Audio UX for Blue-Collar Workforce

At BetterPlace, we largely deal with a population that isn’t well-educated and hence cannot read or write much. This presents us with very interesting problems to solve as we navigate our way to creating products that can be seamlessly used by our target audience.

One such product of ours is Attend-desk, which allows multiple employees to mark their attendance via facial recognition. The problem we were facing here was that our users weren’t able to understand if their attendance had been marked or not, which created a lot of confusion and also led them to mark their attendance multiple times, just to be sure.

The product managers shared with us how an Audio solution would be better able to provide feedback to our users along with the visual screens. As a UX researcher, I felt like I was in uncharted territory, the only audio I had been exposed to was Yo Yo Honey Singh. So after half a day of feeling overwhelmed and terrified I dip my toes into secondary research and begin concept mapping of Audio UX.

Concept Mapping of Audio UX

Building blocks of Audio UX-

1. Understanding that feedback is essential, in terms of telling our users if their attendance has been marked or not, if not then what should their course of action be so that they can do that instead of feeling confused and dissatisfied.
2. Conditioning is one of the primary principles involved in audio UX as the more our users will be exposed to the audio feedback the quicker they will be able to recognise successful attendance or error states.
3. It is important to use auditory solutions that draw user’s attention and help them differentiate between an opportunity or a warning sound.
4. Sonic Metaphor is a concept which is eliciting emotional responses through audio. Eg- Using chirpy/happy sounds upon successfully completing a task.
5. Understanding the importance of keeping the sounds as simple as possible so as to not confuse the users.
6. Make sure audio and visuals go hand in hand to create a more compact feedback system for the user.
7. It can also be used as a branding opportunity for the company to familiarise users with the company’s voice/sound.
8. Through this foundational research, we were also able to identify some things that should be avoided while trying to deploy an audio UX system such as ambiguous sounds, lots of motion graphics, heavy text and lengthy sentences.

The research also helped us form more questions to be asked to the field managers, site supervisors etc to gauge the present situation. We conducted internal user interviews to better understand current noise levels on-site, the sound systems we would deploy to support audio UX, what kind of apps employees generally engage with, any type of audio they are currently exposed to on-site.

Competitive Analysis

After our initial findings we again deep-dived into our competitors to take a closer look at how solutions are being practically implemented, it helped us gain insight into different elements of audio UX such as voice, pitch and tone.

We were able to deduce that an Indian female voice, maintaining an even tone in English and Hindi language was being widely used.

A multitude of factors affects an employee’s attendance, from not enough light on their faces to that user not being registered in our system. Audio cues for all these error states would confuse our users so we decided to club our error states and came up with 3 distinct audio statements,

1. Attendance marked successfully- rather than 2 sounds for login and log out
2. Please Try Again- in cases where user’s face is not framed correctly
3. Kindly Contact Supervisor- In cases where user has been terminated, suspended etc.

Testing out solutions

Part of being a UX researcher is making sure we test our solutions with as diverse a population as possible so that we can be sure that our solutions will be easily adaptable by everyone. We took on-site supervisors, managers and blue-collar employees across Delhi, Gurgaon and Bangalore as our target population.

We made a sound guide which consisted of audio cues along with sonic metaphors and Indian female voice feedback.

Upon testing we discovered that,

i) Almost all of the users were able to identify the voice for attendance marking, meanwhile sounds received a more ambiguous result of 66.7%.

ii) Users showed a preference of voice over sounds.

iii) The main concern of voice taking up a bit of time post attendance marking is easily solvable as error states have already reduced a lot due to the new visual system and an even bigger decline is expected as more people get used to both audio and visual system

iv) Users are most comfortable with Hindi and English.

vi) Some statements that directly came from our users were,

“Jiska hoga uska naam bol dega ye, bahut sahi rahega aap abhi ye dal do tablet me”

“Jinki attendance mark ho gayi hai uska naam bata de raha hai aur jiska nhi hua usko bol raha hai ki Try Again”

This was our entire journey of Audio UX research for over 2 months, we hope this blog finds anyone who is interested in this concept or learning about it.

Credits-

UX Research conducted by

and

Blog Writing by