We hosted a free-wheeling conversation with Hardik where he shared a lot of great insights about growing as a designer, working with teams of different scales; solving problems at different scales. Originally scheduled for an hour, we ended up talking for 2 hours.
We transcribed the entire conversation in order to share it via newsletters. After a week and many hours of typing, we had 20,000 words.
This is a lot of content; we should have trimmed it to it’s gist; we should have hired a professional editor. But we also wanted to publish it in a timely manner, so people can read it.
Yet, life took over and we ended up delaying it for a few months. Finally, here it is.
Hardik Pandya has been working as an interaction designer at Google for the past 1 year. A self-taught designer with academic background in Electrical Engineering, he worked with InstaMojo, Ola Cabs before joining Google in 2017.
Jui works as a User Experience designer at Aubergine Solutions, a UX design and engineering studio based out of Ahmedabad. Her satisfaction of existence as a designer is fed knowing how a designer’s job is to meet both- users’ needs as well as business goals.
How did you get into design?
I got into design purely by chance when my university needed a blog. I taught myself to design and write code in HTML/CSS to build the blog. Finding this interesting, I then built my own website with a few “uninvited redesigns” as portfolio pieces.
Someone from Bengaluru saw my website after we had a conversation on Twitter. They called me and asked me if I would be interested in a job for a designer they had at this company called InstaMojo. I said I was interested and had a talk with their CEO later in the day. This call happened on a Monday and Wednesday I was in Bengaluru working at InstaMojo.
How did the move from InstaMojo to Ola happen?
I worked at InstaMojo for 1.5 years. I got in touch with Sunit Singh, who was then leading design at Cleartrip. He was starting a new consulting company called Design Capital and I joined them. While at Design Capital, I got to work with Ola Cabs on cabs and Ola Play products.
And from there to Google?
Last year I was ready to make a move and a friend of mine had just got into Google. I applied and the overall interview process took about 3-4 months. Now a year into my role, I work on G-Suite and my secondary role is working on Google Photos. I also work closely with Gmail and Google Groups teams.
Working in companies of different scale and sizes
I thought I knew things while working at InstaMojo and I was already wrong when I joined Ola. Now at Google, I am even more wrong about what I knew. As they say,
The more you know, the more you know what you don’t know.
That has been literally the case and also what one calls growth. Now I always ask designers to strive to work on larger products as it increases the scale at you work on, but also because you also open yourself up for a wider array of challenges. That is how you can grow. When you work on a small set of problems and can apply knowledge from that without having to learn anything new, it’s time to move on.
That is also why the agency model did not work for me because I only had to become good at a certain things. Then I could keep succeeding within that known circle. I had to pick up new challenges and change that mental model.
The work we do at Google ends up making the company billions of dollars. But at the same time, the experience is humbling. You go out and talk to people and understand bigger and better products are out there. That’s generally been the difference.
Now when someone asks me a question, I step back and think. Earlier I used to give answers straight away. My answers are now more measured and I try to understand the context.
That maturity that comes from working with people who have done a lot of work and have failed at a scale is now there.
Feedback and design
What is your feedback loop at Google? How do you realize soon enough that you are wrong?
At InstaMojo we were 3 designers which was a really tiny team. When we came across problems or opportunities, we did a brainstorming session and, went out and built it. There wasn’t really a vetting, or validation until the product went out.
With larger products, the ramifications are also larger and you need to be more assured of your proposed solutions. So now the feedback loop involves more designers, and us talking with the users before building anything. You need more eyes looking at and more brains thinking about the solutions as it affects a larger array of products.
Those feedback loops are in place so you would generally not be working in isolation and things go through a hierarchy to get vetted. That started to happen at Ola and happens more at Google. Smaller startups do not have that luxury so they have to figure out a different process, but feedback loops are dependent on how you shape your teams.
Designing for enterprises
How do you design enterprise products where the decision makers are different from the actual users?
When I made the jump from Ola to Google, G Suite was hiring and I was assigned to them. I am glad I joined an enterprise team. It’s really challenging in a very different ways and it impacts design and product the most, engineering not that much. For design, the end user now is not only the person using the product but it has to sell to a CXO person as well.
In that case, design is hard. I don’t have a happy answer for you. Lot of design decisions are taken to tick boxes and to match the competition. Features need to be there to win contracts even when they are not perfect. But you need to walk that kind of environment to understand what actually sells the design. Whether you are selling to a CIO or a product admin, you have to understand the psychology of how to sell to people, how to prioritize your product in order to sell. So I wouldn’t say design is absent in that scenario, it just changes the shape.
So you have to shift your perspective - it’s always easier to convince one person to pay you a than convince thousand people to pay you . Consumer apps struggle to make money, while enterprise products make money. That gives the designers the freedom to try out new things, learn new things.
It’s always easier to convince one person to pay you a $1000 than convince a thousand people to pay you $1 each.
But if you notice Slack, their model is different. They get buy-in from the users who then ask the decision-makers to purchase the product.
Yes. Their (Slack’s) model is bottoms-up. They also focus on a different segment which is Small and Medium Businesses. Google Enterprise works in a top-to-bottom way. Bottoms-up works mostly for smaller product companies like Uber, Twitter etc. For products like that, the decision is made in a different fashion. CEOs think, okay my team likes this, let’s buy. For larger enterprise solutions with security and compliance come into the picture, the sales are made at the top and CXOs need to be involved.
So to go back to the previous question - having a feature is often more important than having it the right way. That is why the design - it’s quality and execution - is very different in enterprise products versus consumer products.
