing of alternatives to find the ideal design for their e-mail offers (e.g. subject line, image used in offer, number of configurations listed). A leading telecommunications company in the Middle East – after a series of campaign tests – realized that percentage discounts were more attractive for its customers over direct price cuts, even in cases where they had less benefit for the customer, and resulted in more profitability for the company (e.g. 15% discount offer on $10 phone bills receiving better response than $2 discount). Other leading companies report 100% increase in response rates simply by reaching out to the customers at the right time (triggered by events or time of day). Such improvements can turn the least successful campaigns into success stories, and considering the very limited level of investment required for them makes analytics an inseparable part of campaign development.
Using the wrong offer, channel, timing or message not only means a waste of resources in terms of communications costs, but also means decreased responsiveness for future campaigns, as most customers today are over-stimulated by overflowing offers they receive from companies. There is no better way of alienating customers than bombarding them with dozens of irrelevant campaign offerings.
There exist four main steps towards optimizing both the offer and the communications for campaigns:
1. Building the Data Structure: Developing campaign optimization models requires detailed campaign results be on hand so that the data can be analyzed. Every offer, whether successful or not, needs to be recorded, together with the information on the timing, channel used and content of offer presented to prospects (along with the prospects demographics or segment). The offer and communication approach need to be recorded in all their details and categorized to ensure analysis can be conducted (e.g. the keywords used in the message, level and type of discount offered, even the colors used in printed offers or the gender of agent talking to the prospect).
2. Running Campaign Tests: In order to identify what combination of offerings and communications parameters would generate the highest return, companies need to test alternatives and identify which get the highest return with the lowest cost for each customer or segment. For companies which have already performed hundreds of campaigns, this is a matter of analyzing historical data for the response rates. For the others, either market research or pilot campaign runs with test cases will yield findings. Since it is practically impossible to test every possible scenario, companies should follow concept testing approaches, such as conjoint analysis for this purpose. Below is a sample scenario of test cases, where different combinations of campaign offerings are tested for the same target segment to identify the ideal mix:
3. Developing Optimization Models: Each of the six main components listed require different econometrics, statistics and data mining models to be utilized, ranging from price elasticity curves to decision trees. Yet, the question in each case boils down to responding to a simple question based on the learning from campaign tests: “What is the likelihood of this customer to respond positively to this offer presented in this specific way?” Having an answer to this question for each prospect and each possible combination of the offers and communications alternatives provides the ability to select the option with the highest impact on customer value for the company.
4. Utilizing Optimization Models: Once a company runs enough tests to identify the ideal campaigns for each customer and develops optimization models, the next step is putting them into practice, using them for each and every campaign the company has to offer. Below is a sample of ideal campaign offerings and communications for two different customers, which demonstrates the outcome of such use:
Companies should see their campaign offerings and prospect bases as two separate portfolios and use these models to perform the ideal matching between them. This implicitly requires centralizing all campaign decisions, in order to select the ideal offering for each customer and the ideal customers for each campaign, which means limiting the freedom the product and segment managers have over running their own campaigns independently. As cases such as Royal Bank of Canada – which had developed a central campaign execution body for this purpose and became one of the most well-known CRM success stories – demonstrate, the impact is worth the trouble.
These four main steps simply build the foundation required for analytics-driven campaign development. Putting them into practice and making the most out of them requires analytical thinking and a collaborative approach to marketing across the organization. Buy-in and training across marketing and related functions are keys to success; hence companies pursuing this approach should not forget managing change in culture and way of doing marketing throughout the process.
Use of analytics to optimize campaigns requires continuous testing and development, as expectations of customers, competitor offerings and companies’ own value propositions continuously change over time. Companies should incorporate this approach into their day-to-day marketing activities. As the next step, companies can carry the same principles into non-campaign interactions with the customers, boosting customer satisfaction via using the right channel, time, and message for their communications with the customers across their life-cycle.
To learn more about making most out of your campaigns, please contact email@example.com.