Voiceable: Voice-enabling the web

The brand BEFORE 

Voiceable’s original branding and product design was disorderly and didn’t leave an impactful impression on customers or end users.

The brand and product AFTER

We completely redesigned in order to convey to potential customers how helpful this technology could be, focusing on explaining the potential benefits to retailers: unique customer insights and an increased likelihood of purchases. For end users, the UX and UI were simplified for clarity and improved for usability, so it was clearer where to press and when to talk.

UX / UI and usability 

I led the team in conducting usability tests of the product to see what conversational UX and copy was working and what wasn’t. We saw that users were not reacting well to the lengthy copy; they preferred the voice assistant to say lines that were much more concise, and with a faster tempo. We also noticed an activation problem: more than 80% of users who pressed to talk - seemed to not actually say anything at all. 

After running usability tests, we saw that:

1. Users sometimes pressed-and-held the talk button, rather than tapping it, which meant that although they spoke, the assistant wasn’t picking up their voice. So we added support and animations for the press-and-hold gesture as well.

2. Users often began to speak to the assistant without allowing for microphone permissions. Timing exactly when the microphone permission pop up appeared increased activation rates by 25%.

Branded conversational copy

Each voice assistant embodied a unique persona, which was crafted around a template of the basic conversational design to meet end-user needs. Below you can see three versions of copy for the same action, using three different assistant personas.

Understanding customer needs is critical to success in a conversational product such as Voiceable’s. If you are looking to understand your target customers in more depth, tap here to check out my research methods.