You'll get access to over 3,000 product manager interview questions and answers
Recommended by over 100k members
Design a Vending machine for blind people
What is a vending machine?
A machine that
Offer a set of pre-determined products
Select the product
Pay for the product
Collect and consume the product
Clarifying questions
Is this a physical vending machine for snacks and/or drinks
Does this machine accept cash / coins and credit cards / UPI or all
Is this product built only for blind people or normal ppl as well
Is this product for fully impaired, partial, color blind or all 3
Answers:
This physical vending machine is to dispense packaged snacks. This vending machine is designed only for fully impaired users.
For this product, we would focus on the individual who can speak and hear but is fully impaired.
User Journey
A. Normal People - Visually abled
User walks to the Vending machine
User sees the items that are available in the vending machine
User sees what is not available in the vending machine
User makes a selection
User sees the price
User enters the quantity
User makes a payment (cash, card)
User takes the dispensed product
User walks away from the vending machine with the purchase.
Considering the above user journey, listed below are user pain points for the Fully Impaired (FI) individual
Locating the vending machine
Unable to determine which food items are available for purchase and which are out of stock.
Not knowing where to input the item to purchase and quantity
Not knowing the price of the item to be purchased
Not knowing where to swipe the card, security of the details being inputted (shoulder surfing)
Not knowing where to insert the cash, notes to use
Counting and inserting the right amount in the vending machine could take more time as compared with user having normal sight.
Finding where the product has been dispensed.
Solutions
Pain Point | Solution 1 | Solution 2 |
Locating the vending machine | Implement an audio output in the vending machine that plays every few 10 seconds | Implement a pathway that is FI-friendly so the user can reach the vending machine spot. This may not work in case the machine is moved away from the designated spot. |
User positioning wrt vending machine | Implement laser and camera sensors to determine whether the user is within the vending machine area and guide the user using audio so the user is aligned in front of the vending machine. |
|
Determine the food items available for purchase and their price. | Audio commands as output | Audio commands as output |
Making a payment | Designated payment area that plays a sound notifying the user that it's time to make payment. | Designated payment area that plays a sound notifying the user that it's time to make payment. |
Dispensing the product | A designated dispensing area that plays sound and vibrates notifying the user that the product is dispensed and picked by the user. | A mechanical arm that hands over the product to the user. |
User Feedback | Audio output – Voice input |
|
Solution 1 is more user-friendly. For the 1st iteration we can go with Solution 1 and later on, implement the braille keypad from Solution 2 if deemed necessary.
Metrics
# of users that completed the transaction
# of users that completed the selection but were unable to complete payment
# of times the user completed the selection, and payment but was unable to pick the dispensed product.
# of times the user was unable to browse through the available items and make any selection
# of drops due to voice input not being received by the vending machine.
Put signs in Braille on the outside of the machine. Design the signs in a way where they can be easily updated by the person filling the machine. Signs placed above or below product or product image on the outside of the machine.
Attach slots with ashesive, the slots are designed to have plates put into them (similar to the slot a name plate is put into on the outside of a door). The plates have on them written in Braille: product name, order code, price. These are three separate plates which can be inserted independently, to customize the product indicators for each machine. To prevent accidental removal of the Braille plates, give the plates a magnetic backing and use either metal slots or attach a magnetic strip on the slots. Alternatively, you could have one closed end and a clip at the open end of the slot to block the plates from slipping out accidentally.
To address product availability, use a buzzer which noticeably vibrates and signal "out of stock" with two consecutive short buzzes and signal "in stock" with one long buzz. Place the buzzer beside the "coin return" button/lever.
To indicate a product has successfully vended, utilize the buzzer to signal this with two consecutive long buzzes. If a product does not successfully vend, either by machine malfunction or because of the user (incorrect payment amount, unrecognized product code, payment denied, or vending canceled by user), use three consecutive short buzzes to indicate product has failed to vend.
There are several variations to the outlined solutions, depending on machine layout, product options, and user experience goals.
Instead of buzzes, you could play tones (low or high pitch or both simulatneously, if you do not have any users who are both visually and hearing impaired).
