One of the biggest challenges facing today's enterprise organizations is how to manage and digest large quantities of data. For most companies, web analytics is only a piece of the puzzle. There are places outside of marketing where data is constantly being collected and needs to somehow be turned into something useful: call centers, online ad serving systems and offline marketing efforts are a few examples that come to mind. In most cases, these different data "silos" are not only managed by different groups within the organization, but they each use a different data model making it difficult to share data across different systems or multiple business units. In fact, in his whitepaper entitled "Developing High Quality Data Models", Matthew West suggests that anywhere between 25-70% of the cost of current analytics systems can be attributed to the interfaces required between systems that must share their data with one another. Fortunately, Adobe’s (formerly Omniture’s) SiteCatalyst tool comes with a wide array of options for importing and exporting data.
There are two basic ways in which SiteCatalyst can be useful in helping organizations centralize their data and reporting capabilities. The decision you first have to make is whether you want to use an internal system as your data and reporting hub, or whether you'd prefer to use SiteCatalyst for this purpose.
Using an internal tool as your data hub
If your organization already has a system in place (such as a business intelligence tool tied to your data warehouse) for managing data and reporting, then it makes sense to include your web analytics data in that system. Having your web analytics data alongside other “offline” data enables you to have an understanding of how a customer or prospect’s online activities influence what they do offline (or vice versa). Of course, in order to combine these data sets together, we first need to establish a “unique key”. A unique key is a value, which uniquely identifies a set of data records as belonging to a specific entity – in this case, an “entity” would be a person. For instance, if a person has an account on your website, chances are that you maintain a database somewhere that contains a profile of that person. The database already contains a unique key for each person it has a record for. Therefore, if someone is logged in on your site, you know who they are and can use the unique key from your database as a way to identify a user in SiteCatalyst. Setting a custom variable with this unique identifier will allow you to export your analytics data at a later date, at which point it will already be associated to the correct person for easy inclusion into your data warehouse. For users that haven’t logged in, you will be able to see what a given person has done online (based on their automatically assigned visitor ID from Omniture) but you will not be able to correlate this with any potential offline activities.
Once you have the right data in SiteCatalyst, the question becomes “How do I get the data back out?” SiteCatalyst offers a number of facilities for exporting data. These include Data Warehouse, data feeds and the Web Services API.
Using SiteCatalyst as your data hub
Perhaps you don’t already have a data warehouse in your organization, or you don’t have a good reporting interface available for it. Well – SiteCatalyst to the rescue! It contains a number of features like report scheduling, dashboards, and alerts which make accessing and using reports simple. In addition, a feature called Data Sources allows you to import information from external systems in three different ways: “summary”, “transactional” and “full processing”.
Using the “summary” option, you can import generic metrics that do not relate directly to a user’s online behavior. For example, if you wanted to be able to report on how many calls were received in your call center for a given product or service offered by your organization, the “summary” option would allow you to do this. Data imported via this method is only available in the SiteCatalyst interface, and cannot be exported using any kind of data warehouse or data feed (as such, it’s usually a good idea to keep a backup of the original data just in case).
“Transactional” processing allows you to set a transaction ID for a given online transaction, and later upload offline data with the same transaction ID. One common use case for this functionality is when companies send traffic offsite to complete a purchase or other process. Since it is not always possible to implement your analytics tracking on a 3rd party site, you normally would have to look at two different sets of reports to understand how many people were sent offsite versus how many actually completed their purchase on the external site. If you were to assign a transaction ID for each “lead” sent to the offsite link, however, and the external site was able to provide you with a report containing this transaction ID for any completed purchases, you could import the external site’s data back into SiteCatalyst and have a full picture of the user’s activity from site entry to offsite purchase.
Finally, “full processing” gives you the ability to import nearly any type of data you want, including multiple dimensions and metrics. Unlike the “summary” data source type, “full processing” lives up to its name by making the data available via all of SiteCatalyst’s export methods.
As you can see, SiteCatalyst can be a great tool for helping your organization to become streamlined and data-driven. We have helped a number of clients centralize their organization’s data and reporting capabilities, and we’d be happy to talk to you about how we can assist you in those efforts. If you’d like to know more, feel free to e-mail us at firstname.lastname@example.org