This post was written by Booodl CTO, Rajat Kulshrestha.
By now you may be up to speed on our latest app release that went live last week: George has documented the journey thus far, our Product Manager walked through the thinking behind our map-based and chat-centric UX, and I want to talk about tech.
We had to tackle some rather large and complex technical issues with the release of this app. The simple proposition we explored, based on community feedback, was that users search for items differently and don’t always know exactly what they want. That took us from a catalogue-lead app, to one that embraces elastic search and caters to the needs of a much wider user base.
For example: instead of selecting a specific Aesop moisturiser from the brand’s catalogue and adding it to your list, you can add a vague want like ‘moisturiser’ or ‘face cleanser’. You’ll then get a notification when any nearby store stocks a product matching that description, whether from Aesop or any other skincare retailer.
We think it’s a much more valuable way to approach the problem we’re tackling and would love to hear your feedback on it.
We’ve gathered a huge amount of data to make this possible. To have a vague search product capability like this, you need to know all the brands and categories stocked by various stores. What we then do is:
- Interpret the search query: what type of product is this?
- Match it to nearby stores: find all stores that stock that brand and category.
- Return the result: serve those results back to the user in the form of an interactive local map so they can see where to buy what they want.
There’s a lot going on in the background to make it seamless for our users. Furthermore, the product is always learning and improving based on what users search for and add.
What do you think of the new version? We’d love to hear your feedback and ideas!