For machines with a set product selection, such as branded beverage vending machines, the Braille product indicators can be permanently affixed to the machine with the option of having removable Braille price indicators.
Product slots and plates can be made by a third party, the vending machine manufacturer, or provided by manufacturers of the products stocked in the machine. If this is a solution for an individual business/office that has a vending machine, it can be fairly easy to create your own product indication system by hand (use magnetic strips with adhesive backing and affix these in front of products/product images on the outside of the machine, use a Braille label-maker to make the product/code/price indicators and attach these to magnetic strips with an adhesive). The only way to implement the "out of/in stock" and "error/success" buzzers or tones is for that to be done by a certified mechanic employed to the task by the vending machine manufacturer (though if you are not violating the terms of use for the machine by doing so, you can certainly rig this yourself if you have experience with circuit boards).
I hope this helps.
Clarify:
1. Is it a physical vending machine?
2. The blind people we target are totally blind, but can hear and speak?
Assuming the answer is YES.
Identify User Groups
Blind people, but also technicals and operators, and for business purpose, we also need to support normal people.
I will focus on blind people since it’s our goal.
Report User Needs
1. When blind people pass by, they need to know where is the vending machine and how to get there
2. When they are in front of the vending machine, they need to know the items on sale, name, price and stock
3. They ‘choose’ a specific item and pay for it
4. Then pick up the item they bought
5. They can leave the feedback
Cut Through Prioritization
I guess 1, 2, 3 user needs are essential, while 4 could be the same as normal ones, since not that much hard to get it, and 5 might be considered in the future.
List Solutions
1. App for the blind people, it will notify them when there is a vending machine nearby, and ask whether they would like to have some snack or drink via voice and vibration
2. The vending machine can detect there is a blind customer there via App, or there is a big physical button for blind people to wake up the vending machine
3. Automatically introduce the items by voice, including name, price and stock, blind people can respond to it
4. Blind people can directly say what they want via voice input, vending machine give them voice feedback
5. Pay by the credit card or app without inputting password (normally no password required under $200)
6. App can be voice controlled in advance with a specific item or the user preference
Evaluate the Trade-offs
1. Impact 5, Cost 4
2. Impact 5, Cost 2
3. Impact 4, Cost 3
4. Impact 5, Cost 3
5. Impact 5, Cost 2
6. Impact 3, Cost 3
Summarize
I will choose 1, 2, 4, 5 solutions to provide a vending machine with voice and app support for the blind people buy some snack and drinks.
Design a Vending Machine for Blind People
What is the purpose of a Vending Machine – Offer a choice of pre-determined products to someone and help them make a selection and pay for the item.
Clarifying questions –
Is this a physical vending machine for snacks and/or drinks?
Does it accept cash/coins or credit card or both?
Does this vending machine need to work for blind people and normally sighted people or only blind people? – Both?
Do we assume that our blind users are 100% blind (legally blind)?
Lets say that the answers to all these questions is Yes.
What are the main needs for users of vending machines and what are the gaps for blind/visually impaired people?
1. Product Discovery – A vending machine has a stock of different (but limited categories) of snacks and or drinks. People usually “see” these selections to make a choice. People “see” if a product is in stock or out of stock. Gap for blind people – They cannot see the categories and selections visually.
Snacks are usually in the categories of chips, pop corn, bars (chocolate, granola), nuts
Drinks are usually in the categories of water, juice, soda
2. Product Selection – There are usually unique numbers associated with each product selection that a user has to key in or touch to select the product they want. Gap for blind people – They cannot see the number or the location of the product in the vending machine to key in or make a selection for what they want.
3. Complete Transaction – The user is shown the price of the product either up front or once the selection is made. They insert cash, coins or swipe their card to make a payment. Once a product has been selected and paid for, it comes out of the vending machine. The user takes the product out and goes on his/her way. Gap for blind people – They cannot see the price of the product and it may be hard for them to insert the exact cash/coins or swipe their card for payment purposes. It may be hard for them to know where the product came out once the transaction is complete.
Since blind people cannot rely on sight, they will have to rely on their other senses – touch, sound and smell. We can eliminate smell for this use case as the products will be packaged in a vending machine.
