Ayushi Gupta, founder of Traxn

Hello, I'm Ayushi.
Most marketing budgets answer the wrong question.

I help B2B SaaS and AI companies find the right one — then build the positioning and go-to-market plan that follows from it.

Book a first call Taking projects this quarter

Founder-led, no handoffsEvidence before recommendationsIndia · US · UK · GCC

The meeting

It's the quarterly review. Someone pulls up the dashboard. Traffic is up, leads are flat, and pipeline is the number nobody wants to say out loud.

So the room does what rooms do. It debates channels. Someone suggests more content. Someone else wants to try events again. Budget moves from one column to another, and everyone leaves with something to do.

Nobody says the quiet part: five people in that room would describe the ideal customer five different ways, and not one of them has spoken to a buyer this quarter.

The argument was never about channels. It was about a decision nobody had made.

What indecision costs

Serving three segments doesn't divide your results by three.

Winning a deal needs four things true at once: your message fits how they talk, you have proof from a lookalike company, their objection is pre-answered, and you're credible live. Those four multiply — they don't add.

Focused on one segment 41% Four factors at 0.8 each
Split across three 6% The same four at 0.5 each

Three segments at 6% each is 18% of a market's worth of effort, against 41% from one. You didn't lose a third — you lost most of it. Figures are illustrative; the multiplicative shape is the point.

The turn

Better questions first. Everything else is downstream.

"How do we get more leads?" has no answer. "Who is choosing us today, and why?" has one — and once you have it, the channel argument settles itself, because you finally know who you're talking to and what they're comparing you against.

Better questions firstClarity over complexityEvidence over assumptions

Services

Four ways in, one line of thinking

Each starts with customer evidence and ends with decisions your team can act on. Open any one for scope, duration and what you walk away with.

Positioning

Brand and product positioning

Decide who you serve, what you replace and why you win — then say it in the words your buyers actually use.

Customer understandingDifferentiationMessaging
What's included
Duration
Two to three weeks
Your time
Two working sessions, six to eight hours total
You get
A positioning statement, the evidence behind it, and messaging your team can use the same week
Fee
Fixed, quoted before work starts

Start here if your homepage leaves visitors unsure what you do, or sales, product and marketing each describe the product differently.

Go-to-market

Go-to-market strategy

Choose the markets worth winning, plan the launch, and get sales and marketing telling one story.

Market selectionLaunch planningSales alignment
What's included
Duration
Four to six weeks for a full plan; two weeks for a market-entry sprint
You get
A market and segment choice with reasoning, a launch plan with owners and dates, and the rejected options written down
Fee
Fixed, quoted before work starts

Start here if the pitch works at home but doesn't land abroad, or you're chasing three segments and winning none.

Growth

Demand generation and growth

Build demand programmes that match how your buyers research, with measurement honest enough to tell you what to stop.

Demand strategyConversionMeasurement
What's included
Duration
A three-week strategy phase, then an optional monthly programme
Commitment
Monthly programmes have a three-month minimum
You get
A demand plan, a measurement model your team can run without a data stack, and one decision per monthly review

Start here if there's plenty of activity but pipeline isn't moving, and budget reviews keep ending in an argument.

AI search

AI search and content visibility

Find out how assistants describe your category and who they name, then earn a place in those answers.

Search researchExpert contentVisibility
What's included
Duration
A two-week snapshot and plan, then quarterly re-measurement
You get
Your current visibility with the full prompt set, what's driving it, and the content work that would change it
Fee
Fixed, quoted before work starts

Start here if AI assistants recommend your competitors, describe you wrongly, or don't mention you at all.

The method

Understand. Define. Build. Learn.

Every engagement moves through the same four stages. The order matters — skip one and the ones after it get weaker.

Stage 1

Understand

Calls, reviews, lost deals, competitor messaging, search and AI answers.

You get: what buyers actually say, and why they choose you.

Stage 2

Define

Name the real problem and rank the opportunities by worth and difficulty.

You get: one problem statement everyone agrees with.

Stage 3

Build

Create the strategy and a roadmap with owners, dates and the first things to ship.

You get: a plan your team can start on Monday.

Stage 4

Learn

Measure what moved, cut what didn't, refine the next decision with better data.

You get: a review that ends in one clear next step.

Case studies

How the thinking plays out

Traxn is new, so these are worked analyses rather than client engagements — built from public information and in-house experience, and labelled as such. Where a figure appears it is modelled, with the assumptions shown.

Case study / 001

The company that was winning deals and losing the market

A 34% win rate looked like proof the product was strong. It was the symptom: they were winning the deals they got shortlisted for, and almost never getting shortlisted. A narrow funnel was flattering the average.

They described themselves as a "people operations platform". Every buyer interviewed said "payroll software". Repositioning into the category buyers actually searched for would roughly double qualified opportunities at flat spend — modelled, holding win rate and budget constant.

The lesson: a high win rate is one of the most dangerous numbers in B2B marketing. Always read it against the volume of opportunities that produced it.

Discipline

  • Positioning
  • Category language
  • Full engagement

Sector

HR tech · India

Results

Modelled, assumptions shown

Illustrative · not client work

Case study / 002

Invisible to the assistant, and nobody noticed for two quarters

Rankings were stable, traffic was stable, demo requests were quietly falling. Sales began hearing "we had a shortlist already" — naming two competitors the team considered weaker.

