In partnership with
Every other job posting on LinkedIn right now says "Growth Analyst." Startups, fintechs, edtechs, D2C brands,everyone wants one.
You look at the job description. It says things like "drive user growth," "optimize the funnel," and "improve retention."
Sounds exciting. It also sounds like it could mean absolutely anything.
Let me tell you what it actually means. Because I've watched analysts either thrive or completely flounder in these roles depending on whether they understood the job before they accepted it.
Here is the secret: Growth Analysts don't actually "grow" anything. They measure the business, find out where it's leaking, and plug the holes.
The Leaky Bucket
A company gets users. Some of those users become active. Some stick around. Some pay money. Some tell their friends.
That is a funnel. Your job is to measure every single stage of that funnel and figure out exactly where people are dropping off.
The framework most Growth teams use is called AARRR (often called "Pirate Metrics"). It stands for:
Acquisition: How do users find us?
Activation: Do they have a good first experience?
Retention: Do they come back?
Revenue: Do they pay us?
Referral: Do they tell others?
Your job is to track these, spot the drops, and work with Product and Marketing to fix them.
A Real Week in the Life
Monday. You check the weekly activation dashboard. Last week, 15,000 people signed up for your food delivery app. But only 5,400 completed the first key action (placing their first order). That's a 36% activation rate. Two months ago, it was 42%. Something is bleeding.
Tuesday. You dig into the activation funnel step-by-step using SQL:
WITH signup_cohort AS (
SELECT user_id, signup_date
FROM users
WHERE signup_date BETWEEN '2026-02-02' AND '2026-02-08'
),
funnel AS (
SELECT
COUNT(DISTINCT s.user_id) AS signed_up,
COUNT(DISTINCT CASE WHEN p.profile_completed = TRUE THEN s.user_id END) AS completed_profile,
COUNT(DISTINCT CASE WHEN a.first_search_date IS NOT NULL THEN s.user_id END) AS searched,
COUNT(DISTINCT CASE WHEN o.first_order_date IS NOT NULL THEN s.user_id END) AS first_order
FROM signup_cohort s
LEFT JOIN user_profiles p ON s.user_id = p.user_id
LEFT JOIN user_activity a ON s.user_id = a.user_id
LEFT JOIN orders o ON s.user_id = o.user_id
AND o.first_order_date <= s.signup_date + INTERVAL '7 days'
)
SELECT * FROM funnel;
The Result:
Signed up: 15,000
Completed profile: 11,200 (25% drop)
Searched for a restaurant: 8,900 (21% drop)
Placed first order: 5,400 (39% drop 🚨)
The biggest drop is between "searched" and "first order." People are looking at restaurants but refusing to buy. Why?
Wednesday. You form hypotheses. Are delivery fees too high? Is the checkout broken? You run a query to check if users who saw a "Free Delivery" promo converted better.
The Result: Users with the promo converted at 65%. Users without it converted at 57%. Insight: The promo helps, but it doesn't explain a massive 39% overall drop.
You pivot. You run another query looking at Session Time. Insight: Users who don't order are spending an average of 6.2 minutes on the search page. Users who do order spend 2.1 minutes.
Thursday. You present your findings to the Product Manager. Not a 10-slide deck. A simple Slack message:
"Activation is down 6 points WoW. The biggest leak is the Search-to-Order step. Free delivery promos only help marginally. The real issue is search friction—churned users are spending 6+ minutes scrolling without clicking a restaurant. They can't find what they want. Recommending we A/B test a personalized 'Top Picks for You' section at the top of the feed."
Friday. The Product team starts designing the experiment. You help define what "success" looks like, what metrics to track, and how large the sample size needs to be.
That is a Growth Analyst's week. Notice: No machine learning. No complex predictive modeling. Just funnel measurement, hypothesis testing, and clear communication.
LEARN AI By reading this best newseltter.
Tech moves fast, but you're still playing catch-up?
That's exactly why 100K+ engineers working at Google, Meta, and Apple read The Code twice a week.
Here's what you get:
Curated tech news that shapes your career - Filtered from thousands of sources so you know what's coming 6 months early.
Practical resources you can use immediately - Real tutorials and tools that solve actual engineering problems.
Research papers and insights decoded - We break down complex tech so you understand what matters.
All delivered twice a week in just 2 short emails.
The Metrics You Will Own
Walk into an interview knowing these, and you're ahead of the pack:
DAU/MAU Ratio: Daily Active Users divided by Monthly Active Users. If you have 30k MAU and 6k DAU, your ratio is 20%. This measures how "sticky" the product is.
Activation Rate: The percentage of signups who hit the "Aha! moment" (e.g., placing their first order, or adding 5 friends).
Retention Cohorts: Of the users who signed up in Week 1, how many came back in Week 4? If Week 4 retention is below 20%, you have a leaky bucket. No amount of marketing spend will save you.
LTV to CAC: How much revenue a user generates in their lifetime (LTV) vs. what it cost to acquire them (CAC). If LTV is ₹4,000 and CAC is ₹1,200, print money. If CAC is ₹5,000, you are going out of business.
The Interview Question They Will Ask You
"Our Week 1 retention dropped from 45% to 38%. Walk me through how you'd investigate."
Weak Answer: "I would build a churn prediction model using Python." (They didn't ask for a model. They asked for an investigation).
Strong Answer: "I would segment the data to isolate the drop. First, by acquisition channel; did we just onboard a batch of low-quality paid traffic? Then by platform (iOS vs Android) to rule out bugs. Finally, I'd look at behavioral data: What did the retained users do in their first 24 hours that the churned users didn't?"
One Thing to Do This Week
Pick any app you use daily (Swiggy, Spotify, Uber). Map out its funnel on a piece of paper. How did you discover it? What was your first action? When did you first pay?
Now, think about exactly where you would have dropped off if the experience was slightly worse. That is product thinking. That is what Growth Analysts do.
Next week: Selection Frameworks: How Analysts help companies make million-dollar decisions.
— RAJ
Real Analyst Jobs
P.S. Added new 25+ fresh remote jobs: realanalystjobs.com
Added new free projects : https://realanalystjobs.com/projects
Document your 2026 Journey: https://realanalystjobs.com/journey
Talk to me : https://realanalystjobs.com/raj