We can rely on audio and touch for helping with product discovery, selection and transaction
Features
1. Product Discovery and Selection: Product Discovery and Selection can be done via initiating user input or by reading out a menu that’s not tedious/long to hear. When the blind person comes near the vending machine,
• A very simple way to solve this problem would be to have a visually capable human being sit next to the vending machine and help the blind person make the product selection and choice.
• In the absence of a human being, by facial recognition through cameras, the machine can detect that this person is blind or as a simpler solution – a voice could simply ask if the person needs assistance with selection
• A voice prompt near the machine could ask for – Would you like a snack or a drink today? If the user says – snack- they can be prompted for chips, bars or popcorn. When they say chips- they can be given the 4 different options within the menu and the user can make a choice via voice input. Alternatively, the user could interrupt the voice prompt to specifically ask for a particular product they like and the machine can either confirm or offer a similar product/substitute. Over time, the vending machine could get smarter by learning the user’s preferences.
• Or there could be a mobile solution: There could be integration between an app on the user’s phone and the vending machine for the user to handle these voice interactions through their own mobile phone which would then communicate with the vending machine via blue tooth. Data would be stored about the user’s preferences on the vending machine app on their phone. The phone could also help with detecting the location of the nearest vending machine through voice assisted navigation. The phone/app could even let the user know if the product is out of stock and offer an alternative.
• Once the machine has understood the user’s choice, it can confirm the choice by asking the user and by asking for quantity, amount to be paid.
2. Transaction – Getting the product from the vending machine can be done in multiple ways
a. The selected product could fall in to a mechanical arm attached to the vending machine which would then start buzzing – prompting the user to move their hand towards the arm. Once they place their arm on the mechanical arm, it could guide the user to pick the selected product from there.
b. Payment information should be pre- updated in the mobile app account associated with the vending machine or in the vending machine payment system (similar to Uber or Lyft). A friend or a relative of the blind person could help setting this account up front. The mobile app interaction of selecting the product and confirming the purchase via voice would deduct the payment amount from their account.
c. Another option is to use crowd sourced apps where the blind person can take a picture of the bill/coin they have and ask the mobile app if it’s the correct amount. A company had done this recently using Amazon mechanical turk but I like the first pre-paid option much more.
3. Prioritization
a. Product Selection – I would build the voice activated mobile vending machine app instead of creating the voice processing capability in the vending machine itself . This would make the mobile app more generic and potentially help it work with different kinds of vending machines with different product selections. (I am not going in to data set up requirements here)
b. Transaction – I would build the buzzing mechanical arm to deliver the snack near the blind person so that they don’t have to worry about pulling the snack from a difficult position in the machine.
c. Transaction – I would build the pre-paid account capability to make payment processing easier for the blind person.
This would constitute my MVP.
4. Metrics and Launch Plan
a. Adoption and Engagement– How many blind users are downloading the vending machine app (app downloads), monthly active usage (how many are using the app regularly at vending machines), how many vending machines are converting to this model and integrating with my app (app integrations)
b. Customer feedback – I would get audio NPS feedback from my blind users to see whats working and not working – are they able to select products easily, is the NLP working well?
c. I would first test it in schools or colleges with blind students to see if they adopt and engage with this vending machine – app combination product as well as collect customer feedback. I would continue rolling it out to more schools, colleges, cities across US depending on nature of feedback, adoption and usage improvements.
d. Monetization – I would like to offer these machines at a discount and get revenue from the vendors whose products are being offered in the machine as this is for a socially relevant and inclusive cause.
The first step will be to clarify and confirm the question prompt.
Clarifying Q1 - Is approach path to the coffee machine clear? In case of any obstruction, do we have means to ensure basic accesibility to the blind individual?
Clarifying Q2 - Is there any manual assistance available in case of any difficulty to the blind individual?
Assumption - The machine is placed in a location which has very good accessibility for visually impaired people.
Next, we identify the user groups - Frequent users (power users), Regular users, Rare users. Of these three user groups, we focus on power users.
Third step will be to identify user needs and prioritize these on the basis of impact, revenue, and effort.
