Three reasons why your search may be prehistoric

ArnoldIT wondered today why we were bothering to announce an upgrade to the venerable dtSearch engine, when they “weren’t aware of too many people still using that software”. Perhaps it’s time for a quick reality check here – we regularly see clients with search engines that many would consider prehistoric still in active use. Here’s some reasons why that might be so:

  • Search isn’t seen as essential. If your accounting software goes down, nobody gets paid: but if the search engine has gradually degraded in accuracy, doesn’t always contain the most recent documents and is generally too hard to use then most of your users will try and find a way around it – they’ll Google for content on the corporate website, dig slowly through the filestores or call up a colleague to ask. Of course, all of this will take time and there’s the risk they won’t find anything useful (or worse, find something inaccurate or out-of-date), but time is only money, surely?
  • The magic has gone. The sharp suited salesman who told you all the magical things your search engine could do – it could understand concepts, human language and the meaning of life – is a distant memory. Somehow those magical features were never implemented, perhaps the unexpected extra cost put you off (surely the magic came as standard? No?). You’ve also probably turned off a lot of the clever features of your engine as either no-one could understand how to use them, or they affected performance so much that search results took minutes to appear.
  • Upgrading search is hard and expensive. Small changes to the existing engine can cost huge consultancy fees but if you change supplier, you’ll have a whole new team of salesmen to meet, lots more buzzwords to learn, there’s expensive new license fees to pay, you’ll also have to overhaul your content management system, your metadata, your front ends…better to leave everything alone, surely?

There are search engines out there, chugging away quietly behind a corporate firewall, whose antiquity would astonish. Any chance of a support contract has long gone as the supplier would prefer it if you upgraded to their latest-and-greatest version – that’s if the supplier still exists at all. However there is always a way to upgrade that reduces the risk and cost – an incremental, agile and open-source based approach will prevent future lock-in to a single supplier and give you more control of the code your search engine depends on. Recently we’ve used this approach to help clients successfully upgrade search applications based on dtSearch, FAST ESP and Oracle and in the near future we’ll be doing the same for clients with several other well-known engines – and a few lost in the mists of time!

Share this postShare on FacebookShare on Google+Tweet about this on TwitterShare on LinkedInShare on RedditEmail this to someone

Leave a Reply

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