Nagesh Koritala, product manager

Product judgment. Engineering depth. AI-native execution.

I’m a product manager at Kohler, leading work across the conversion funnel. I turn customer insight into clear priorities and align teams to deliver.

Nagesh Koritala in a checked blazer over a black sweater, smiling, beside a cut-paper collage of the product work he does: a chat, a payment form, finish swatches with one circled in red, and charts.

Meet Nagesh

A product manager who started as an engineer.

I’ve worked in digital for 14+ years, and I started as a frontend developer on the team that built the store before it launched.

I find where customers get stuck deciding, then get design, engineering, and the business looking at the same evidence. And I still build: working prototypes, AI evals, anything that turns a long debate into a short question.

Read my full story 

Watercolor of Nagesh in a denim shirt at a screen showing a customer journey with one step circled in red, pointing to it while a data analyst in a wheelchair, a content designer in a hijab, and an engineer look at the same spot

Selected work

Complete your install

For this product

  • Faucet$289.00 (included)

We recommend

  • Drain assembly$34.99 (not included)
  • Supply lines$18.50 (not included)
  • Installation kit$12.75 (not included)
  • Faucet cleaner$9.99 (not included)

Selected (1)$289.00

Add to cart (1)

Illustrative concept

The easy path, set aside:More faucets like this one

Our call:Everything it takes to install this one

LiveRecommendations

Helping customers finish the job

The shopper’s question“What else do I need to install it?”

Once a shopper picks a faucet, more faucets don’t help. The parts to install it do.

I noticed customers coming back within days of a big purchase to buy the parts they needed to install it. I worked with merchandising to show those parts on the product page.

How we knew it was workingAverage order value and units per order both went up from these recommendations.

See my work (access code)  access code

Nagesh Koritala in a charcoal sweater, reading a tablet, beside a cut-paper collage of a shopper’s chat and a grid of product facts, with a gap circled in red.

The easy path, set aside:Patch weak answers one by one

Our call:Fix weak answers at the source

LiveAI shopping agent

Earning trust for an AI shopping agent

The shopper’s question“Is this the right one for me?”

Shoppers need answers they can trust, so we agreed on what a good answer looks like and keep checking the agent against it.

My leadership team brought the idea. I’ve been its product manager from the first day through launch, and I still am. I wrote the guardrails, and I keep improving it from what the evals and customers tell us.

How we knew it was workingOur evals showed answer quality improving because we fixed the underlying data gaps.

Read the case study: Earning trust for an AI shopping agent (access code)  access code

Nagesh Koritala in a navy blazer over a white T-shirt, beside a cut-paper collage of payment cards whose arrows meet at one checkout form, circled in red.

The easy path, set aside:Add more payment methods first

Our call:Fix failing payments, then add choice.

LiveCheckout and payments

Fixing failed payments at checkout

The shopper’s question“Why did my payment fail?”

A payment that keeps failing breaks trust.

When a payment fails at checkout, it hurts both sides: the customer can’t finish buying, and the business loses the sale at the last step. I brought the problem statement to the team. Together we concluded that failing payments had to be fixed before we added new ways to pay, and leadership and stakeholders backed that order. We fixed the systemic errors in our checkout APIs and payment gateway, and rewrote the error messages so customers can fix the problem themselves.

How we knew it was workingFewer payment failures at checkout.

Read the case study: Fixing failed payments at checkout (access code)  access code

Full case studies are shared by access code. See all case studies 

A short film

Customers rarely get stuck on a feature. They get stuck on a decision.

0:00
Original music, written for this film.
Read the transcript
  1. When we can build more, what is worth building?
  2. AI has increased our ability to build faster than our confidence about what deserves to be built.
  3. So we don’t start with a feature list. We start where customers get stuck.
  4. Not the newest idea. Not the loudest request in the room.
  5. One problem worth solving.
  6. We test a few ways to solve it, and watch what customers do.
  7. The goal isn’t building more things.
  8. The right thing.
  9. When an idea falls flat with customers, we let it go.
  10. We keep looking until one really works.
  11. The next shopper doesn’t stop to ask.
  12. They buy what they came for.
  13. So what’s worth building? The thing that solves the customer’s problem.
  14. Worth building.

How I run product discovery

What I do before we build.

You can’t decide what’s worth building from a conference room. So I go find out. I talk to customers, put rough prototypes in front of them, and watch what they do. It keeps the customer first, and it keeps all of us honest about what we actually know.

Three overlapping circles. Desirable, marked start here: Do people want it? Feasible: Can we build it? Viable: Does it work for the business? The goal is where all three meet: worth building. $ DesirableFeasibleViable Start here Do people want it? Can we build it? Does it work forthe business? Goal Worth building Three overlapping circles. Desirable, marked start here: Do people want it? Feasible: Can we build it? Viable: Does it work for the business? The goal is where all three meet: worth building. $ DesirableFeasibleViable Start hereDo people want it? Can we build it? Does it work forthe business? Goal Worth building

Every idea has to get past three lenses.

I start with people. Then I check that we can build it and that it works for the business. The ideas worth building sit where all three meet.

  1. Desirable Start here

    Do people want it?

    First, the customer: what they’re trying to decide, and what’s in their way.

  2. Feasible

    Can we build it?

    This is where my engineering background helps. I check the data, the systems, and the integrations early, often with a working prototype.

  3. Viable

    Does it work for the business?

    The company has to be able to sustain it and make money from it.

The work is a loop, not a line.

Five steps, but rarely in a straight line. What we learn while testing often sends us back to the start. That’s not the process failing. That’s it working.

Empathize · User research, interviews

Listen first.

I listen for what people are trying to do, in their own words, and where it goes wrong for them.

At work, this meansSorting logged shopper questions to see where the AI agent lets people down.

Nagesh walking in with a coffee.
their words, not mine

Frameworks adapted from the Interaction Design Foundation (interaction-design.org).

See how I work

Learn with me

I teach what I practice.

Free and interactive, for product managers at any stage. Every lesson links its sources.

See everything I teach 

Product basics as one picture: three parts. The job, lessons 1 to 8; Advanced, lessons 9 to 11; Interview prep, lessons 12 to 15. Arrows lead from one part to the next. Lesson 1 is marked in red: start here. The job12345678Advanced91011Interview prep12131415 start at 1, then go at your own pace
The Product basics course map.

Published writing

What I’m learning.

I write about product judgment, customer decisions, and commerce as AI changes it.

Read all writing

Say hello

Deciding what’s
worth building?

So am I, every week. I’d love to hear what’s on your list, and what you’re leaving off.

Nagesh Koritala in an olive overshirt, sitting on the edge of a desk with a coffee mug, smiling with an open hand as if mid-conversation, beside a cut-paper collage of two armchairs, a small table with mugs, and a speech bubble.