User Needs | Impact | Revenue | Effort |
Navigate to machine | H | H | L |
Select Item | H | H | L |
Place the mug at the right spot | H | H | M |
Complete payment transaction | H | H | M |
Get notified when the mug is full | M | H | L |
Provide Feedback | M | L | L |
The solution requires building a machine that has a sound alarm to notify the visually impaired person of the machine's location. Next, the person should be able to place their mug at the right place for the dispenser to collect the coffee. A marking and a slot at the lower end of the machine can enable the person place their mug at the right spot. Next the person selects the item through a braille enabled keypad and audio instructions. Once the selection is done, the individual completes the payment transaction using a combination of audio instructions and braille key strokes. After the machine dispenses coffee, it sounds a notification indicating that the mug is full. Lastly, the user has the option to provide a feedback.
In sum, we start by clarifying on key questions and identify the user groups. Next, we pick one user group and list down the key user needs. Thirdly, we build a solution to address above mentioned user needs.
Clarifying questions:
are we a big tech based company or a startup? How much resources can we realistically invest into this product’
Experience with vending machines, some limitations of use right off the bat
I’m definitely don’t use vending machines in my day to day life unless I needed to grab a quick snack - I do notice that vending machines can be hard to Navigate for the visually impaired. This is because
1) the items inside can often only be seen and there currently aren’t ways to know what items are inside unless you can visually interact with the machine
2)the placement of each item cant be known without visual interaction
3) some are cash only, this sometimes isn’t friendly for the visually impaired
Competitors:
None
Goal:
To design a vending machine that makes it intuitive for the visually impaired to access
north star:
revenue
Users:
Blind people
The general public
Where vending machines typically are stationed:
Schools
Work places
Streets
We will focus on workplaces as workplaces have more regulars who know of the vending machine and its location, as well as having money to spend on the snacks and beverages the vending machine has to offer, for schools and streets where different people go in and out it can be hard for our users to notice and use our machines.
Painpoints:
Not knowing that the vending machine is even there
Not knowing what items are inside the vending machine
Not knowing where the specific buttons and placement of items, resulting in not knowing what numbers to press and where to press those numbers.
Not knowing which payment methods are accepted, cash or card- or where the censor, cash slot is
Proposed solutions:
Having braille letters printed on the machine for each item as well as their numbers
Having a voice based mode of interaction for menu activated by motion sensor, allows user to navigate through items in the machine (with description if what the item is) and add multiple items to cart with voice commands, so the entire machine doesn’t need you to press buttons
Incorporate beeping sounds into the site of payment - the POS for cards and cash slots
Constraints:
Need to make sure the vending machine has no sharp corners and every component is safe to the touch
how do we prevent our users from being embarrassed by the attention the voice command system draws to them.
Questions:
As what company?
As a Startup
Time limitation or constraints in the engineers?
No
Let’s define a vending machine. What kind of products do you want to sell here?
Anything. You can decide
Analysis:
Users
Define the types of users that we have:
Blind people
Adult
Children
Elder
Support animals for blind
Friends or support for blind people
Not considered → no special needs or not different than for non-blind users
Let’s define their needs and pain points:
User Type | Needs | Pain Points | Comment |
Blind | See stores, vending machines, … as he needs to buy specific items | Not seeing: not seeing the kind of products, the price, … | |
Dog |
| not reading
No way of communicating what is available in the machine | No sense in pursuing with this user type |
User Journey
Our user has a need to buy something (e.g. Water)
Can buy from a local store (theoretically)
First issue: he needs to see the vending machine
Second issue: he needs to know what the vending machine is selling and if it is selling anything of his interest
third issue: he needs to see the price
Fourth issue: he needs to find the number corresponding to the product he wants to buy
Fifth issue: he needs to insert the coin and know when the coins are enough
He needs then to take what he ordered, when given by the machine
Eventually, he needs to take the change (connected to point 5)
Pain points
Pain points identified with the user journey:
seeing the machine
knowing the products sold
see the price
find the number / order it
insert the coin / pay
Notice that point 5 is already solvable if the machine is using cards; we can also assume that it is not an issue, as coins have different sizes and they are recognizable in the touch.
