Overview:

 

 

Sage had numerous data bases in various states and with various integrations. Complete and accurate customer information for targeting and suppressions was a challenging and time-consuming process for all North American teams. When we began to look at the time the ad-hoc processes were having on our various teams we knew we needed to act and streamline our process. Our goal was to get to a nightly refresh of all customer data stored within a single Eloqua CDO.

 

 

 

 

 

Solution & Implementation

 

Sage needed to get to a single source of its North American customer data that refreshed on a nightly basis and could be stored in a CDO.  Completing the B2B: Convert with Custom Objects course really helped with set up. We reviewed the current state of where and how data was coming into the system and how much time it was taking to complete all the ad-hoc processes.

 

The Set Up:

 

Assemble a team

 

Assemble a team to help drive the project forward. It is necessary to have people familiar with the platforms you are extracting the data from, people familiar with the data and it’s challenges and someone to handle the Eloqua side of things.

 

What do you need?

 

Discuss with multiple teams on what is needed and in what formats, for example, products may have descriptions, part numbers, material ID’s etc. Making sure you know what format is the easiest to understand will allow for filtering within Eloqua. Consider all aspects such as products and add ons, service plans and dates.

 

Identify your data sources

 

Do you have one or many systems where the data you need is kept?  After identifying what you need, determine where it is located and what is available to you for extracting the data from those systems.

 

Understand the challenges

 

You are only as good as your data. CRM systems and other databases where the data you need is stored are not always the cleanest. Taking the time to identify issues that could impact the success of your project is critical.

 

Here is an example:

 

Identifying data obscured to protect for privacy

img 1.jpg

Here we see the same company has three different accounts within the database.

 

img 2.jpg

img 3.jpg

 

The problem is compounded at the contact level, here we see one contact is listed twice on each account, one account using the same email address across two accounts. When bringing the data into Eloqua where we want the email address to be the unique identifier this will be a challenge creating multiple CDO records.

 

 

Determine your process

 

Once you have your data, does it need to be cleansed? Do aspects of it need to be concatenated or transformed in another manor?  How will you get it to Eloqua, API or data import via SFTP?

 

Eloqua Build, Testing and QA

 

Review the file for import and have a good understanding of what the data you are importing looks like.  Do you need text fields, large text fields or date fields? Prepare your CDO with the fields you have identified. Review the file and conduct an import.  Set up filters for the various scenarios you will be running and compare these counts to what you would expect to see if you were doing a manual list.  Are the counts close?  What are the differences?  Keep in mind any cleansing of the data you may have done and how this will affect counts.

 

Summary

 

Sage is bringing it’s North American data into a CDO with a CSV file via SFTP.  We started with complete data from the systems and then moved to a delta file of new and modified customers on a nightly basis. We realized an ROI on the project within 10 weeks of implementation.  We are saving over 2,000 man hours per year with this solution! Campaign managers are no longer going to data teams to request lists of current customers to provide to the marketing automation team to upload for campaigns.  They simply provide their criteria to the automation team who utilize pre-built filters to complete segments for campaigns in seconds by product line, plan status, primary contacts, cross sell opportunities, it is all right there at our fingertips!

 

Hours

Team

Previous State

Current State

Data Teams

36 hours per week

2 hours per week

Marketing Automation Team

10 hours per week

1 hour per week

Total

46 hours per week

3 hours per week

 

 

While our goal was to reduce the time and effort these ad hoc processes were having on the business, we realized additional benefits right away. Rather than constantly pulling customer lists from various systems, the data team has had their time freed up to focus on additional projects to benefit the business, such as propensity to buy modeling. Our targeting accuracy within campaigns has improved as well. We are now easily able to identify existing customers and can adjust the message for them vs. a prospect when pulling them into campaigns or choose to exclude them altogether if offers etc. do not apply, which is saving calls into our call centers. Accurate customer information has also improved our lead routing to sales teams. New customer acquisition teams are not making calls on existing customers and account managers are flagged when there is an activity with their accounts, as we are more easily recognizing other interests and increasing cross-sell opportunities.  Having access to the data is shifting how we approach campaigns and we are moving to automated lifecycle nurtures for our customers and have the confidence that we can build such programs and continue to get more complex as the data points to drive the programs are there.