But be the champion for your users. You may get contracts by selling to CXOs, but your users will stay only if the product is good. How do you re-win the contracts? That is only by making usable, good products. As a designer your job is to make sure the product is still good even when the table-stakes are taken care of. If you are an enterprise product operating at and offering to smaller companies, making a better product is your best bet.
Designing for different platforms and devices
Have a platform team. At InstaMojo, we set-up a design system which helped us move quickly to the market. Any time a new designer joined they had a ready toolkit. Even when we were 3 people we made a strong foundation. I joined when they were redesigning the product so we had the chance to create the foundation.
It was difficult at Ola, because Ola did not have a structure in place. I joined when they were 20 people and they doubled the team. We started to implement new designs without retro fitting it, so all new features would have new design language.
It’s hard to reset design for products because few companies have that type of budgets. So you have to pick your battles and decide from what point onwards you want to implement new designs. Not many people want to fix problems created by other’s work so often it doesn’t make sense to fix old designs. Focus on what’s coming up.
For company like Google, they can dedicate teams to maintain visual language and have separate roles for that. As an interaction designer, I don’t have to work on the visual design. I only solve interaction and flow problems, not the visual problems.
But if you are small team, start with time-boxed efforts to create design system and implement it. You start with basic components and iterate from there. Set principles in place and take it from there. Definitely involve engineers, otherwise you will find it difficult to have their buy-in. Design Systems need engineering counterpart and in it’s absence you don’t have a system.
Designing with processes
When you are selling design, you are selling design processes as well. How do you make sure the processes are followed? How do you ensure impact while doing “stupid work”?
Design has two pillars - there is the user pillar and the business pillar.
As a junior designer, you are not very concerned with the business side of things. When you start out, you learn tools. You pick up Sketch, Principle, Photoshop, HTML/CSS - whatever you need to learn the craft. Then you learn the architecture and design screens. You learn about your user and how the user will achieve their goal through flows - where you should use a modal flow, where the buttons should be placed, etc. You understand your user bit by bit.
Growing beyond that you learn that your own biases are not helpful. So you realize your own way of using products is not how all users use them. That is when you work with researchers. When you work with great researchers, and understand how they gain insights you grow more in the user pillar. Now you know what the users actually want and you have the skills to design for that.
Now your user pillar expands more.
This is when a lot of designers hit the glass ceiling. They are doing great job in given problem, now they can replicate the same from one company to another. At this stage our designer is already at a mid-level company and she can jump to other start-ups and do the same thing.
After certain time it stops being a challenge because you are repeating the same things. So they wonder what next and how to grow more.
My theory is this is when you start learning psychology and human behavior. This is when you learn why people make certain choices, how they make them. That is when you are pushing the bar towards the maximum.
When you start to understand psychology, human behavior, history. That is when you start to understand the economy, the society and why we have built systems the way they are. That is when you understand why Uber as an idea works and how it impacts the economy, the society and how it impacts urban planning. That is when you start to think about the larger problems. That is when you start to become a designer. The Jony Ives of the world don’t think about UIs or products. They think about the impact on humanity. It gets a little philosophical but I think it does have to get that. You have to think about the larger picture at that point.
The other pillar is interesting one, the business pillar and this is where you start hearing things like “seat at the table”. If you search for the phrase, you find hundreds of articles about design and the seat at the table. I always feel design always had a seat at the table but designers weren’t asking for it. They were never ready for it and never preparing for it.
Designers only get their rightful seat at the table if they bring the same level of accountability to it. Who gets a seat at the table? Engineers, managers, CTOs, Product leaders - they have accountability and responsibility that they will bring ideas to execution and take ownership of it’s failures. Designers don’t do that. Designers are not taking up that responsibility enough. That is why we don’t have that seat and designers are considered at step in the process instead of a continuous process it’self and stakeholders.
How do you do that?
We can only do that by understanding business and we don’t do that enough. We don’t take interest in how our companies work, how our balance-sheets are formed, who are our suppliers and customers, what are the marketing and sales strategies. Until we understand that, we are not very important. If your only output is mocks, you don’t matter.
If your only output is mocks, you can be replaced tomorrow by a person who doesn’t bitch about it.
If you want more accountability, increase your coverage. Look at people who have done great things, like Julie Zhuo at Facebook. The only way to become a design leader is to help your company solve business problems. There are hundreds of problems within organizations that the product managers are grappling with, and unless we understand those problems and empathise with them, we can never be stakeholders. In this entire process, there is no hand-off. “I am there during the entire journey, I will help ship the product and I will also close the feedback. I will help with how the product landed.”
Launch is not measure of success; you can launch new products and features all day. Landing is the hard part and it takes accountability. Designers need to take that accountability.
It requires a change in process but it can literally start with a single designer. I was lucky to have a team of designers at Ola who got regular face-time with Bhavish (Ola founder) and that was incredibly helpful in understanding the business. He understood we brought something of value and we went out of our way to bring insights that were business-related; choosing the right development platform, right messaging and the right MVP. Designing should be part of all of these. That’s when you start the business pillar up and that is the only way to grow as a designer. You can become a principal designer and all the way to CDO. Those roles do not have any responsibilities with tools at all. You are not responsible for tangible outcomes, but more for strategic and business outcomes in the long-term. So you are measured along those yardsticks.
The funny thing is, the user pillar you can learn on your own. You do not need a lot of stakeholder help in that. You can read and learn on your own to get to the top of the user pillar. You cannot learn business on your own. You have to get out of your chair, talk to your product and marketing leaders; you need to go out and understand what you and the business are doing. You should understand what process shortcomings you can optimise.
