Improving Your Search Engine

Jeff Lawrence writing on the Celerity blog, has an article: How to Improve Your Search Engine: Unlocking Basic Features.  Jeff discusses three basic routes:

  • Review the server and search logs
  • Review your search interface
  • Review the features and functionality offered by your existing search tool and check if you’re using them all to the fullest extent

The item is not specifically about DAM, but since search is integral to it, I suspect that was one of the more widely understood underlying use cases Jeff had in mind when he wrote the article:

Understanding the basics of an effective search strategy is key to unlocking your search engine’s potential, improving customer satisfaction, and increasing ROI. It’s so important to build an utterly simple and helpful search user interface. If you make your customers work to find what they’re looking for, the user experience will suffer and they’ll most often look elsewhere.  If you simply turned on your new search tool out of the box, you’re probably only getting modest results. Most modern search tools include a number of overlooked options that are not pre-configured or require additional set up to work effectively.” [Read More]

Jeff offers a breakdown of some common search features offered by search tools and a description of how they usually work, it includes some well known ones you will encounter in DAM like synonyms, faceted search and relevancy as well as some more esoteric or specialised options.

On the first point, one refinement I would make to Jeff’s advice is that (if you have the choice) choose DAM solutions which provide an audit that can be accessed via the standard administrator user interface, not one you need to go off and ask an engineer to get for you.  Ideally the audit should be oriented around the system activities as you understand them, not a bunch of obscure URLs and complex technical messages, but more along the lines of “user x logged in at 12:45pm on Tuesday 24th June 2014” or “user y searched for ‘new logo’ and found 236 assets”.  You need the detail, but still to be able to understand what it is telling you without a developer or database engineer to translate and explain the meaning of each entry.  The audit should be one you can review from either a user or asset perspective – as these are usually the two most important entities in a typical DAM system.

That aside, the article is good at providing an overview of the various search strategies and techniques on offer.  An issue with current DAM solutions is that typically you can’t replace the built-in search feature with an alternative, in part due to the fact that the indexing (the bit you don’t see which the search relies on) is usually tightly integrated into an associated database asset index.  With that said, the majority of DAM solutions now use a search/index component that someone else developed, usually based on something like Lucene and most of these components now either include those features or can be extended to support them.

I still live in hope that some day more DAM developers will eventually re-factor their applications to allow the kind of best of breed modularity and flexibility which would make this possible (and not just for search)  but I have to acknowledge that this isn’t something that most end-users currently appreciate the need for (nor are willing to subsidise the cost thereof).  Those vendors with the requisite foresight (and development budgets) may be able to anticipate demand for that changing as interoperability and scalability both become increasingly important components for a successful long-term DAM strategy.

Share this Article:

Leave a Reply

Your email address will not be published. Required fields are marked *