Improve communication effectiveness ten-fold: add relevance to the marketing mix

Marketers have frequently professed an objective of delivering offers to the right customer at the right time. Meeting this objective has not really been possible until the advent of CRM, and in particular, the implementation of event-driven campaign management systems.

So far in this series of editorials, I have avoided this subject, partly because it is the functional area where I can lay some claim to have made a contribution, and so there's a danger I'll go on about it too much. I played a significant role in the specification, design, development, and implementation of Mind, a multi-channel event driven campaign management system in the late 80's. Mind was implemented at 20 or so mainly Retail Financial Services clients around the world, and some of today's leading campaign management companies have been kind enough to recognise that Mind played a role in defining requirements in this area.

Enough of blowing my own trumpet. The purpose of raising the subject is to try and highlight the many important uses of multi-channel event-driven communications in a CRM world.

Let's start with the big idea. Segmentation and propensity modelling (techniques outlined in the editorial last week on Customer Knowledge), will define a target market for a particular product or service, or more generally, a profit opportunity. However, the members of a target market will not usually make a purchase decision at the same time. What drives that decision is often events particular to those individuals.

So house purchase, for example, may be triggered by change of job, birth of a child to the family, financial inheritance or other change in financial status. So if we can deliver communications to the target market when those events occur to individual members, we are likely to raise the relevance, and hence the response to the communication. This turns one-off large scale campaigns to the whole target market, into a continuous process of trickle-feeding communications to members of the target market as and when the communication is relevant to them. To implement this we need some technology which will take the definition of each of these campaigns, apply it to our customer base on a pre-defined basis, and select those customers ready to receive that communication just now.

So the big idea is to use events happening to customers to trigger communications to them, in order to make the communications relevant to the recipients.

There are a number of caveats to introduce immediately. Firstly, depending on your culture, the use of customers' personal events by large organisations may be considered intrusive and constitute mis-use of data. Secondly, it may be extremely difficult to identify events accurately. For example, in some Christian societies, a woman changing her surname may be an event identifying that she has got married, or is it that she's got divorced, or perhaps even more likely, is the spelling of her name by the company being corrected??

For these and other reasons, in my experience the events actually used tend to be changes in transactional behaviour by a client. For example, in financial services, changes in the credit and debit turnover of a transactional account may identify saving or loans opportunities; a failed cash withdrawal at an ATM may be an opportunity to increase a credit limit; impending redemption of a life assurance policy may be a savings account opportunity, etc, etc, etc.

Outside financial services, changes in the usage patterns of a phone, or airline travel patterns may open opportunities for companies to respond relevantly to their customers. In retail, a first purchase of Pampers or equivalent, may indicate a need for other baby-related products. Note that in such circumstances companies are responding to the customer's use of their services so it is hard to argue that this is intrusive, indeed if the relevance is high, it is usually perceived as good customer service.

So hopefully we now understand the event-driven aspect of a campaign management system, but what about the multi-channel aspect? Most companies have more than one channel for communicating with customers, and more to the point, customers like to use all available communication channels to a company as they see fit.

To manage customer relationships effectively it is key that all channels should deliver the same message to the customer. In proactive communications, we may also want to chose the most cost-effective channel for specific communications.

To meet these requirements, each channel needs access to a common store of customer and contact data, and our campaign management system needs to be able to coordinate communications across these channels. In practice, this second requirement can be met by an event-driven campaign management system, provided it has access to contact and response data, as we can use a response (or lack of a response) as an event to trigger further communications, in the same or a different, channel, assuming there are interfaces between the campaign management system and the various channels.

Perhaps some examples can make this clear.
If we send a letter to a customer, and the customer doesn't respond, we can trigger a follow-up call X days after sending the letter, provided there is still no response. Alternatively, a positive response to a telephone call making an offer might be used to trigger a salesman's visit to the customer to try and close the deal. So a multi-channel campaign management system can be used both to deliver relevant communications to customers, and manage the delivery of subsequent follow-up and fulfilment communications.

