Kiran Kadambi (EBay) on Innovating within an Existing System

Voiceover: Welcome to The Inter-Disciplinarian, where product managers share their stories and insights from the field.

Alex: Thanks for joining me. I’m your host, Alex Cowan and today you’re going to meet Kiran Kadambi who is a product manager at eBay. Now, particularly if you’re at a place like eBay, there’s a large successful infrastructure. The features that you’re gonna work on as a product manager don’t operate in isolation. They’re part of a much larger infrastructure and while you’re innovating and thinking about how to make the customer experience better, you still need to have an eye on that big picture so that the whole system and the whole experience still make sense.

Today, Kiran is going to tell us about what it’s like to introduce new features in the context of a much larger system.

Kiran: Thanks, Alex. It’s great to be here. Just to set some context, I’ve been at eBay close to 4 1/2 years now, and one of the key big projects that I was involved in in my time here was to build the in-store pick-up products for eBay. Now, in-store pick-up is kind of a new form of delivery option where a lot of retailers are building lockers or letting people buy online and walk in-store and pick it up. Our idea was to- how do we get the same concept onto a marketplace like eBay so that eBay becomes your one-stop destination to buy whatever you want and get it delivered however you prefer? So, along those lines, when I started working on in-store pick-up, what I needed to do was understand the full ecosystem that eBay had because eBay has a lot of legacy infrastructure as well as a multitude of features that need to work together.

Alex: So the basic idea was that I buy something on eBay and I can walk into a location a few miles from my house or my work and I can just pick it up, I don’t have to worry about how it gets dropped off at my house, or the shipping or anything like that.

Kiran: Correct. Correct. That’s perfectly right, Alex. It also gives you more convenience and control about how you want to take control of your package. So, if you’re on the subway and you want to stop for a cup of coffee before you head home and the store is right there, you send it to that store and walk in to pick up your item.
So, having said that, given that eBay has this complex marketplace with a lot of features in place to make it safe for our buyers and sellers, one of the key things that I had to keep in mind while building this product was to think about how this product would fit in to the whole ecosystem. I’ve used this term before, but what this entails is that I had to use systems thinking to understand how this feature would play in.

For example, what was the seller experience like? How would sellers upload their store inventory and let us know that, hey this is the store location, and postal code 95054, and how would we surface that inventory to buyers and how would we [inaudible 00:03:38] inventory list. What I’m trying to make is the whole ecosystem are right from listing, putting an item for sale to buying through refunds, through any claims, so it was essential for me to think through the whole value chain, in trying how to build this product. In order to be successful you need to account for all the key flows and to do that one of the key characteristics a product manager needs to have is something that I call “systems thinking.” This would enable you to put what you thinking of the product and it’s features, so what you’re planning to build, into perspective of the bigger ecosystem so that you make the right decisions and things don’t surprise you later on.

Alex: Hmm-mm-hmm (affirmative). How did you do that?

Kiran: So, the different tools which can help you actually do this and all the tools that I use was something known as creating a mind map. What that means is, it basically lays down, at a high level, lays down all the features that your product will use in some sort of- giving it a visual representation, so that as you go down this list, you can use it as a check-list to make sure you’re thinking through all the aspects and use it to communicate this to a wider audience and use it as a brain-storming tool [inaudible 00:05:05], and as you-

Alex: So you-

Kiran: Go ahead.

Alex: So you’d start with a blank sheet of paper and you would write down all the major chunks of the infrastructure of eBay and how they would interact with your new feature. Is that right?

Kiran: Correct. There are a couple of approaches. One approach is to write down all the list of features and build on how they would interact with your product. The other approach – I’ve use both approaches during my time here – the other approach would be to look at the other way around, which is more top down, in which case, you start off with your product and have a quick summary of what the key attributes of it are, and then I drill down and into it. For example, I know eBay does seller’s list items and buyer’s buy items, and then there is shipping and then there is returns and refunds and such.

Alex: Hmm-mm-hmm (affirmative).

Kiran: So, the second level of drill down after the initial summary would be that these here are the big areas being impacted, now let’s write down selling. Then I, again, expand that and blow-up selling into it’s other components which eventually helps me drill down to a level of detail which I can then use to brainstorm and fine tune in terms of what would be the different areas that I should look at.

Alex: How do you progress from that to implementation specifics with the rest of your team to actually get all this stuff working and in front of the user? What happens next?

Kiran: That’s a good question, Alex, and it’s an intricate process. Once you identify how different the systems are impacted and what are the different systems, the initial process was for me to talk to these people to identify the scope of things that need to change in order to support this feature. In those conversations, sometimes things might be very easy, sometimes they might be difficult, and having been through those conversations, you eventually end up a very good idea of what it needs to take, what is the scope of the activities that are needed. You use that information to go to the next aspect of it which is in planning and prioritization and budgeting.

Alex: Got it.