So to summarize, our pain points are:
seeing the machine
knowing the products sold
seeing the price
finding the number / order it
Prioritization
We cannot prioritize one unique set of these needs, as all important to complete the user journey and finalize the purchase with the vending machine.
Solutions
Vending machine that incorporates a voice recognition feature
User can communicate with the machine as with a human, as the machine would read aloud the options
Note: this does not solve for the “discoverability” of the machine
Connected to an app on the device that indicates where the vending machines are and read aloud the options
Subject to a network effect: we need to have many machines using the app for users to download it; but then potential for scale
They could for example include the stockings in the future and the location (usually now it is more often the case to run into a vending machine)
Touch device: write the content in a way that it is readable for blind people
Not very useful for other users
Smell device: e.g. vending machine that smells like coffe or the other products sold
Difficult to detect - if it is selling different thing at a time for example
To determine the solution to prioritize, I would focus on two things:
Scalability of the solution - if the solution can be useful also to other nieches
Potential impact on the users
Complexity
Solution | Scalability | Complexity | Impact |
1 | high | low | medium - we do not solve the discoverability problem |
2 | high | medium - on the business side, we need to have enough vending machine sold so that users would download the app | high |
3 | low | high | low - only affecting blind users |
4 | low | high | low - potentially not even affecting blind users |
To summarize
To conclude, ideally we should move forward with either solution 2, if we can confirm that we would be able to scale and apply it to enough vending machines.
On the strategic side, it can be applied to several types of vending machines, it can be used to foresee inventory levels, and it can help vending machines to substitute stores when closed (in some countries for example, some stores are closed on the weekend and allowing users to know in advance the inventory level before taking the trip to the vending machine can be advantageous).
Goal: Build a vending machine for visually impaired people to help them successfully purchase an item without the need to be dependent on caretakers.
-> Can I assume that the person is in front of the vending machine - Yes
-> Can I assume that the vending machine is for snacks - Yes
Users:
-> Mildly visually impaired
-> Partially blind
-> Fully blind
Since we have would like to design a solution that would help the visually impaired to be self able without the care takers, I would like to focus the solution around fully blind.
Use Cases
As a user, I should be able to
Understand the Items available | P1 |
select the item | P1 |
Pay the money using card | P1 |
Let the user know when the payment isn't successful | P1 |
Collect the dispensed item | P1 |
Ability for the user to be able to cancel the transaction | P2 |
Let the user know when the Machine is faulty | P2 |
Report when the item is not dispensed after making the payment | P2 |
Solutions
Solution | Impact | Effort |
Normal Vending machine with a small key box with item names written in Braille with an arrow mark to the payment slot | M | M |
A voice enabled system with a conversational experience to buy and dispense the product | H | H |
A sensor based vending machine which helps the user with the name of the item when they show their hand in front of the item | M | H |
Generally the vending machines are a 5*5 matrix
- Chips, Chocolates, Cakes, cookies, Ice creams
Wireframe looks like this
The Vending Machine identifies the user the moment they come in front of the Machine and starts by saying
Machine : Hello , How may I help you
User : I'm looking for chips or can you help me with the menu (bot would list the menu or if the user directly selects what they want, jump to the different items available within that category)
BOT : Sure, we have the following , could you please let me know which one who you like
User : I would like to have lays onion and sour cream
Bot : Item selected, could you please help me with your phone number
User : provides the phone number
Bot : I sent you the payment details link for this transaction
User : Makes the payment via the smartphones. Smartphones are enabled with voice over which reads the text on the screen and makes it easier for the visually impaired to effectively communicate just like the sighted people
Bot : The payment is successful and the item will be dispensed shortly (after few secs)
Bot : The item is dispensed. Thank you for shopping with us today!!
Metrics :
- No. of intents that were answered correctly by the BOT
- No. of Successful purchases
- No. of Failed purchases -> Wrong intent or due to the card issues
- No. of Intents that were started but dropped off
Risk
- The above solution does not work with people who do not have smart phones, per 2021 US statistics, 80% of the people buy smartphones
- Sometimes users do not like to share their mobile number or personal information and this could be one reason for disengagement. I would like to review the stats around the number of people who started the transaction and walked away.
