Search is a powerful and necessary aspect of many websites. It is an entry point for many people, allowing them to quickly access or find content that will be most interesting to them. However, creating search features can be a cumbersome and lengthy process for both developers and designers. Allowing little control over the implementation or making future additions a painful task.
A certain degree of balance is to be achieved between integration, time constraints, and usability. As developers, designers and users ourselves, we never want to find ourselves in a position where we are providing a sub-par experience. Constantly striving to make it the best possible experience not only for a client, but also the very people we’re creating for.
We identified a few key points and features in what we were looking for in the search;
- Flexible enough to have full control over design and UX.
- Allowed for quick, easy development.
- A fast API allowing for live, on the fly searching with suggestions.
- Geolocation search allowing people to find local inspiration.
- Metrics and reporting of searches.
WordPress’ built-in search, though offers some great basics, could not offer us the flexibility that we were looking for and would require a large amount of development on top of what is already there. 3rd party implementations such as Google Custom Search, despite being powered by Google, generally do not offer the level of integration that is required, for a complete experience, for the creative work we as developers and designers do. ElasticSearch came close to what we needed as it is powerful and allows for a complete, flexible search. However, the required setup and maintenance was too much for our needs, we found it still needed additional work to fulfil our key points which ruled it out. Creating our own engine would have been out of the question also, we do not have the resources to build a full featured search engine in comparison to existing ones, such as Algolia for example.
Algolia and their WordPress plugin allowed us to have as much or as little control as we wanted over any search-based functionality on Remodelista. As well as taking out large amounts of work in setting up an engine like ElasticSearch while processing posts and taxonomies that would be required for search. By choosing Algolia we had the power of ElasticSearch within a couple of clicks, while also meeting all of our criteria on our key points and features list.
One of these key features in Algolia, one that helped us make the decision to use the services, was the Travel section of the website. This feature is one of the more complex elements of the entire site, where search is concerned. The technical requirements meant that we needed to query the data based on geolocation (longitude and latitude) and boundary areas of a Google Map view. We found that Algolia offered a near perfect implementation, allowing us to create a powerful feature that enabled people to not only view posts on a map, but also gave the users greater control; such as filtering by type of location within single queries.
Algolia equips our team of developers and our clients with analytic data and metrics around their site’s search. We are able to learn about what people are searching, when they are searching and whether there are commonalities within their search terms. This might have included common spelling mistakes, synonyms or even localisation.
Fortunately when it comes to these issues, Algolia offers a number of key features to serve up the content a person might be looking for, giving us opportunity to cater and provide the right content for the person searching. Search term synonyms allowed us to serve content for terms such as ‘refrigerators’ when when they were searching for ‘fridges.’
Needless to say Algolia is powerful, a pleasure to use, and has become an important piece of our toolkit.