Kiran: So, once you take that information through that process, you are better positioned to take it to the next step which is working with your team to flesh out the details of individual aspects and then-

Alex: How do you do that? Do you use user stories or do you guys sketch it out, make prototypes? How do you do that with your teams?

Kiran: Well, working with different teams in a larger organization what you’d notice is that different teams have similar yet subtle different approaches. Now, if the team I was working with- for the most part, we had user stories. For the team that we are working with so we should [layer those 00:08:14] stories out, I would use two Balsamiq layered rough mocks or sketches in terms of what we have in mind.

Alex: Hmm-mm-hmm (affirmative).

Kiran: Part of this phase also was working with design and research, where we come up with more high definition proof of concept of what we’re thinking which we can then [inaudible 00:08:35] work with and conduct user research to kind of dig deeper and make sure that what we’re designing is in line with what our user expectations are.

Alex: Got it. That’s great. So, you get to this point where you have something you feel like you’re ready to put in front of the user, let’s say, then how do make sure that it’s really delivering the results that you wanted, especially in the context of all these moving parts? What were the metrics, for instance, that you used to test the different facets of this in-store pick-up feature?

Kiran: Sure. One of the key metrics in the eCommerce world is what we know as GMV, or Gross Merchandise Volume, so it’s pretty much how much sales happened on eBay. The other aspect that we are really attuned to is the conversion rate. So, out of the X number of people coming in and attracting with your feature, what percentage of them go on to complete the transaction. That is one of the key metrics that is at the heart of any eCommerce business.

Alex: Sure.

Kiran: Those are some of the high level metrics that we use and how it boils down, and I can give you an example of that which will speak well to your other question. So, one of the features of an in-store pick-up that we use is to let our buyers pick a store location near where they live or where they work, which would be convenient for them to pick up the item that they bought on eBay. In order to to that we had initially thought about providing an interface where our buyers would go ahead and see a list of stores and make a selection and then they would take it from there. But what we found out was that the conversion for that feature, which was not a key aspect of those was not really that great initially. Then, using some user research, we came up with the hypothesis saying that this could be because our buyers could not spatially locate this location of the store in reference to areas of location that they understand. So this lead us down the path of-

Alex: So, basically, if I know my route home, from work, I want to make sure that I can just drop in and pick this thing up, but the types of locations that you were presenting them it wasn’t easy to figure that out. Is that accurate?

Kiran: More than not being easy to figure it out, I think the issue- I’ll give you a good example. Let’s say that you bought an item and you want to pick up the item at Best Buy store on Ivy Road, and Ivy Road is a little generic, when you give the name of the store as XYZ on Ivy Road, it distracts the buyer from the purchase flow because they had to go and look up the address on Google maps or some other application before coming back to complete the purchase. The more users leave your site or leave the flow, that’s less conversion for you. So what we had to do was understand why users needed to do that, and we did that. One of the reasons we found was that users prefer – or rather generally people love to – when you give them an address, get a quick visual cue about how it stands with respect to other landmarks in the area. So, if you show them a quick map of, hey, here is the store, it’s right next to the intersection of 238 and Ivy Road. That immediately gives me a location on the- the pinpoint of the location. That’s on my way. That is right next to my workplace and I’ll just drive down and pick it up.

Alex: Got it.

Kiran: So we used that and we ended up including a map component within the store widget that we have on eBay. We took it to testing. We did some user research using that version and we were really surprised to see the results with conversion increasing by a huge margin as opposed to other conversions on eBay. That was a good learning for us and we incorporated it into our product and we saw a huge gain in terms of conversion and eventually GMV which helped make this a successful product on eBay. Eventually we expanded it to other countries where eBay has a presence.

Alex: Originally we talked about the other systems that you would need to kind of attend to and look at to make sure that the insertion of this facility worked with the rest of the eBay infrastructure, both customer facing and internal. Can you give us a couple of examples of places where you had to spend a little bit of time to make sure you understood that you were compatible with the rest of the infrastructure?

Kiran: Sure. So, keeping the same product in mind as we walk through this conversation, I initially touched on how our sellers had to give us the item location and the store information and such. Now, this is something new to eBay because earlier the eBay world involved sellers shipping their items to purchasers of their products, to buyers. In the new world what we learned while doing this is that we had to find out a new way or a new method to capture this data efficiently and make it easily accessible to different systems within eBay which required this location and inventory information.

So this was, for example, something that came out of that exercise and that lead us down the path of working with the right teams to design the system in such a way that it works with the eBay inventory system but adds the element of location and quantity per location concept with it.

Alex: Got it. Before we close, I’d love to take a few minutes and talk about your career story. Before you were at Darden, you were a solution architect at HP. Can you talk about- what was that job like and what things that are important to you now did you learn on that job?