- Since this is a voice enabled system and is built in a language that everyone can understand , we should go with the global standardization(English) and language might be a barrier for people who do not know how to speak in English
- Voice enabled system -> Needs a lot of training with sample data and needs to be thoroughly tested before launched.
1/ Clairify what the problem really is
Vending machine that sells widgets. I assume the widgets can be anything from snacks and drinks to those in the airport that the distibute things like headphones and tooth brushes?
Assume that this machine needs to be used by people who are both blind and not blind.
Assumes it needs to accept various forms of payment such as cash and credit cards
2/ What are the goals of the product?
Goal of the product is for blind, and non blind people to purchase a product from the machine.
Metric goal here is the number of transactions completed
3/Brainstorm who the potential customers are
Blind people ← focus group
Regular people
Own
maintain the machine
4/Brainstorm what their problems are
Do some foundational research
I need help knowing what products you are selling
I need to know the price the products you are selling
I need to know how to use the machine to select and pay
I need to know how to pick up the item
Do some foundation data analysis
How often is someone with limited abilites using this
What type of products do they need.
Do some foundational competitive analaysis
Look at other automated machines that have this problem like atm machines, ticket vending machines for trains and get some inspiration from them on how to solve the problem
Force rank what the problems our people are having. Get an estimate of what kind of impact solving these problems will have on our metric and progress towards our goal
Pick a customer and the problems they are having.
5/Brianstorm some solutions
I need to know what products you are selling?
Show a product list in braille similar to mcdonalds - simple printed out list
Have an audio jack similar to atm machines that have similar requirements.
I need to know how to use the machine
Braile interface with button selection. IN the previous step tell us which buttons to press for a selection similar to a call tree.
I need to know how to pay
Have prominent place to input cash and credit cards with braille above it.
I need to know where to pick it up
Make sure people know where to pick up something. A simple bell or sound of the product dropping or chime.
The noise of the opening.
Have an AI interface with asks what you would like, you approach it and you tell it what you want.
User walks up, machine asks whaat do you want similar to a teller or barista
Non blind people say whatever they want,
Machine can detect if it’s a blind person and asks what they want and provide
Same thing for payment for security purposes.
Pros and cons of each solution
Solution one is an easy mvp to cook up
Solution two is something that could be what our long term vision is.
6/How do you decide what to do first?
Focus on impact in terms of solving the user problem + progress towards our goal
In terms of impact - I would probably go with an all braille + audio jack solutions as its high impact and low effort to get signal. To get metrics signal I would see how many people use the audio jack and complete a transaction.
Third identify risks
The biggest risk associated with all of this assumes that we solve the discoverability problem of getting blind people to the machine.
The second risk associated with all of this is making sure things are always stocked, in the event that something is not stocked we need a simple way to tell the user the item is not in stock. A simple buzzer or voice to could help. Best way to solve this problem is notifiying the owner or maintence to stock the machine when it gets low on an item.
Top Google interview questions
- What is your favorite product? Why?89 answers | 263k views
- How would you design a bicycle renting app for tourists?62 answers | 82.5k views
- Build a product to buy and sell antiques.54 answers | 66.8k views
- See Google PM Interview Questions
Top Product Design interview questions
- How would you design a web search engine for children below 14 years old?36 answers | 42.9k views
- Design a library for the future.25 answers | 23.5k views
- How would you design a consumer application for a scooter sharing business?21 answers | 18.6k views
- See Product Design PM Interview Questions
Top Google interview questions
- How would you improve Google Maps?53 answers | 228k views
- A metric for a video streaming service dropped by 80%. What do you do?50 answers | 135k views
- Calculate the number of queries answered by Google per second.45 answers | 78.5k views
- See Google PM Interview Questions
Top Product Design interview questions
- Build a product to solve the dog poop problem.13 answers | 9.4k views
- How would you design a "Google Refrigerator"?13 answers | 8.3k views
- Design a social travel product.12 answers | 13.4k views
- See Product Design PM Interview Questions