Most Relevant Results First — Personalizing Search in online shopping UX
Takeaways from UX Heroes episode featuring Carina Urban-Dasgupta, head of Webshop Management
How hard is it to design an intuitive online shopping experience? I have been in this field for a while but user behavior never fails to surprise me in some use cases. So when I listened to this incredible episode of UX Heroes where the head of webshop management at Humanic explains what they did to win the Best Online Shop award and how they personalized their search; it improved my own design process so much I decided that it was too good not to be shared with the whole wide non-German-speaking world.
You know me; diese Mädchen ist in Sharing verliebt 😉 So here we go.
I know it’s obvious but: Research Competition
That’s a no-brainer for all designers, of course 😅 . But to build any successful product it’s important to research your competition; see what they do right, what they do wrong and think about what you find out. See what might work for your own product, don’t just patch up a product of all the nice features and behaviors in other people’s products.
Let’s assume that you like the checkout flow of a certain app but the technology your dev team is using has limitations that prevent building a similar flow for your app. In that case you take what you liked about that flow and tailor it to the technology that you are using. And so on.
Really take the time to do competitor analysis, you will notice the value later. Or that’s what Carina promises 😀 (it’s true though).
Focus on Search — Placement and filter personalisation
- Make search the hero. Use your UI super skills for that 😎.
- Personalize search: I know you want to show off your products and get customers to dig through categories and see all your shop has to offer but believe me, they came with at least a vague idea of what they want. You will have your chance to show them all the other products that may interest them, but your goal is to:
Show the most relevant results as soon as possible
For that to happen, you need Data. That’s where marketing personas come into play:
Create Marketing Personas > audience building
use online and offline user behavior data to plot all the possible search scenarios and then split the personas who go through these journeys into groups by: age, gender, preferences or whatever makes sense according to the data you got.
That is not easy due to the sheer size of available data but the trick according to Carina is to start with the simplest, most straight-forward journeys first and then work your way and personalize, bit by bit.
For example; a working mother in her early or mid thirties may come looking for a pair comfortable heels for work but will definitely be interested if you show links to kids/teens shoes in a strategic point in the journey before she checks out, how about you throw in a discount and hep her save a little by buying more for less? be smart about how you design the flow.
Filter arrangement
Filter arrangement really does make a difference; for each persona group, the filters they need the most could be completely different.
If you come looking for a really expensive designer or brand items the price filter will not be the most relevant filter to you for example, you will probably care about how new a design is and what other variations of this item the store has to offer (style, color, …). It’s all about understanding the people and categorizing them correctly to show the most relevant results first and complete the journey as easily as possible.
You will keep changing and tweaking as you see users react to your slight but studied changes to search; keep testing and collecting data and use it all to plot the perfect journey for each group.
Common Pain Points in online shopping experiences
- typos & autocomplete woes 🤦🏻♀️; if you can build a powerful enough search engine that suggests the correct search terms or fixes typos you will take off a huge roadblock in a shopping journey.
- Dead ends: when a search does not yield any results; suggest something as close to what they may be interested in as possible. Always hold their hand and help them get out of a dead end and steer them in the direction of things they can explore.
Advice
- Always; collect data about how people are using the app. It’s your one source of truth.
- When you decide to make a change; it pays to do an impact/effort study to be able to prioritize the changes that really make sense at any point in time with your teams availability and resources.
Most common mistakes
- Sticking to own ideas as true. We all do that, there is no shame in it. We just need to grow up and treat it as a hypothesis that needs proof 😂. More often than not it is not. Data is king; it decides and splits Real Truth (yes, capitalized) from mere assumptions. Carina says that wrong assumptions teach us more about users than those we get right, which is 100% true.
- Being afraid of personalization if your business is small and your resources are not that huge. Everyone can benefit by improving their shop’s UX, very small changes can completely revamp the shopping experience for your customers.
It really is a great episode. You don’t win a prize for the best online shop by winging it, the Humanic tam know what they are doing 😎 I wanted to share and if I missed anything, excuse my German 😅.
Until next time! Lots of love and #keepdesigning!