That is how you can have more say as a designer. User pillar is completely self-served, and business is completely collaborative. You have to decide as a designer where you want to grow. I wanted new challenges in the business pillar to grow. You may decide to become a solid “user designer”.
Working with deadlines
How do designers work creatively within dead-lines?
The role of a designer is to initiate conversations. In any stakeholder meetings, designer can start conversations around trade-offs, prioritization, what can be done within deadlines.
Ask questions. When you are short of time and you feel like your process is being shortened because of lack of time, start asking stakeholders why more time cannot be spent on design. It may not always win the battle. Often times the deadlines are made up, unless there is a strategic advantage to be had by being first in the market. When I worked at startups, people communicated wrong deadlines all the time. So for design the time is so short, that you actually have no time at all. The only way to change that is to ask questions - why is the time so short? Are we doing the right thing? Send these questions to the stakeholder - along with receiving work, you should also send this work to them. Unless you have all those answers, you should be comfortable in saying that you don’t have enough information to work on. “I can design but I have no way of being sure that this will be effective design.” It’s perfectly acceptable to push back. I think we do not push back enough. It’s hard to do in a lot of companies, I know, but it’s perfectly alright. You may be afraid it comes off as you don’t know the answers but that is fine. Good teams understand when you do not have any inputs to design with. We put time on engineering efforts but not on UX efforts. We need to make stakeholders understand that for good UX, we need information and inputs and we will not move a pixel unless we have that. Otherwise it’s wasted effort. Explain that just how devs do not do throwaway work in engineering, we cannot do throwaway work in design.
The Agency Life
You are asked to give estimated hours. How can we even estimate hours for any product where solution is yet to be designed!
It’s hard as an agency in a transactional environment. You cannot go to a shop and simply buy design. So agency life is tough. It’s not the right model to become a product designer. You can be a great freelance designer but you have to work with that mindset. If you want to become a product designer with ownership in the final product, move to a product company. I am saying that because you would be spending your time wisely in that kind of environment as the design gets really challenging in such environments. In agency, the processes become challenging. You do need to overcome the challenges thrown by the processes but not because the work model is such.
As I see, there is no growth in agencies either. You might land better clients and better paychecks. And I realize I am saying this from a privileged position. Lot of designers do have to start as freelance designers in agencies, but know that that is not the end goal but a stepping stone towards more ownership.
There are freelancer designers in the market who have the respect in the market that allows them to install right practices and processes in organizations. Because when a product manager client is wrong, you should be able to call it out irrespective of whether you are a consultant or an employee. Push back should not depend on that. You should be able to push back that a food app does not need to look like a cab app.
The point is people need to get off their laptop and do the work in the field. The insights are not on internet; they are out there in the market. So as any type of designers, collect information that makes the lives of product managers easy. Educate people why their approaches or assumptions are wrong. Hitting the assumptions is the right way of questioning people. When you question their assumptions, it’s impersonal. If an assumption is wrong, it may make the approach wrong so discuss that.
That has helped us at Google in becoming objective and impartial stakeholder.
Do not question people and their approach; question their assumptions.
Everyone makes assumptions, not just designers. CEOs make assumptions. What is a strategy? Strategy is an assumption that market will behave in a certain way and we will be there if and when it does. Strategies fail when you assumptions were wrong or misinformed.
Communicating and delivering design
What should design deliverables focus more on? Functional aspects or visual aspect?
I have a solution that also works for product companies - how do you communicate your design.
At different stages it works differently. Now a days, when I am working on a design problem I take notes. When I am done taking notes, the design is done. When you decide that these are the steps to a goal, this is what and how the information will be presented, your design is mostly done. Now all you need is to break it down in terms of space and time. Space as in layout and time as in what comes after another.
Visual design is about space and time.
When we communicate with PMs I can communicate with notes and maybe flow diagrams.
Basically it will be pen-and-paper so we can define what’s required and feasible, and a general representation of the design manifestation is complete.
With clients, they want to see tangible things. So I have a system for that. I was at Figma office recently and I am working with them to provide some feedback on their app, in how it could be used to communicate design. It’s a great tool for this purpose; it isn’t there completely yet but it will get there soon. So for now, tools are not the answer. Processes are the answer.
There are two communication streams - upstream and downstream. Upstream is with your product leaders and stakeholders and downstream is with your devs. There are different fidelities involved with both these audiences. If you send flow diagrams to a CEO, they will not get it immediately. You need to know who is your audience for that deliverable and what feedback you are looking for. With a product manager, you want to know if the flow and step is right so you have to ask those questions. A level-up, tell them the entire context of your flow and low-fidelity deliverables might do the job.
Above that, you have a CEO who needs to see the product the way a user would see. This is where hi-fidelity mock-ups would be valuable. Treat your CEOs as your users; just like your users have no value in wire-frames.
To communicate at these levels, you need different fidelity. So set up a component system. Reuse components to churn out different flows at those fidelity.
Annotate your flows and share it with the stakeholders so they understand how complex or easy a flow is and they have the information to make a decision. There is no perfect solution yet, but we are getting there.
Every stakeholder is looking for different things in a mock, by the way. An engineer is looking at how many APIs are going into this screen to support the design. Product manager is looking at how much time it will take the user. The CEO is looking at if it makes sense to the user and whether it looks good. Everyone is selling this deliverable to the upper party. Help them sell it and help them move this process further. Don’t be a hurdle.
Battles are fought at every fort in the company and you have to provide each stakeholder (with) the right ammunition.
I have had CEOs ping me to ask for mock-ups to share in investor meetings. Investors will never invest on wireframes, they need to see a product.