So we now understand the two core features of a CRM campaign management system: to be able to deliver event-driven communications to a target audience defined by segmentation and scorecard data; and to co-ordinate those communications across all the organisation's channels, including follow-up and fulfilment communications. How can these techniques be used at the business level to deliver value to the organisation? A few or the many applications are outlined below:

1. Adding relevant sales messages to a customer service contact centre:

The application where we have seen most impact, both financial and in terms of customer relationship, is the use of this sort of tool to deliver relevant communications to customers in a customer service call service or contact centre.

Correctly integrated into a call centre environment, such a system can add relevant sales calls to a service environment. A customer can call into a customer service environment, and whilst that service call is being handled by an operator, the system can check whether there are any relevant event-driven communications to deliver to that customer. At the end of the service call ("How can I help"), the operator can be prompted to offer any relevant sales message ("Can we do this for you").

Delivered in a service environment, with the right quality of staff, this approach can deliver significant benefits. The best case study available to me is still First Direct, the first implementation of Mind, which can be found at:

2. Delivering relevant sales and service messages to customers:

The use of event-driven techniques does not have to be limited to a call centre environment. Within a conventional marketing department, an event-driven campaign management system can be used to maintain a library of campaigns which are run on a regular basis (daily, weekly or some other cycle) to deliver relevant sales and marketing communications to customers.

Two interesting things in such an environment are the improvement in effectiveness, and the number of campaigns that can be run. Typically when segmentation and propensity models are combined with event-driven techniques and implemented effectively, we see perhaps a ten-fold improvement in response rates.

Identifying the target market through segmentation and propensity models using transactional data frequently improves response rates 3-fold. For example, in Retail Financial Services, campaign response rates have moved from norms of <1% to 2% up to 3% - 8%. Incorporating event-driven techniques have then often led to response rates of the order of 25% with early campaigns, and as experience grows in the use of the technique further improvement can be expected.

3. Converting product-based campaigns to 1-1 communications:

There has been a lot of discussion of 1-1 marketing techniques over the last few years, but in my own view, there are almost no companies where it is financially feasible to have a customer strategy per customer.

However, if you accept the premise that people or companies can be segmented into groups who behave similarly, but at different times, due to the events in their lives or businesses, then the use of event-driven techniques allow one to deliver relevant messages to customers, and these are likely to approach 1-1 communications. The significance of the relevance of a message is hard to ever-emphasise. If you receive an offer which really meets a need you have at that moment, it is not perceived as an intrusive sales message, but as good customer service. Imagine, you're sitting at home, dreaming of the new yacht you can almost afford, and an offer of a marine mortgage lands on your door-step. That's not junk mail - it's a highly relevant message which allows you to achieve your dream. (I'm still dreaming).

4. Managing the life-cycle of cyclical products such as credit-cards:

With some products and services, there is a well-defined life - cycle with key steps defined which have to be managed. For example, many credit-cards go through five key steps in a yearly cycle: from initial customer acquisition; through activation of the credit-card; stimulation of usage, and in particular activation of revolving credit usage; tri-aging of credit risk situations; and finally renewal of the credit-card after a year (similar to customer acquisition).

With the high value of transactional activity, customer segmentation and propensity models combined with event-driven techniques can be used to manage each of these events and potentially combined into a customr loyalty program. You can find a case study on such a program at:


and I also highly recommend a report from The Council for Financial Competition, Washington, called "Towards the information bank" which covers this topic in depth.

For an example of how to manage churn in mobile communications:

5. Managing a salesman's diary appointments:

In some industries, (e.g. frequently Insurance) a large-scale sales force is key to the sales process. In such a situation there has long been a view, that the salesman is too valuable a resource to spend his time generating leads, and converting them to prospects. He should be focused on closing the deal with prospects who are 'ready to buy'.

This idea has been difficult to implement, but multi-channel event-driven campaign management systems provide the technology to support such a strategy. Basically, the idea is to use the most cost-effective channel to undertake each of the steps of getting new customers: normally direct marketing channels (<$1 per communication) to generate suspects; suspects are qualified via call centre communications (perhaps $5 per communication); which in turn generate appointments (perhaps $75 per visit) for salesmen to visit prospects to close the deal.