Twenty buyer-phrased prompts across three assistants showed why. The competitors didn't rank better; they were simply described more often in the sources assistants lean on — comparison pages, review summaries, forum threads.

The lesson: you can hold every search ranking you ever had and still vanish from the shortlist. Anyone promising a guaranteed position in an AI answer is selling something that cannot be bought.

Discipline

  • AI visibility
  • Prompt measurement
  • Content strategy

Sector

Support SaaS · India & UK

Method

Prompts published

Illustrative · not client work

Case study / 003

The demo that wowed, and the deal that died in procurement

Every first call went well. Prospects called it "magic". Then the second call never got booked — or it got booked with someone asking where the data goes, what happens when it's wrong, and who signs off on it.

The company read this as a sales problem and hired two more AEs. But the enthusiasm and the silence came from two different people: the demo converted an individual, while the purchase required a committee assessing risk.

The lesson: for AI products, capability wins the first meeting. Accountability wins the contract.

Discipline

  • Positioning
  • AI products
  • Buying committees

Sector

AI · Series A

Format

Worked example

Illustrative · not client work

In their words

What clients will say here

The layout is built and filled with sample quotes and invented client details, each one badged, so you can see exactly how the finished section reads. Real quotes replace them as engagements finish and clients approve the wording in writing.

Sample
We came in wanting a campaign plan and left having finally agreed who we sell to. The second one turned out to be the thing we'd been missing for two years.
Rhea NambiarCo-founder & CEO · Payloom (HR tech, Bengaluru)
Sample
The research came back saying something we didn't want to hear, with the receipts attached. That's the part I'd pay for again.
Daniel OseiVP Marketing · Arcwell (support SaaS, London)
Sample
One page, three decisions, no deck to decode. Our new hire read it on day one and knew what we stood for.
Meera RaghunathanHead of Growth · Sundeck AI (Series A, Pune)

These three are samples — the people and companies are invented. They are here so you can see the finished layout with realistic names in place, which is why each carries a Sample badge. Leave the badges on until real quotes replace them: publishing invented endorsements as genuine is misleading advertising, and one prospect asking "can I speak to them?" would undo the credibility of everything else on this page. When a client approves wording, swap the name, role and company and delete the badge.

What you can check

Judge the work, not the adjectives

Until client proof exists, here is what you can read, attend or re-run yourself before you spend anything.

Published thinking

Human in the Loop

A newsletter on AI and the future of work, written by the person who would run your engagement. The clearest sample of how she reasons.

In the room

Speaking on go-to-market

Including eChai Ventures' GTM and Distribution for Startups in Bengaluru. Come and ask hard questions before committing to anything.

Checkable research

The AI answer study

We publish the prompts alongside the findings, so you can run them yourself and see whether we reported honestly.

Ayushi Gupta speaking at a go-to-market event in Bengaluru

Who you'd be working with

I've sat in that meeting more times than I can count.

My career has been spent in product marketing and go-to-market inside B2B SaaS and AI companies — positioning, launches, demand generation and growth — and in the particular problem of taking Indian products into the US, UK and Gulf markets, where the product stays the same and almost everything about the story has to change.

That work left me with one conviction: the expensive mistakes in marketing are rarely execution mistakes. They are decisions nobody made, postponed until a campaign calendar quietly makes them by default.

I lead every engagement myself and join every working session. No account managers, no handoffs — the person who does the research is the person who presents the recommendation, and the person still there when you want to argue with it.

Based in
Bengaluru, India
Works across
India, US, UK and GCC
Focus
B2B SaaS and AI, seed to Series B
Writes
Human in the Loop

FAQ

A few things, answered

What does an engagement cost?
Fixed price per engagement, agreed before anything starts. After the first call you get a single page: the problem as we understand it, the scope, the deliverables, your time commitment and the price. Indian companies are invoiced in INR plus GST; companies elsewhere in USD.
How much of my team's time will this take?
We tell you upfront and then stick to it. A positioning engagement is typically two working sessions and six to eight hours of your team's time in total. If the scope needs to change midway, we re-agree it in writing before continuing.
Why are there no named client case studies yet?
Because Traxn is new, and we won't publish a client's name, numbers or words until they have approved them in writing. The analyses here are labelled illustrative for the same reason. You can judge the thinking without us borrowing someone else's results.
Do you do the execution too?
We do the strategic work and the first layer of execution that proves it. If you need a full execution agency for ads, design and video, or a brand identity built from scratch, we'll say so on the first call and point you to people who do that well.
What if we're not a good fit?
We'll tell you, early and plainly. If you don't need us, or need something we don't do, that's a thirty-minute call rather than a wasted quarter. We'd rather lose a bad-fit project on the first call than three weeks into it.

Start a conversation

What's the question you keep circling back to?

Tell me where you are, what you're trying to achieve and what's in the way. Thirty minutes is usually enough to know whether I can help — and I'll say so plainly if I can't.

Message me on LinkedIn ↗ See the services again

Bengaluru, India · Email on request via LinkedIn