CRM’s identity crisis: Duplicate contacts

At the core of customer relationship management is “who am I talking with?” In a simple SFA or CRM system, it’s obvious: you called them, or they called you. But in enterprise CRM, it’s tricky to identify exactly whom the interaction is with, and every new data source seems to make it harder. The problem occurs at two levels: contact information blur from multiple databases, and avatar confusion from multiple entry points into your company’s web and social networking sites. This week, we’ll cover the top layer of the problem.

Multiple contact lists

Nearly every employee in your company has an address list in their e-mail client. On analyzing a large number of those address books, you’ll find a high degree of overlap. Of course, the data entries in each address book will have minor variations, setting the stage for a massive duplication problem if you ever tried to consolidate all that contact information. Fortunately, in most companies there is little reason to try this for most employees’ contacts. The relationships just aren’t relevant enough to the overall business.

But in professional services firms and investment banking/private equity/venture capital, the contacts of nearly every employee can be a valuable asset. It is very tempting to move towards a centralized list of contacts, making all of them visible in the CRM system. But the chaos of data in the individual address books makes that very hard. Here are approaches we’ve seen work, with the pros and cons:

• Using a shared address book for corporate contacts, supplementing the private address books for each user. This approach is simple, can be set up to automatically synchronize the shared address book with your CRM, and doesn’t involve a lot of cost. But it does require a change of user behavior: users have to remember to keep the contacts in the corporate address book updated, and to put new business contacts into the shared address book rather than their own private one. Further, the effort of deciding which people should be in the shared list, and deduping that shared list once created, is a visible and painful startup cost that stops many organizations from using this strategy. Finally, this won’t really work if you’re a Mac shop.

• Using the CRM system in place of the e-mail shared address book. This is simple to understand and execute, but it requires an even bigger change to user behavior. People like to live in their e-mail client, and this approach means telling users to log in to their CRM system to get the contact info they need all day long. This strategy can work in sales, marketing, and customer service, but for most professional services organizations or senior management, this e-mail-plus-CRM approach is the kiss of death.

• Using plug-ins to synchronize your e-mail client address books directly with your CRM system. There are a variety of plug-ins from CRM vendors, dedicated contact management systems, and third party products that try to seamlessly represent CRM contacts as a natural extension of the user’s individual address book. While each of these products has potential plusses, the user experience with them ranges from the acceptable to the horrendous. I won’t mention any names in print, but feel free to contact me if you need the real story. The bottom line over the long term is that none of these products is really satisfactory without implementing careful policy, extensive testing, or expensive integration. Sometimes, all three. And if you’re a Mac shop, there’s almost nothing on the market to support this strategy.

• Using Microsoft Dynamics. You have to admit it, these guys own more than 80% of the e-mail clients, 65% of the e-mail servers, and control of the desktop. Not surprisingly, they’ve done a bang-up job of integrating with Outlook and avoiding the bug-bombs that seem to come with each of their patches. That said, users generally don’t rate Dynamics very high, the system can be overly complex, and they are in no way a leader in SaaS CRM. So Dynamics isn’t on the short list for a lot of companies — particularly if they’re professional services firms where SaaS matters and Macs abound.

• Using people to offset the weaknesses in technology. Believe it or not, this one can work the best because it makes no tough demands for behavioral change. In this approach, an admin or intern is designated as the keeper of the corporate contact file, and they make all updates — typically to the CRM system. Busy executives like this because it means they don’t sit around typing updates. Periodically (probably daily), the admin/intern runs a push program to update the corporate addresses that everyone uses (typically a read-only shared address book). Because all the data is updated in one place, there’s much less risk of duplicates or data corruption. Ironically, this can be one of the most flexible and least cost approaches, particularly if you have the admin/intern in a low-cost labor market.

Even with the best of these solutions, there are still wrinkles. Thanks to Google Mail, BlackBerries, and differing user work-styles, there will be multiple address books that are at best partially synchronized. The bottom line for this layer of the CRM identity crisis: policy and business process need to be refined if you are ever to get this problem under control. So don’t invest deeply in a technical solution until you’ve got the people and policy issues nailed.