Posted on July 27, 2013, midnight
Note that this post discusses the technical points of a CRM and marketing automation sync, and does not have anything to do with marketing. Instead this discusses various ways of syncing the two systems to ensure data quality.
In the past few weeks, I’ve gone to a bunch of events and mingled with some of the smartest people in CRM and marketing automation. On Tuesday, I attended the Marketo Jumpstart Tour in Boston, a Marketo Usergroup meeting in New York City, and a Salesforce Usergroup meeting in NYC. It was a pretty busy week. When I’ve spoken to people who manage both CRM and their marketing automation, I’ve typically asked them about how they’ve gone about their sync: do they leverage the recommended sync - where you have a marketing automation database filled with marketing qualified leads (MQL) that get pushed into a CRM database that contains a subset of records. These records are only pushed from marketing automation into CRM when they’ve reached a specified Lead score – or when sales goes and cherry picks them because they’re having a slow day. The alternative is a 1:1 sync – each MQL record has a corresponding record in CRM. The records should not be assigned to a sales rep; the records would instead be assigned to a marketing queue.
The recommended setup would be extremely problematic for RingLead since a good portion of our Leads come from the Salesforce AppExchange and are created via API (pushed straight into Salesforce). If there is an existing record in marketing automation, and a new record is created in Salesforce, then there is a chance that the two leads might not be matched up.
The ideal (unrealistic) scenario is illustrated in Scenario A in the graphic below. It illustrates a bunch of MQL records in the marketing automation system, with a record being pushed into CRM when it hits the Lead score threshold.
The problem is, as illustrated in Scenario B, that when a record is created in CRM, there is a chance it could be unmatched to an existing record in marketing automation.
Here’s an example: a prospect spends time on your website and finally downloads a whitepaper. A few days later, they call your sales line and speak with a rep. The sales rep does his or her job by doing a search for existing records in CRM (or, better yet, they’re using Unique Entry which does the search for them as they’re typing the record into Salesforce) but the record does not exist in CRM at this stage – it only exists in marketing automation.
This problem is overcome by abiding by Scenario C, which abides by a 1:1 relationship of records in marketing automation to records in CRM. The MQLs that are not yet ready for sales can simply be assigned to a marketing queue – not a sales queue or a sales rep. This is ideal for preserving data quality in CRM and marketing automation.