Next AMA: On Product Management [with Shashank Mehta of Razorpay]

I have a B.Tech in Electronics and Communication. The courses I loved the most in college were all humanities courses! Of course, correlation not causation etc. applies here.

As I wrote in another answer (link), if you understand people, you can do product thinking. Humanities definitely helps with that.

I don’t think it will ever become a hiring parameter, given the dearth of great product managers. However I strongly believe it will help good PMs become great. I wish I could establish through a series of logical statements but I can’t. And hence it isn’t a hiring criteria either.


We have followed a very common-sensical approach to scaling Razorpay. We did whatever felt right to us at that time and tweaked it as we scaled and things no longer seemed right.

Our entire product org (PMs, Designers, Engineers, Tech Writers) report to Shashank Kumar who is the co-founder. He is the CTO and CPO for us. We haven’t ever had PMs reporting to engineering folks, which is the norm in some places. It never made sense for us.

Our product team is broken into independent units: Core Payments, Payment Apps and Platform, X and Capital. Each has its own product leader who reports to Shashank Kumar. These are normally at Director level. The PMs in each unit report to Director or Associate Directors (who report to the D). These two positions are the only people management product positions.

Side note: Marketing is part of product org too. We are building exactly what we are selling (APIs, software) unlike Flipkart where you are building the ecommerce platform to sell the mobile phone, as an example. So the marketing team, which is marketing the product, which is what we are building, sits along with product team.

1 Like

this is brave :slight_smile: most PM teams suffer through this.

Partly answered it here: Next AMA: On Product Management [with Shashank Mehta of Razorpay]

We prefer to have end-to-end ownership on product segments. It’s not easy to setup though. For example, in payments product, do you separate ownership by:

  1. Payment APIs/Checkout for clients
  2. Backend processing systems working with banks


  1. Card payments (from client APIs/Checkout to processing)
  2. Netbanking payments (same)


We used to follow approach (1) which worked when we had way more engineers than PMs. Now we are shifting to (2) as we try to create self sufficient pods which can own the segment end to end.


While working with large demanding enterprises, how do you ensure that you don’t deviate from your core product philosophy? Did you ever come across a make or break “feature” deal which was not in your roadmap / not fancied by the majority of your other clients

Every business is a mathematical function. Master the variables and you are golden.

Eg: A super simplified example
Rate of growth of revenue for Netflix = (rate of growth of new users - rate of churn) x pricing

It’s your standard mathematical form of f(g(x)), really.

So I strongly recommend understanding the levers of growth of the company you have joined. Then look at things from this lens. Things like vision, customers, features etc.

Another thing I recommend is to understand the reason why every job role exists in the company. Why is there an account management team? Why is there a digital marketing manager? What are their KRAs? This will give you a crash course on what matters for the company and why.

This isn’t an exhaustive list. So I have tried to cover some not so usual models that I like!


But surely, the ‘dearth of great product managers’ should mean that a relatively lateral entry to the field must be encouraged? A company like Google already does it since so much of the universal appeal-ification of their products depends upon knowing people and cultures as much as technology. But I do concede that it is perhaps a luxury that only large companies can afford.

1 Like

Your PM tools? What’s the product stack at Razorpay?

  1. How to select domains to expand into? We looked at things which are related to payments and where we have an inherent advantage due to our payments business
  2. How do you structure the team? If you treat them as features, you risk putting lesser focus than you need. We created fully separate business units. RazorpayX has its own dedicated product, engineering, design, business teams who are laser sharp focussed on growth of X
  3. How do you let these teams operate like a mini startup and not a 4 year old business?
  4. How do you launch a new platform without confusing your existing customers?
  5. How well finished does it have to be given that there’s already a quality expectation from Razorpay. We can’t afford to look like a brand new startup in terms of platform quality

These are some things from top of my mind. Last one year has been absolutely crazy as we launched X (neobanking) and Capital (lending)!


It’s become way more science than art! And in a good way.

As the first PM at Razorpay, I spent most of my time trying to keep my head above water. The number of engineers was way more than the PM:Dev ratio ever recommends :joy:. So I picked my battles and focussed on things where the requirement of PM was necessary. Also, with the team size being smaller, we had less dependency on structured documents to convey product direction, concept notes, roadmaps etc.

Now we have structured product management. Closer to what you would expect from good tech companies. We do quarterly planning, write concept notes, PRDs, manage GTMs in a way more robust manner etc.

Honestly, I love how well we do our product management now. The structure has never felt constricting in any way, which is one common fear when you add processes.


nice ! What’s the PM:Dev ratio? Is that even a think you guys are focused on, from a strategy point of view?

Last 5 minutes fellas. We will close the AMA

This is really good. Simple yet effective tool for founders/PMs to come up with metrics critical for growth

1 Like

I made the transition from engineering to PM at Razorpay. But was even tougher was going from an individual contributor PM to a manager of product managers! I had to learn how to have product catch ups, how to review and critic concept notes without having the luxury of spending as much time on it as the PM had. I had to learn how to get predictability in feature delivery. I had to learn to take decisions with 60-70% of info. I had to let go of trying to double check all the facts, trying to do my own first principle thinking to all problems. It just was not possible.

Managing PMs while leading the product strategy was way harder than I had imagined it to be. This is a topic I would love to talk about in more detail some day!


Do you change product strategy mid-way? Let’s say - when you have to, how have you dealt with such situations?

As a Product Leader, my day looks very different from the Product Managers who are IC.

This isn’t all in one day but a mix of things across days

  1. Check health metrics of the product
  2. Clear mails/slack
  3. Review a few concept notes, PRDs
  4. Talk to few clients (call, support mails)
  5. Talk to biz dev folks and handle client queries through them
  6. One-on-one with one of my reportees (I do one every single day)
  7. Brainstorm on problem statements that my team is currently tackling
  8. Review quarter plan / OKR progress
  9. Write strategy docs giving direction to product lines
1 Like

The decision parameters are:

  1. OKRs which represent where we want to take our product and what’s our priority
  2. Impact of the feature, especially in terms of which OKR will it help push
  3. Effort to ship it out. We pick high effort tasks too, but we spread those over multiple sprints then

In my opinion OKRs in prioritized order give a lot of direction in this sense. The art aspect here is to make sure you don’t spend 100% on top OKR but distribute your efforts.

1 Like

Our blog post on PM hiring covers a lot of these in pretty good detail:

For example, would you, as a hiring manager, consider someone who has built their own product a viable candidate?

Definitely! A candidate coming in next week was running his own startup for a few years and we’d love to hire such folks!

1 Like

We are an extremely ambitious bunch of folks with finite time and bandwidth :slight_smile:

So how do you achieve the results in best possible way in the fastest possible timeline. These two rarely go hand-in-hand and as a PM this is your responsibility!

The other major challenge is in managing different kind of customers. We cater to largest of enterprises like Airtel to top startups like Ather Energy and to smallest of SME businesses. Each has different kind of requirements. How do you build a product that can cater to the entire gamut of client base. Most suggest you pick your persona and go deeper. This isn’t necessarily the best advice in payments space. Managing this becomes a huge challenge.

1 Like