The evolving nature of design
How do I handle the evolving nature of design in a design system? How to add new patterns without killing the consistency?
Walk before you run! How you create an MVP across all your functions is a critical thing. The development teams need to create frameworks that are minimal and viable; similarly, design also needs to create deliverables that can be validated. Can you validate that one feature? Based on the validation, you can make case for spending enough resources to the stakeholders. You need confidence in the feature before you invest in it and that confidence comes either from running validations with users or from launching. Many companies are agile enough to run an experiment with a small set of users and learn whether the proposed solution worked or not. Then based on your learning, you can micro-pivot towards optimizations. No one needs full feature apps right away, so you can always build piecemeal featurettes and learn from them. Agile processes have solved this and you have to implement the same in design also.
How much design is enough for you to validate your concept? That is a simple solution. Let’s take an existing example. Let’s say Uber ratings don’t exist and you want to know how the ride experience was for both the end consumers and drivers. What would you do first? You can validate by putting a feedback form in the cab that the users can fill out at the end of every ride. People can simply tick check-boxes to let you know their feedback. That’s when you know people really want to do that. Now you want a centralized system where you can gather all the feedback, and then create a system so the drivers are affected by the feedback. So now you want to make the process electronic and on a server, and show the interface in the app.
Your MVP was a form; when you gathered enough forms and saw that 100s of people actually ticked the form you had the validation of the idea. When this has already happened, it’s a no-brainer to sell this idea to the stakeholders. Now you can approach this in incremental fashion in the product. You can start with up and down thumbs, then move to granular star ratings. Then loop the granularity of ratings to business logic.
You didn’t even need technology to validate your concept. You did not even need engineer or mocks. You still designed a feature with the help of a traditional method. That is what restaurants do with feedback forms. They gather the feedback and when they know how to synthesize the data, maybe they will use an app for that. You don’t have to be Agile with your mocks, you can be agile with your approach. That you can do without any tools.
Designing with data
As a designer, what are some learning paths to understand data to design better? In your experience, how does that impact design at Google?
Understanding the basics are a good start. Understanding sample sizes, variance, probability, market segments, your measuring methods and what you are measuring - these are all valuable topics to learn.
I recently worked with a researcher who collected a lot of information from support tickets. We went through those tickets and identified that the problems we were trying to solve in products, should actually have been solved in processes. The way the help center article was worded gave an inaccurate impression about something in the product.
How did we analyze the problem? Simply by looking at the data at a scale. I often hack my way through this - for example, searching for keywords in our support tickets.
If you want to learn about data at scale, you can plot the data in different ways to understand it’s impact. For example at Ola we noticed a lot of our bookings were coming in at specific times. We wanted to build a solution for the daily commute, because that’s a guaranteed market-share. How do we cater to a demand so that we could make sure the supply was available when the demand rose? We looked at the data for all the booking timings and locations found that certain residential areas were sending lots of cabs to tech parks. And in the evening, there was demand from tech parks fanning out to all these area.
We dedicated part of our resources catering specifically to this demand. And this data analysis gave us a huge raise in revenue. We had to forgo demand in lot of other geographical areas to cater to this structural demand, but those are the decisions you make in a product and business.
There are various ways to look at the data, but as a designer you have to figure out how to use it to your advantage and to fail less often.
Working at Google
Does Google specifically place you in a role so you know the boundaries of your role and that of a data analytics person?
Yes. There are things in my role which are “must do” and there are things which are “good to do”. The Must-dos I have to cover in order to go up the promotion ladder. My role at my level is clearly defined. Each employee needs to tick the must-do boxes in order to be eligible for promotion. Promotions are not based on potential but on retro-active performance. So you should already be doing job of next level to earn your promotion. So the good-to-do things are what you can show to build your case for promotion.
Google uses a lot of data for design decisions. Have there been instances when data has picked a decision no one expected to win?
We have a saying that data always picks your least favourite child.
Data always picks your least favourite child.
If I want to work at Google, what should I be doing?
There are different roles, although interaction design I see as a blanket role, but you may target a specific area of expertise and then gradually branch out so you may say that I enjoy Visual Design and I want to do that first. If you enjoy Interaction, UI/UX, although I can’t find differences in how the life would be but yeah if you find an area of expertise, I think the only way to build enough credibility is to create a body of work that speaks for you, create a lot of work even if it is unsolicited design of favorite apps of yours. That’s how I started, I liked Tweetbot back in the days and I still love it and I redesigned one of the screens of the original Twitter app to make it look like what if Tweetbot designed it. So just mimicked the whole thing and made it in Sketch, and tried to make it as polished as possible because that’s the only way I could measure design at that point, if something was polished, it looks good. Now I know different, but… So yeah, create a lot of mocks, lot of interfaces, lot of different ways of solving the same problem. For example. If you want to create five different ways of Sign up, what if you want a sign-up with least number of inputs, what if you want to create a sign-up with max number of inputs. What are the pros and cons. In what kind of a product would a lengthy sign up make sense, in what kind of a product would a really quick sign up makes sense. That’s when you understand probably, a person who wants to quickly book a cab and just downloaded the Ola Cabs, doesn’t care about the sign up, all you need is a phone number and that you can take after booking the ride - so you don’t need a sign-up there versus (if) you’re signing up for lets say a Credit card app and you definitely need all the information that you can to not have fraudulent sign-ups. Create fake case studies, what would you do if you were to design a simple thing as sign-up and show that you can put a thought to it even to a seemingly simple piece as that. That’s one way to do it. You can also show UI design skills in that, you can also show illustration skills in that, you can also show little bit of interaction in that, how does sign-up not just a screen, but sign-up experience feels like. What if you want to ask for OTP and what does that experience feel like. So start that way, do a lot of work, you don’t need somebody giving you a lot of work, you can just pick up unsolicited work, let me go and re-design xyz app, what if I was the designer working on it, how would I do it? That won’t get you to Google, that can get you to a company that can get you to Google.
