First thing first! Let’s get rid of the bullshit bingo lingo: “Relevancy tuning” in search is a fancy description for something that’s not very magical, even if it sounds like just that. It’s about getting the right results on top of your search result. End of story. If somebody asks you a question, you should [...]
Our mental model for the seasonal recipe app is helping people use the best ingredients for any particular time of year is the goal for our little demo search app. Since a lot of people in Norway actually go into the nature and forage, fetch, pick, shoot and fish their own food, we wanted to [...]
So, what’s this you ask? It’s a series of mini-hackathons some of us at Comperio are doing to achieve a small list of goals: Learn more about search, both tech and UX. Show that a nice search user experience doesn’t need a search input box. Show a search that is light weight. Generally, build something [...]
In Comperio, we create and develop a lot of interesting search solutions for our customers. The UX designers do user interviews, create personas, user stories, concepts analysis and interaction design. The developers follow up with content analysis, installation of software, configuration, development, and a lot of relevancy tuning: “How can we ensure that the right [...]
What do you do when you have a lot of content, but not many users interacting with it? Ideas for Comoyo on how to go from boring lists to a suggestion engine, letting the users do the heavy lifting. User driven content and search Check out the presentation on LinkedIn: “Ideas on how to find [...]
Today, most search solutions will give you the results as a list from 1 to 10. Problem is, they’re not very appealing, and don’t do the task at hand very well. At the top of the list, it’s okay. Number 1 gets most clicks, number 2 a little less, number three even less, but then [...]
Faceted search and facet amounts are one of the functions that makes searching in huge result sets bearable. And having the facet amounts explained with numbers to tell how many results each facet contain is a must. Problem is, it’s not very readable or scannable. This is an easy problem to fix.