Another thing you can do is just straight up ask your users if they think specific recommendations are good. Just like you can ask for explicit feedback on items with ratings, you can ask users to rate your recommendations too! This is called measuring the “perceived quality” of recommendations, and it seems like a good idea on paper, since as you’ve learned defining what makes a “good” recommendation is by no means clear. In practice though, it’s a tough thing to do. Users will probably be confused over whether you’re asking them to rate the item or rate the recommendation, so you won’t really know how to interpret this data. It also requires extra work from your customers with no clear payoff for them, so you’re unlikely to get enough ratings on your recommendations to be useful. It’s best to just stick with online A/B tests, and measure how your customers vote with their wallets on the quality of your recommendations .

Quiz

We’ve covered a lot in this section so far – but it’s important to understand what makes a good recommender system before we start trying to build one. So, let’s reinforce some of what we’ve learned with another short quiz.