Google doesn’t hire kids?
They definitely do and one of the sure shot ways of getting into Google is to get an internship. They do open internship slots, the interns in Google are properly vetted. So it’s like half of the hiring process is done at that point and most end up converting full time. So that’s definitely one way to enter Google. The sad thing is that in the internship program in India, the seats are limited. Definitely limited to the full time openings. So there are a few in Gurugram, one in Bengaluru and there is one in Mumbai. So there’s not many, as many as in Mountain View. In Mountain View, kids literally join Google as an internship. that’s a de facto- Google, Facebook and Whatsapp. There, they are more willing to lengthy, mature internship programs. Sadly, not as much here. So we focus more on full time hires than interns. Positions are limited, but you can always try.
You met Luke, right? I’ve been following him for almost a year now and I find him very unconventional (compared) to what designers are. He analyses past trends which is not something that designers normally do. They focus in the present, the future, never the past.
So, Luke, I think it’s been his passion. Documenting evolution of devices and documenting evolution of human behavior, thumb positions and things like that. He has always been like that and that’s why he works in two projects right now at Google that are actually leveraging his expertise in those areas. He doesn’t work on any specific product right now, he works on horizontal things and he works in Product Excellence, that is something that is horizontal and helps a lot of other teams in Google build better products. He threw his insights and knowledge. Products is- ’Prod-ex’ as we call it, is definitely a vertical that Google has invested a lot in. He is really a unique case, if you get a chance, do pick up a talk of his on youtube, ’Future of Mobile’, I attended in person in meetup in Palo Alto. That’s one talk he keep giving all over the world, it’s a beautiful talk. He talks about the whole evolution of how we started with these mainframes and all the way to these tiny computers in our pockets. it’s very interesting, and I would say if you have not been following him at all, the talk is a great way to start, to know what he is all about and then just follow him on twitter and go through his infographic, they’re really informative, they’re really good. And he is generally responsive on twitter.
He is not a designer, by the way, now. He is a Product Manager now, he changed the vertical. He is in the product manager ladder and he has designers reporting to him which is very unique, which is not the case anywhere in Google. He is the only one who is a product manager, but has designers reporting to him. Generally, in Google, designers report to designers. Unless you are at a VP level where you could be VP of design reporting to a VP of product.
I watched a movie last week, ‘The Internship’, is it true or it was just a concept?
I think Google internships are slightly harder than that. You are basically doing a job, you just have low level of accountability and you are willing to screw up. You’re there to learn, in that if you end up helping us as a team, fair enough. But that’s still secondary objective. You learning is the primary objective and that’s why the internship program is designed in a way that helps all the students and people from academics sharpen their skills and produce a body of work that can be presented at their university for a final semester project.
It’s definitely not cycling around the campus and eating free food. It’s more than that.
Internships at Google
Internships are team-specific. Every team has internship positions open, if you end up joining a team that is an AI, they don’t have an AI team in India yet, we’re setting one up, so maybe in future we have an AI internship but there are many internships in the AI teams and assistant is one of the biggest AI products in Google. Daydream is another one, daydream has certain AI components. There are some other verticals as well, unannounced products and Pixel is one of the other verticals. Depends on what product you join, and the definition of AI would be around that part of AI, not AI in general. So there are no generic AI internships, but you would be working with AI in one of these really specific product areas.
Interns always work on real projects, they don’t work on imaginary projects or any briefs that are not going to be built into products. They actually work on actual products, and they have deliverables given to them. So they work on 6 months to 9 months cycles and they are given a specific agenda, at the end of that many weeks, they have to go through reviews and they have to match a certain mark and that’s when Google writes a certificate for you that says that this body of work is eligible for a final year project and while also being meaningful for the company. You may not always moving mountains in the company but you might still be helping company out with the very key part of the project.
More about work at Google …
Teams in Google work as startups. Google Photos is not the team I work very closely with. They are not corporate by any means, they are 17 people design team which is small in Google. In a company of 3000 designers, 17 people is a tiny team. These teams work like a startup. They work with the very iterative design approach, they are a small team, you don’t have to get a lot of approvals and getting things executed fast, they are very quick and dirty in their approach, they want to try out different things. Google doesn’t necessarily go out of their way to position themselves as a startup because the teams are basically like that, unless you are in a huge org. My day job is in Google Cloud, Cloud is a 20000 people team vs Google Photo- it’s a 500 people team. There are so many teams in Google that operate as startups. Google Duplex, it’s a startup team, Assistant used to be a startup team, they’re pretty big now, multi-location. Maps, definitely not a startup, NBU used to be a startup, Tez worked as a startup under NBU.
Contextual Design and Transparency
What do you think about the system gathering some data about the user, and user’s awareness about it. For eg. Netflix records where you stopped last time, and next time you visit, it resumes from there.