Such systems are frequently implemented as 'tele-appointing' systems, and as well as cutting costs by using the most cost-effective channel for each communication, they also improve the effectiveness of the salesman. In implementations I have been involved in, the client has seen a 35% increase in salesman's effectiveness as measured by sales made per saleman per week. You can find case study examples of this approach in financial services in a presentation at:


6. Follow-up and fulfillment of conventional direct marketing campaigns, including response analysis:

In the 80's Peter Rosenwald, a guru of Database Marketing, (where are you Peter?), had a presentation called "The Golden Shoe-box". In brief, the said shoe-box contained all the leads that marketing generated, which in many organisations do not get processed at all, with subsequent loss of business.

The problem for such organisations is often that inter- departmental communications are not necessarily good, and the process to convert all the leads into business is often inefficient.

Our multi-channel event-driven campaign management should be able to solve this problem, provided it allows us to capture the details of campaigns run manually. Normally, conventional Direct Marketing campaigns are not run through such a system, because of the difficulties of interfacing with the marketing agency, bureaux services, bought-in list providers, etc, so we need to be able to record such external campaigns within the system, and record responses against those campaigns, with our campaign management system managing the follow-up and fulfillment of the leads generating, emptying Peter's golden shoe-box.

Because we're managing the responses, this also means we should be able to use our standard response analyses tools to measure the effectiveness of those campaigns, as well as manage the fulfillment process.

7. Implementation of champion / challenger approaches:

In last week's editorial, we touched on the importance of response analysis as a means for evaluating the effectiveness of marketing campaigns, and marketing, provided we have an adequate response capture mechanism.

The event-driven campaign management system, because it provides us with a tool which allows us to define and run many complex campaigns, provides us with a means of implementing a Champion / Challenger approach.

There are always arguments and discussions on the 'best approach' to persuading customers to take up a product or service option. Different creative, offers, contact strategies etc will be advocated by different parties. The champion / challenger approach is a methodology for testing which works bests. In any campaign, 80 - 90% of the campaign volume is reserved for 'current best practice' with the remainder being devoted to 'tests' of different approaches which 'challenge' the current best practice.

Contact history records of which customers received which offer are maintained, and responses from customers are captured. Provided the difference in responses between tests and current best practice (champion) are statistically significant, we can decide whether or not to stay with the current champion, or promote a challenger to become champion. Adoption of this methodology lets us set up a process of continuously improving our marketing effectiveness, and adopt to changes in the marketplace.

We have no specific case studies of this technique, though most case studies will make some reference to this, and the Direct Marketing industry has long adopted this approach, frequently running 10s or 100s of tests in large-scale campaigns.

Our multi-channel event-driven campaign management system helps implement such a methodology by allowing the details of all the campaigns and tests to be captured, and issuing the resultant communications.

8. Co-ordination of e-channel and conventional channels:

We've left till last one of the trickiest areas, which is the integration of the e-channel with conventional channels. In one sense the e-channel is a channel like any other, so all we need to do is integrate it with existing channels, and we should be able to include it in our marketing activity.

At one level this is true, but at another, the e-channel imposes a number of additional requirements. The fundamental issue is that in the e-channel, the customer interacts with the organisation with no direct intervention of the organisation's staff. This requires a level of accuracy in the targeting of communications and the accuracy of the data which is not so necessary in other channels, as there is usually a human filter between the communication and the customer. Also, data gathered in the current interaction with the customer in the e-channel may itself change the communication we want to make, and this may lead to needs for almost real-time selection of communication. These difficulties, and an architectural approach to resolving them are outlined in a presentation entitled:

Real-time CRM: Technological Horizons at:

Not surprisingly given the subject, I have made reference to a number of documents in the CRM-Forum library in this editorial, and for ease of reference I repeat them below. I am sure you'll find something of interest in them, if you've found this editorial of value. The papers referenced are:

Capturing value through CRM - First Direct:

CRM & Credit-cards in Asia Pacific:

Churn in Mobile Communications:

Creating Value through CRM in Retail Financial Services:

Real-time CRM: Technological Horizons:

As usual, we'd appreciate any feedback to this editorial, either as a comment attached to this editorial, or in our Discussion Forum at:
or by email to me at [email protected]

Share this content


Please login or register to join the discussion.

There are currently no replies, be the first to post a reply.