Get these factors right and you're well on your way to building a supreme site-search experience.
Understanding context is something we humans take for granted, but teaching an artificial intelligence to 'get it' can be tricky.
It doesn't help that there are a gazillion different words for the same thing, like 'sport shoes', 'trainers', 'pumps' and 'sneakers'. For this reason, it's paramount that your site-search understands synonymity, allowing it to interpret synonyms and serve up the right results every time.
Another contextual challenge is the opposite of handling synonyms – handling polysemic words, or words with multiple meanings. This is important because you don't want users to search for a pair of 'pumps' and be presented with devices for inflating their bike tyres!
To begin, you need to think about data laundry, which is all about keeping your database clean and streamlined. It involves removing duplicate index entries, normalising attributes, and merging your repositories.
We've done our best to explain each of these challenges, describe why they need to be solved, and offer solutions and fixes in our deep-dive. Check it out today and ensure your site-search delights visitors and performs as well as it can.