Back in
business.
Overdue invoices were also a growth problem. Payment follow-up helped partners resolve issues and get properties moving again.
Account Executive Intern · Bengaluru
GROWTH · SYSTEMS · AIBENGALURU, IN
AN INTERACTIVE STUDY
ORDER FROM COMPLEXITY
ORDER FROM COMPLEXITY
A PORTFOLIO BY SHAHEER RAHMAN / VOL. 02
TURN SIGNALinto↗MOMENTUM.
I find the friction between
a good idea and a useful result.
Then I build a way through.
THE WORK↙
ECONOMICS MIND.
BUILDER’S INSTINCT.SCROLL TO EXPLORE
BUILDER’S INSTINCT.SCROLL TO EXPLORE
01 / A TRACK RECORD IN THE MAKING
Less friction.
More forward.
Partner conversations. Messy data. Repetitive work.
A few places I’ve helped turn a bottleneck into a next step.
Showing all 3 field notes.
Back in
Give the team
their time back.
A property-data deduplication workflow that turned an estimated 100-hour manual task into 1–2 hours. Plus a self-service knowledge agent for recurring questions.
Context is
a growth tool.
Prospect mapping, structured discovery notes, and clear next-step ownership. Small systems that made a handoff easier to act on.
Business Development Intern · Pune
400+property reactivations supported
1,000+accounts in the prioritisation pool
420partner-performance calls
Booking.com internship results, as reported in my résumé. Open a case for the context and scope.
02 / THE INTERACTIVE GROWTH LAB BUILT TO BE USED
Change an input.
Challenge a bet.
More traffic isn’t always the answer.
Move a slider. Find the leak.
Take the useful part with you.
Make a better bet.
What if activation improved before you spent more on traffic?
10020,000
THE SAME TRAFFIC. A DIFFERENT OUTCOME.SCENARIO / NOT LIVE DATA
30→42
expected first successful actions
Cost / first success
₹500 ₹357
Change in first successes+40%
Improving activation from 25% to 35% adds 12 expected first successes at the same spend.
Expected values, rounded for display.
Illustrative inputs only. This models an assumption; it does not prove a lift. First success means a defined useful action, such as a successful API call. It is not necessarily a paying customer.
A LITTLE LESS “WHERE DID THAT COME FROM?”Give every link
Give every link
a paper trail.
Build a campaign URL with consistent tags. Values are trimmed and spaces become underscores.
This adds tracking parameters to a URL. It does not send data or connect to an analytics account.
03 / HOW I THINKA LOOP, NOT A STRAIGHT LINE.
From noise
to next.
Useful work starts with a good question.
Then it needs somewhere to go.
Start where
it gets stuck.
Partner conversations and recurring team questions are useful signals. Understand the friction before choosing the fix.
Build the smallest
useful version.
A workflow, a landing page, a better handoff. Make the idea concrete enough for someone to try.
Earn the claim
with evidence.
Define the action that matters. Separate activity from outcomes, and a useful result from an encouraging click.
Make the next
bet better.
Keep what works. Question what doesn’t. A result should change the next decision, even when the idea loses.
04 / A FIRST BET FOR SARVAMPROPOSED · NOT YET RUN
Don’t sell the API.
Show the possibility.
A 30-day activation experiment I’d bring to the team.
Concrete enough to ship. Measurable enough to question.
THE HYPOTHESISShow the use case.
Show the use case.
Shorten the distance
to “I made that.”
A builder arriving from a concrete voice-note demo may be more likely to make a first successful API call than someone arriving from a general feature overview.
A hypothesis to validate, not a claim about Sarvam’s current funnel.
PRIMARY MEASUREVisitor → first successful API call
within 24 hours · eligible new visitors
within 24 hours · eligible new visitors
DIAGNOSTICS
Signup conversion · key creation · time to first success · repeat usage within 7 days
Thirty days. Four honest steps.
SELECT A WEEK TO EXPLORE ↘Get the baseline before the big idea.
Review the existing onboarding path with the team, listen to developer questions, and agree on the activation event. Confirm that source, variant, signup, and server-confirmed API success can be linked with approved instrumentation.
DELIVERABLE / A funnel map, baseline rates, and an agreed measurement plan.A PARAGRAPH FOR THE APPLICATION
In my first 30 days at Sarvam, I’d test whether a use-case-first voice demo helps new developers reach their first successful API call. I’d start by validating the activation event and baseline, then compare a concrete voice-note-to-text entry point against a general API overview using randomly assigned traffic. The primary measure would be first successful API calls within 24 hours per eligible visitor, with signup conversion, time to first success, errors, and seven-day repeat usage as diagnostics. I’d ship a small test in week two, work within an agreed sample and budget, and use the final two weeks to decide what to improve, stop, or extend. The funnel simulator and campaign URL builder in this portfolio are small examples of how I approach that work.
05 / THE PERSON BEHIND THE SYSTEMSASSAM → PUNE → BENGALURU
SHAHEER RAHMANBENGALURU, INDIA ↗
Economics trained
my questions.
Building tests them.
I’m an economics postgraduate and a Marketing PGDM candidate at IMDR, Pune. My work has taken me from hotel partner conversations at Booking.com to prospect research at NoBrokerHood and product demos at Mayasabha XR.
The thread running through it: understand the person, organise the information, and make the next action easier. AI helps me build faster. Asking better questions helps me build something useful.
Download my résuméEDUCATION
PGDM · Marketing
IMDR, Pune · Expected 2027
MA · Economics
IGNOU · 2022–2024
BA · Economics
Dibrugarh University · 2022
IN MY TOOLKIT
MS Excel & MS Office
CRM documentation
Booking.com Extranet
AI-assisted workflows
English · Hindi
The journey so far ↘
JUL 2026 — PRESENT
Booking.comAccount Executive Intern · Bengaluru
MAY — JUN 2026
NoBrokerHoodBusiness Development Intern · Pune
OCT 2025 — JAN 2026
Mayasabha XRMarketing Intern · Pune20+ demos/onboarding conversations · 50+ prospect touchpoints · 10+ restaurant & technical stakeholders
06 / THE NEXT GOOD QUESTIONLET’S BUILD SOMETHING USEFUL.