Context matters. it’s very specific to problem statement. Whenever you are reproducing something that depends on something that the user did in past, it’s really a rule of thumb to inform the user why they are seeing this. You see it in Google search today, for example you visited a page 7 days ago, we could simply be putting that result on the top, you may even realize that I was on that page, but then you start questioning how does this thing work, you might think what all do they know about me, where all are they injecting this kind of a content. So I think drawing the boundary of what you are seeing and what everybody else would see if they perform the same action, that distinction is really important and should be done.
e.g. if you search something in a logged in state on Google, you would see different results, I would see different results. And somebody who is new to Google and Google doesn’t know anything about them, they would see different results. Now you could be presenting personalized results with a demarcation saying, ‘Because you searched for xyz in past, we are showing you this. That’s a good way to do it. That’s a logical way to do it. And also tells the users that they are not seeing the pure web, they are seeing the contextual web, they are seeing their web. Especially in current times when the information gets affected by the bottles that we create around us and we have seen a lot of problems because of that. Facebook was forced to do it in feeds, a certain feed entries, why are you seeing this? Because you indicated your interest in XYZ. Linkedin does it all the time, you said you are interested in Artificial intelligence, there are 5 company post about artificial intelligence. You need to inform. It’s also one of the core principles of Google’s AI principles. I don’t know if you read them recently, Sundar publicly published the AI principles because of a lot of backlash they got from Duplex. When an AI is talking to you, you should tell them that you are talking to AI because that’s a breach of trust, you might be thinking you are talking to a human being, but you’re not talking to a human being. It’s a huge gray area. There’s a lot of morality involved in that as well. I think a lot of companies take decisions based on what they want to do, there are business reasons involved. But here’s where I stand, I think users should have visibility in what they need to know, where their future actions depends on them knowing, if they are taking the right action or not, then you should inform them. If your future action doesn’t depend on you showing a distinction, sure, that's an exception, but the other thing that you mentioned about currently watching state- that’s multi session user journey- you leaving things off and coming back again. That is generally good user experience, coming back again to the same product or you were doing something on the desktop and picking it up again on the phone.
Tejas: With Netflix, they may be approaching the problem this way - 80% of our users want to resume, for other 20% we have a very good rewind experience.
Hardik: I have a rule of thumb - ‘Frequent User journey should be easy, infrequent user journey should be possible, not easy but possible.’ This rule of thumb is helpful in a lot of ways, for example, if you know that 80% of the users just resume because they don’t want to watch it again, make it easy for them - and make it possible for the folks who want to rewind. So as long as you have a way to rewind, you would be making 80% people, you can make everyone happy anyway, so you just optimize for larger user base and their DNA.
Showcasing your work and building visibility
Anecdotic-ally speaking, I got my first job on Twitter. Did I do anything extraordinary? Not really.. Just had my website link on my profile and the person who ended up calling me was the one who was trying to figure out Spotify in India and he wasn’t able to login and I was able to help. He looked at my profile and saw who is this guy. And he said okay this guy says he is a designer. Ha. And he has a website on. He looked at the website. I had actually spent a lot of time building the website.
There is a long story on how I bought the domain, if you are interested in this kind of thing. There is this podcast called Design Details, if you follow that. Design Details is a fun podcast by two people Marshall and Brian, very good friends of mine. I was on the podcast episode 250 (240), and I told this exact story. How I bought the domain, built my own website, How I got into HTML CSS and built a website when I had no prior knowledge. If you have time, just give it a listen. So yeah, he reached out me through the website and that’s how it started.
I think the website had enough information to convince the CEO that I was the guy who can help him out based on what they needed. We didn’t even negotiate the salary, they were offering 3x the amount I was expecting. So I was like, “Okay, fine.” I said, “Why this much?” He said, “You’ll probably need this much, Bengaluru is a little different than Ahmedabad” “Okay!” They paid for my flight, I flew to Bengaluru and lived on Domino’s pizza for the next three days because I didn’t have a house, I only had to order, eat out.
Three-Four days of the job I literally had no idea, I was sitting like a zombie in the office I had no clue what I was going to be doing in the company, didn’t know my purpose. It’s a scary situation to be in. They were paying a lot of money for the person who didn’t know what he was going to do. But yeah it all worked out, they asked me to build an app on day 11, because they thought we should give him some work, because he has a laptop and he sit’s all day, he says he is a designer. We have a lot of work so lets give him some work. Five days later I made a complete mobile app for their web product which was 75 screens - largest project I’d ever done. They hated it! They said, ’ha, this is not what we want” and I thought I was packing my bags and coming back to Ahmedabad. What they liked was, this is my insight but I think what they liked was that I put a lot of effort in understanding what Instamojo was, what every one of the folks there were doing, who were we building for, and why we were building the way we were building. I asked some right questions in the meeting that ended up convincing that I was the right guy.
Those people I still consider friends today, they are more than colleagues now. Couple of them I work with regularly. One of them works for a startup name **Dazzin)** in London. ****It’s the largest sports podcast in Europe now. They recently signed Cristiano Ronaldo as their brand ambassador. Every one of our team at the time is doing spectacularly well, thankfully. So yeah, full marks to the guy who actually reached out.
What I am trying to get at is that hiring in today’s world happens in really serendipitous ways. But people can only find you if you put yourself out there, right. I think a lot of really accomplished designers also don’t have their presence out there about what they are doing, what work they are doing, what are they all about, How many complex problems have they worked on, what have they helped the companies with. And I think there are two or three things you can do really, well, to actually fix that.
Have an online presence.
At least a twitter account because most of the design community is on Twitter. Very receptive, very supportive, they share a lot of useful things and I think you would be wise to put up into that, you would be a fool to miss out on that. Have atwitter account, follow the right people, ask the right questions, reach out to them, ask for help, they are all always always helpful. I have met some of the really phenomenal designers I never imagined I would be sharing coffee with 6 years ago, I was fortunate to meetup, of course being in Google helps. But even before that, I had a really strong presence and my identity helps me a lot.
