Experience Sitecore ! | March 2013

Experience Sitecore !

More than 200 articles about the best DXP by Martin Miles

Case study: Collection database for Metropolitan Museum of Art (NYC)

This is my first significant Sitecore project ever, which has been completed two years ago.

The Metropolitan Museum of Art is the heavy-loaded web implementation for the iconic museum based in the largest US city – New York. It is been visited by 6.7 million people annually but the online attendance exceeds physical visitors 5 times, with 33 million interactions globally.

By 2010 the amount of the digital assets exceeded 900K and approached the value of 1M with a third of them being exposed publicly. The biggest problem however was that all the museum collections were kept in numerous incompatible source formats and origins. With dozens of collections and each of them managed by totally independent teams in a non-standardized approach, it was clearly a mess turning into a loss of maintainability.

Met chose Sitecore as the best digital experience platform suitable to centralize the authoring process for all editorial teams as well as improve online publishing and distribution in face of even growing online traffic.

I joined the project at a digital agency called Cogapp that provides services to the largest museums, galleries, and libraries as a contractor specializing in Sitecore and my particular scope was to create a centralized editable Sitecore-powered database for all the existing collections. I ended up creating the first single collection database the museum had ever for their artifacts with almost a million editable records!

Beyond the benefits mentioned above, there were lots of crucial improvements derived from my work:

  • Powerful search. Since every single collection item got standardized, it became possible to use semantic search against the whole collection. That alone came as an impressive improvement in visitors’ user experience, reducing the time for thousands of art professionals of reaching desired content.

  • Versioning and history audit. Since the database went live, it became possible to see all the changes done to every particular collection item back to the initial creation, see who made these changes, when and why, and with the ability to roll back to any previous version. Not just that, an approval workflow process was introduced to prevent faulty changes from going live before getting approved by a senior editor. Having a single “Source of Truth” for the data has fixed the “versioning hell” problem once and forever.

  • External data feeds. Powered by Sitecore, every single collection became exportable by external data feeds for all the Museum associates. They automatically get all the changes provided into a central database upon approval.


Instead of a one-time converting and migrating of the data, I created the Exporter Portal to ensure this process becomes reproducible and could be triggered by an individual editing team so that they do not need to roundtrips to the developers. Since all the previous content teams were mis-synchronized, it was impossible to force them all to start committing to the centralized database from day 1, not to say they had to undertake Sitecore training that was scattered over time. Exporter Portal knew all the original source formats each team used and was allowed to pull the delta into the centralized database. With the portal in place teams took could temporary work as before taking their time to familiarize themselves with a new system before they become ready to start using it exclusively – that prevented lots of spontaneous mistakes caused by lack of Sitecore training.