Kiran: Prior to my time at Darden, like you said, I was part of the financial services team at Hewlett-Packard, more in the capacity of solution architect. So what that entailed there was to think about how we build scalable financial systems for retail banking operations that HP could then bundle along with their other products [inaudible 00:15:58]. In terms of what I- it was a good experience for me to learn more about financial systems, but also more along the lines of how do we- when you think about retail banking applications or retail banks and see how they operate, there are a lot of touch points with technology that I had to understand-

Alex: I bet.

Kiran: (laughs) understand a bit in terms of making sure I build whatever I was architecting at that time to be successful and to be- I mean, if people have to [add up 00:16:33] it has to work in an intricate ecosystem. So that got me started down the path of design thinking and systems thinking and thinking from the perspective of “how do I make sure this works in the whole ecosystem and not just in isolation?”

Alex: Then you went to Darden. Now, you graduated in 2010, is that right?

Kiran: Correct.

Alex: So let’s go back to 2008. You’re the Kiran of 8 years ago, you’re just starting business school. What advice would you give now to that version of yourself? Where would you focus, what things would you do?

Kiran: Good question. The advice that I would give myself from 8 years ago would be have more confidence in myself, focus on prioritization, do a better job at prioritization, stress on systems and design thinking, make more friends and have more fun.

Alex: (laughs) What is it that you learned at Darden, for example, that helped you prepare for your role now?

Kiran: As a product manager I can tie it back to the example I was talking about. I had to work through all aspects of it, including our marketing teams to kick of campaigns for product launches as well as with our accounting and finance teams on how to reconciliation of flows and taxes and such. Coming from- I have an engineering background. Coming from an engineering background, I truly was unaware of most of the aspects in the business world outside of engineering. Personally, for me at Darden, that was one of the core areas where I really learned a lot and enjoyed the process of Darden. The other aspect was also in terms of development of a lot of the softer skills and people handling and people management that you learn at Darden.

Alex: You have a great story about your process from basically graduating Darden to starting your job at eBay.

Kiran: I don’t know if it’s a good story, Alex. When I graduated in 2010 the job market was not that great. We were just coming out of the recession, or actually I think we were still in it, at least that’s how it felt for somebody who’s graduating and looking for a job. I had natural inclination towards technology and during my internship – I had interned out on the West Coast as well – I figured that there’s no better place to be than the West Coast, right in the heart of Silicon Valley, so I said there’s no point in wasting time. I got rid of all my stuff in Charlottesville, put my clothes into my car and I just drove. I started one fine day and drove out to the West Coast and landed in San Jose. Got a accommodation at a place which was really shady. As a new graduate I didn’t have too much money, but went about that and eventually landed my first gig.

Alex: Which was at eBay, is that right?

Kiran: No, my first gig was at Walmart eCommerce.

Alex: Oh, that’s right.

Kiran: So, landed the first gig. What really helped me through this process was an underlying belief in my capabilities and you know if you have it in you, you can just go out there and make it happen.

Alex: How long did this take you? Between you- you land in this dive in San Jose where you were living, to okay you’ve got a job at Walmart eCommerce now. How long did that take?

Kiran: That took me … So I landed in San Jose in early June and I landed this job sometime in September.

Alex: How many interviews do you think you did?

Kiran: I did a handful of interviews. I don’t recall a count off the top of my head, but it was a handful. Too many people were not hiring people, who were cautious about hiring during the economy but I landed a few interviews because I was right in Silicon Valley and knew that I wanted to be in tech. It was the right place to be, so I ended up getting a few interviews and eventually one of them worked.

Alex: What advice do you have to MBAs that are interesting in doing what you do, that are interested in being product managers in high tech, particularly for those that don’t have a stem undergrad or engineering background?

Kiran: Good question, Alex. The first thing I would say is kind of outright, you don’t need a stem undergrad or engineering background or a coding background to be a product manager. In that period I worked with all different types of product managers and some of the really good ones don’t necessarily have a technology background. But, what you do need is a love for technology and an understanding of how things work. By that I mean, understanding, for example, let’s say you have a smart phone and you have apps. So, at the high level understanding what the capabilities of apps are is really required, rather than understanding the code-level features or something like that. Having that and having the people skills to be a PM, the patience, the perseverance as well as knowing how to have fun and get along with people. Those are the kind of skills you need. Not really coding skills.

Alex: Got it. Well, thank you so much for joining us, Kiran. This has been great.

Kiran: Thanks, Alex, thanks for having me.

Alex: All right, so, a couple of things that I thought were really interesting about what Kiran said is, number one, the way that he maps out the whole system and I’m sure that he’s worked hard to improve his understanding of that. So, as we design a feature, in a place like eBay, we don’t just look at our individual feature, we map out all the big, moving parts of the infrastructure that we operate in and we go and use our capability as interdisciplinary innovators to go out and understand or interface those other areas and validate how we’re going to interact with those successfully.

Thanks so much for joining us. If you would like to learn more about building great products and features, I can’t help but mention our agile development specialization, which is available on Coursera at Bitly High Agile, bit.ly/highagile.