Write about design.
Write about problems you’re facing, write about problems you’re trying to solve, if you are facing problems in your freelance world, in your product company, share how you solve them, share the story how you got some quick wins or how you got some really long term major wins. The more you share, the more you self-learn, the more you yourself put your work in a perspective. Everyone’s doing important work. We take it for granted because we are so involved in it that we forget the bigger picture. The moment you start writing about it, you feel how much of an impact it had.
I did that when I built my own portfolio, I realized how much of an impact our small team had in Ola where we made them lacs of money quarter over quarter. You wouldn’t realize that everyday office in an office out, until you sit down and put everything on a page, you realize I think I did a lot of work in a year. Build a case study, build a portfolio full of stories about the kind of work you have done. it’s hard work, it’s not fun often times because it’s growling work. The only reason why I got into Google in 3 months which usually takes 6 months is because of my portfolio. I spent 3.5 months of extra work that I was not getting paid for, after hours, keeping up late at night because I wanted to have a portfolio that spoke for my work before I introduce myself to the interviewer. You need to have your work speaking for yourself before they even have a first conversation with you because you might not even have a first conversation with them if you don’t have your work speaking for you. And design as a field is so crowded, there are thousands of designers in Bengaluru and million outside, you need to differentiate yourself, you need to have an identity that sets yourself apart from the other folks. The leverage is- most people are not doing this. Most people don’t go out of their way to build an identity, to share their work, to talk about their work, be present in the community, share knowledge and help others, they don’t do that. The more you do that, the more you create value and the more people remember you, you create that level of leverage for yourself and the differentiation that people generally seek and good stakeholders, everybody wants to work with great team members, how would you say that you are a great team member? Your mocks won’t say the story, the story only comes out by the words you share, by the personality you share, the way you present yourself in conferences, in meet ups, in meetings 1:1 with other folks in the community
Go out of your way to spend time with folks who are doing great work, share your stories with them, share your stories online, it’s incredibly powerful, it’s like a megaphone that you don’t need to control 24x7. People search for you, they already know more about you than you would ever be able to tell them in a face to face meeting. So why not take that effort! Tools are not important, you putting yourself out there is important and having a consistent personality is really important. And in a crowded world, where everybody is doing the same things, differentiating yourself is the only way to grow in your career, the only way to land really shiny opportunities, and I am saying that from my own experience.
Chase people you always wanted to work with. For example, I did that with Sunit Singh. I had heard a lot about him when I was working at Instamojo, people were all about him- “He is this designer who built Cleartrip and he is one of the smartest people in the industry right now and if you ever want to grow as a designer, you should absolutely be working with him. That night I literally searched him all over the place and I learned about the kind of work that he has done and sent him a message saying, “Hey, I am a designer who works at Instamojo and I want to have a coffee with you and want to talk about design stuff, there’s no agenda as such.” He said, “Okay, lets meet up.” This is a big deal, people don’t normally say yes. I don’t know why he said yes. But after that first meeting, it’s about how… You want that first meeting so that you can have the second and third and fourth meetings. Your first meeting is to make a break.
I presented my work, I shared my story of why I want to do what kind of things I want to do, I wanted more scale, I wanted more product products, wanted to make mass market products. He said okay, I am actually looking for- you know what, everybody is always hiring by the way. You should know that if you come off as somebody valuable to them, you have a job offer, not in the first meeting, second third or fourth meeting. Or they at least remember you to refer you to the right place. So he said, “I am setting this up in Ola, why don’t you join me.” I said, “Okay, yeah.” I thought that was the end of it because in Instamojo there was no interview. He said I am going to put you through this team, you will have to talk to these 3 people. All three of them gave me three separate design exercises. That was new to me, I had never done that. I put two back-to-back all-nighters and got back to them earlier than expected. I was actually worried that when they say give it a week, come back next Monday and we’ll discuss it, me doing it early might come off as cutting corners. They might think he might not have given it 100% efforts. that’s not how it works in startups, thankfully. They said, “You gave it a fair shot, you gave us a solution in two days, I think it looks good, when do you wanna start?” I said, “I haven’t talked to Instamojo yet, so I need a week”
Anyway, the gist of the matter is conversations matter. You meeting people matter a lot, and if you are not in a situation or not in a geography where there are lot of inspirational designers, have conversations with them over Skype, over phone, doesn’t matter. All you need to do is leave an impression that helps you pull yourself out of wherever you are right now. We all want better work, we all want better companies. So look for the people and chase them. This is what I followed. I literally abused Twitter, to the point where I DMed 4-5 people every single week, not in a spamy way, there are subtle ways to reach out to people.
Next 5 years in UX
How do you see next 5 years or coming years in UX, will the design system make the designers disappear in coming years?
Great question, I am glad people are thinking in that direction.
You know Paul Graham, right? He said future is already here, it’s not equally distributed yet. 2
The future is already here, it’s not equally distributed yet.
William Gibson, Necromancer
It depends on the market you are in. If you are in India, next five years are not going to look much different than they are today. If you are in the valley, year is changing year over year, faster than anywhere in the world. The answer here is that there are some skills we as designers have that are easily commoditized, for example. Every job has a certain level of function that can be automated, and that will be automated.
