Archive for February, 2009

Your eyes only – data driven personalised web advertising

Thursday, February 26th, 2009

I don’t tend to cover web related matters here too much, even though website management has formed part of my role at times. However, when specific opportunities come up for integrating or enhancing insight with web techniques, I’m always interested. At this week’s Technology for Marketing exhibition I stopped by online ad serving solution provider Mediaplex, for a run-down on their services. In brief, they hold advertisers’ online adverting inventory and serve it to media sites where advertising has been placed. In this way, the are able to ensure the latest version of an ad, appropriate to the audience in which it is being displayed, without the advertiser having the nightmare of managing myriad different advertising locations and creative executions manually.

Such services get more sophisticated though, in that they are able to recognise repeat visitors to a given site (through the use of cookies) and serve a version of the ad appropriate to such an individual; perhaps building on a creative theme or trying an alternative messaging angle. In addition, whilst at a rudimentary level, this technology is beginning to allow one-to-one advertising, or at least ads that are highly segmented. By linking their ad serving platform to an advertiser’s CRM data, it becomes possible to create customer and prospect segments to which specific messaging and creative can be targeted. By recognising visitors using previously placed cookies, highly relevant advertising can be placed, appropriate to the individual’s purchasing cycle or previously stated requirements, perhaps.

Mediaplex don’t have real-time links from their platform to CRM systems today, relying on relatively unsophisticated batch data transfer to share segmentation and ad targeting rules. The possibilities to build on this readily present themselves though, and apart from the usual privacy and “big brother” considerations, another powerful use of marketing data presents itself for improving relevance and effectiveness of marketing communications.

7 reasons for real time data updates

Thursday, February 12th, 2009

Previously, (see The secret to CRM & Marketing data management?) I’ve written about the benefits and hazards of creating independent marketing databases, and in particular the questions that need to be asked before taking such an approach. I’m currently involved in a debate over the long term approach that should be taken to the management of marketing data, and where it should reside, which raises some of these issues.

Take the real life example of a campaign automation system that is synchronised with a sales force automation (SFA) solution via a real time data adapter. Changes made to customer and prospect contact data in either system are exchanged almost immediately, together with leads and status updates. When it works, it’s fabulous, providing a real time view of data in either system, ensuring Sales and Marketing are seeing the same picture, whilst enabling them to use the best-of-breed system most appropriate to their respective requirements.

A new CMO and the prevailing economic conditions though have lead to questioning whether marketing data should continue to be managed in-house, rather than outsourcing to a marketing service provider. In reviewing the options for outsourcing however, one of the first issues (of many) that arises is how, if at all, should sales and marketing data integration be maintained?

Most out-sourced or hosted solutions tend to rely on much less sophisticated and timely batch data transfers, via ftp or similar mechanism, which are a long way from the real time synchronisation currently enjoyed. Is moving to such a mechanism and the attendant loss of immediacy important? “This is a really worrying trend,” says Shane Redding of business to business digital and data marketing consultancy Cyance. “It is disappointing to see companies make a backward step of this kind, which in my opinion is usually the result of not making the next step of really using the real time data in anger which then demonstrates the return on investment.”

Shane and I are very much in agreement, and here’s why.

  1. Sales and Marketing users don’t, and shouldn’t need to, understand the intricacies of data integration. They just want to know that data in one system is available in the other; a Sales rep entering a new contact in the SFA system wants to know their prospect is available for marketing activity. It invokes much greater confidence if this transfer is immediate, without having to know about or understand overnight batch updates. Once control is lost, users feel disconnected and reduce their ownership of the process, leading to a rapid deterioration in data quality.
  2. The sooner changes made to a record in either system are replicated, the less chance there is of subsequent changes to the same record in the other system being made before the data is transferred, leading to potential anomalies or corruption. This is particularly the case where records are merged or changes are made to many fields at the same time.
  3. Marketing-generated leads need to be transferred to Sales promptly. Research shows that timely lead follow-up is one of the biggest determinants of successful lead conversion. If a lead or response relates to an existing contact or customer, Sales should be made aware as soon as possible, allowing a rep to handle their account in the most appropriate way
  4. Best-of-breed marketing practices, such as trigger marketing based on response or other events, require good data integration. Explaining such requirements away saying “we don’t need to do that” won’t cut it. Your competitors are doing it.
  5. Business is moving ever faster. It is expected that data changes are available immediately, especially between Sales and Marketing systems. Reverting to a batch system is a backwards step that fails to lay the foundations for modern and forward-thinking marketing capability.
  6. System development and testing are substantially quicker and easier if changes in one system are reflected in the other almost straight away, rather than having to wait to see if configuration changes are working as intended.
  7. Much of the complexity in data synchronisation lies in the business rules for handling updates, conflicts, mappings and referential integrity. Once these rules are in place, why not transfer data more frequently, reducing the volume and complexity of batch updates when they occur?

Marketing shouldn’t be ashamed to stand up for genuine business requirements, with demonstrable benefits. Don’t let internal politics or external suppliers tell you otherwise!

With thanks to Shane Redding for contributing to this post.

Top 10 tips for sourcing marketing technology

Tuesday, February 3rd, 2009

Just about this time last year, I outlined a fairly personal set of 6 crucial marketing automation system requirements that it was particularly important to us were incorporated into the system we were about to deploy. Business 2 Business Marketing magazine’s online companion site has just published an alternative take on marketing technology requirements which I thought complimented mine. Rather broader in scope, point four “Pilot your technology” particularly resonated with me, given its message of testing and phased roll-out, rather than taking an all-or-nothing approach (see “Testing, testing, testing” for my thoughts in this area.)

After the problems we’ve been experiencing this year, having rushed into production with key elements untested, this sentiment is particularly pertinent. Don’t let yourself think “I’m sure it’ll be alright” – if it can go wrong, it probably will!