Blog

Test-driven search relevance tuning – for free!

As you may know OpenSource Connections has a commitment to search relevance that goes beyond our expert consultancy and training – we genuinely want to build a community of practitioners and support this however we can. We’re also huge fans of open source software such as Elasticsearch and Apache Solr.

However, we’ve also built a closed source product for test-driven relevance training, Quepid, which we provide as a paid-for hosted service. It’s a tool that we use internally on a regular basis – it’s great for creating test queries for Elasticsearch and Solr and for identifying and resolving relevance issues. We know that many of you have also found it useful and a fair few of you signed up as paying users. We think Quepid is a great way to implement a test-based relevance tuning strategy – but we’d like to encourage even more people to use it.

So, we’ve decided to donate Quepid to the community in two ways: by relicensing the project as open source software and by making the hosted version at www.quepid.com absolutely free to use. From now on there will be no charge to use www.quepid.com and shortly afterwards the source code for Quepid will be released on our Github page – allowing you to build Quepid into your own relevance testing platform, contribute to the codebase and even run it behind the firewall (a feature many of you have requested).

We’d like to thank all Quepid users for your support of the project over the last few years. It’s been hugely valuable to get your feedback. We expect Quepid to continue to develop and improve, and look forward to your contributions!

If you have any questions about Quepid and how it can help you tune search relevance please do get in touch.