There was this funny video that somebody had posted on Twitter, where an eye was literally studying the user interface and coding on the side. I think tools will get there, sooner or later. Not so much in India, but in the western world, yes I think design as a job as we know it. It will not go away, for sure not, it will change, it will manifest into something else, and I have been thinking about this a lot personally because I strongly feel that your tool skills are replaceable and the faster you equip yourself with more skills that are beyond tools, better positioned you are for the future. If you know sketch well, fine. If you know prototyping well, fine. Know that those jobs won’t be jobs on themselves. There will be a part of the job that you’d do in future but they will be nothing called a UI designer in future, for example. An AI would be able to create four different variations of the interface in seconds. So it’s faster, it’s cheaper, it’s more scalable, so people have no reason to pay you, to deal with a person, they can deal with a tool.
Bridging the gap: Digital Design and User Experience
I think technology is trying to bridge that gap now. If you know Peloton, the bicycle startup that started in San Francisco. Peloton is an exercise bike which is internet connected, so that it can collect your health data and create profiles and things like that. Nothing special, but industrial design of it is so well thought out, and the technology behind it of it collecting information about your heart rate, respiration, precipitation, perspiration, muscle mass and all that. It builds a whole data profile about you that would not have been possible with normal exercise bikes. That is a good example of things that are out there in physical world, while also being present in the cloud and having a digital profile. Those kind of products are coming up.
Nest is the one that I think probably started it but I think the future is in experiential design and AirBnB is a great example of that. How do you go beyond your screens. I think there was this great article, I’ll share the link but it went into details like how AirBnB actually hired a design leader who previously didn’t have a background of digital product design, they are from HCI background and some societal work in the past. 1 They designed the end-to-end AirBnB guest and host experience, and that’s what made the company so special. Because they elevated such a person to a position of head of design instead of UX-UI design, because AirBnB and a lot of it is similar to Ola, products like AirBnB have their manifestation in the real world. Even though you purchase it on the phone, that’s 5% of the journey, 95% of the experience happens out in the real world. If your host doesn’t come to receive you at the right time, if your keys are not properly explained to you, water purifier, and washing machine is not properly explained to you; you are in a remote location, at a place where you don’t understand the language, your on-boarding experience is broken, and you would probably not give the company other chance, you would book a hotel because that’s more predictable. When AirBnB started designing this experience, which is comparable to hotel while still being personal and friendly, even in across geographies, across language barriers - this is what they though about - ‘How do we design and experience, not a product’. The experience is the product, ultimately. The moment you start thinking about design as something that is outside screens, outside pixels and outside interfaces, that’s when you understand that you are having a much larger impact even though you are designing screens because there is also this service delivery that happens in the real world that you should probably be thinking about. In Ola, we said that no matter how beautiful the app is, if the car comes with a dirty seat, the mats are not clean, and the driver won’t turn on the a/c, you would not give the company a second chance, you would give a bad rating. Doesn’t matter what you app says, if you promise a service, you deliver the service in the real world. You as a designer need to own that service delivery aspect as well. So we ended up going to partner care centers where the drivers are starting their day, and we understood how they start their day. We even put a program in the training part where we mandate the drivers to clean their cars in the morning because we realized that none of what we do in the office matters if what happens here does not match up the mark. Those problems won’t go away, we just have to know that they are design problems.
Service design problems won’t go away, we just have to know that they are also design problems.
We just have to take them under our wings, we have to make sure that we understand those problems and we help solve them. That’s how you do it, and that goes into physical product design as well, industrial product design as well. If you think an exercise cycle is designed as is, and that’s the final form, you would never know what more could be done there. You need someone who is willing to innovate and has some design knowledge to go and disrupt that whole market. It’s a $2000 bicycle, people are buying it left, right and center. e.g. a lot of bike startups are coming up in San Francisco like Lime and Bird. Why do they come up with those kind of things? Because there are problems in society that technology cannot solve but this is also way design can solve those problems, like transportation, parking, pollution… Broaden your horizon, read things that are not about UI design. I think UI design you can just keep chugging away and be better at it. But these things if you don’t learn and if you don’t understand, you will never be useful tomorrow. You might be useful today, you might be earning today, tomorrow you are replaceable. So, yeah that’s what keeps me up at night.
Expanding your horizon and inspiration
(On taking inspirations) There are definitely design teams out there. But I would say I derive more inspiration these days from history and understanding the society. Reading a lot of books is definitely one way to improve your understanding of why we have designed things the way we have designed today. Architectures is one of my big topics of interest. Another one is Urban design, there are some really good authors that have written books on urban design, like going to designing safe spaces for people. Why are European cities designed the way they are designed vs Chinese cities the way they are designed. I think there is a lot of design knowledge out there that actually play at human psychology and human sensitivity. For eg there was really good anecdotal piece that I wrote about one of the things that I read in books- about the town squares that are there in all European cities. In medieval times, people needed a way to safely gather in a town when outside parties attack. Sure, you had castles, but even when your castle was not able to support you, there will be this town center that has only four corners at the entry points and everything else is cordoned off, and you can literally see in all directions, the horizon is in your visible point, and you can see who is coming from the top, who is coming from the corners. It was easy to protect, it was easy to feel safe in those enclosures, and it was easy to guard from the towers that we there in the adjacent places. Every medieval town in Europe is designed that way. I would have never known about it if I didn’t read about it. That goes into psychology of if we are threatened, we as humanity, what kind of behaviors do we resort to, how do we protect ourselves as society, how do we protect ourselves as individuals, and that’s when we start to understand how our mind works.
There are some great podcasts that goes into these historic things, from really good books, but again this is how you keep updated about the other fields and lot of designers have said this, a lot of designers say that great insights about design are not in the design books, in the field of design, they are outside, and you can only learn if you are willing to indulge into those things.
Again, I think I have a lot of links and books and resources, I would love to share if you are interested. May be we can do a follow up email to all the people who have signed up and I can send some resources.