What Separates a 7–12 LPA Candidate from Everyone Else

7–12 LPA ANSWER — FIRST 90 SECONDS:

Interviewer: "Tell me about a project that
mattered."

Candidate: "Sure. I owned the sales
reporting pipeline, and the thing that
mattered was the decision it changed.

The problem: the sales team reviewed
performance monthly, using a report that
took 6 hours to run. By the time they saw
the number, the month was over — they
couldn't act on it.

What I did: I rebuilt the pipeline — a
star schema in Postgres, incremental loads,
and a Redis cache — so the report runs in 20
minutes. The sales team moved from monthly
to weekly reviews, and in the first quarter
they caught two regions that were slipping
before the quarter closed.

The trade-off: the star schema adds ETL
work — about 2 days to build, and 20 minutes
of maintenance a month. I chose that over
querying the raw tables because the raw query
took 40 seconds at 5M rows, and the
dashboard was unreadable.

The ownership: I did the data model, the
queries, the refresh schedule, and the
stakeholder training. I also wrote the
reconciliation check that compares the
report to the finance total every morning.

I can show you the model and the
reconciliation."

What the interviewer thinks in 90 seconds:
- "They named a decision that changed." ✅
- "There's a before/after: 6 hours → 20
  minutes." ✅
- "There's a business impact: monthly →
  weekly reviews." ✅
- "There's a trade-off with a cost." ✅
- "They owned it end-to-end." ✅

Signal: 7–12 LPA. The rest of the interview
confirms it.

BELOW 7 LPA ANSWER — FIRST 90 SECONDS:

Interviewer: "Tell me about a project that
mattered."

Candidate: "I worked on a sales report. I
wrote SQL queries and built a dashboard in
Tableau. I used joins and aggregations. The
report showed sales by region and by month.
My manager said it was good."

What the interviewer thinks in 90 seconds:
- "A task, not a decision." ❌
- "No before/after — how long did it take
  before?" ❌
- "No impact — did anyone use it
  differently?" ❌
- "No trade-off — why this approach?" ❌
- "They built it; they didn't own it." ❌

What happens next:
- The interviewer asks "What decision did
  the report change?"
- The candidate says "I'm not sure — the
  sales team used it."
- The interviewer asks "How do you know
  they used it?"
- The candidate pauses.

Signal: below 7 LPA. The candidate can do
the task, but hasn't shown the impact.

Why it failed:
❌ Task, not decision
❌ No before/after
❌ No impact metric
❌ No trade-off
❌ The first 90 seconds signaled "I
  completed a ticket," not "I changed a
  decision"

IMPACT ANSWER — WHAT DECIDES:

Interviewer: "How do you know it mattered?"

WHAT THE CANDIDATE SAYS:
"Three pieces of evidence, and each is
specific.

Evidence one: the time. The report went from
6 hours to 20 minutes. That's measured, not
estimated — I timed the old job and the new
one.

Evidence two: the cadence. The sales team
moved from monthly to weekly reviews. That's
a behavior change — they wouldn't do that if
the report still took 6 hours.

Evidence three: the outcome. In the first
quarter after launch, the team caught two
regions that were slipping before the quarter
closed. I can't prove the report caused the
catch, but the VP of Sales named it in the
QBR.

The limit: I don't have a clean before/after
on revenue. The report made the information
faster, but revenue depends on many things.
What I can prove is the time and the cadence.

The trade-off: the star schema costs 2 days
to build and 20 minutes a month to maintain.
I chose it because the raw query took 40
seconds and the dashboard was unreadable at
that speed."

What the interviewer thinks:
- "Three pieces of evidence, each
  specific." ✅
- "Time and cadence are measured, not
  estimated." ✅
- "The outcome is attributed honestly." ✅
- "The limit is named." ✅
- "The trade-off is named with its cost." ✅

Signal: confirmed 7–12 LPA.

Why it decides:
✅ Evidence is specific
✅ Time is measured
✅ Behavior change is evidence
✅ Attribution is honest
✅ Trade-off is named

OWNERSHIP ANSWER — WHAT DECIDES:

Interviewer: "What did you own?"

WHAT THE CANDIDATE SAYS:
"End-to-end, with one exception I'll name.

What I owned: the data model (a star schema
with region and month dimensions), the
incremental load (so the daily refresh takes
90 seconds, not 20 minutes), the queries,
the refresh schedule, and the reconciliation
check.

What I also owned: the stakeholder training.
I ran two sessions with the sales team and
wrote a one-page guide. That's part of the
job — a dashboard nobody uses is a failed
project.

What I didn't own: the source data. The CRM
export is owned by the sales ops team, and I
had to file a ticket when a field changed.
That's a boundary I'd want to change in the
next role — I'd push for read access to the
CRM, so I can debug source issues without a
ticket.

The trade-off: owning the training added a
week to the project. I chose it because the
alternative was shipping a dashboard that sat
unused — and that's not a win.

The result: 30 sales reps use it weekly, and
the VP named it in the QBR."

What the interviewer thinks:
- "They named what they owned." ✅
- "They named what they didn't own." ✅
- "The boundary is a real one, with a
  ticket." ✅
- "They owned the adoption, not just the
  code." ✅
- "The trade-off is named with its cost." ✅

Signal: confirmed 7–12 LPA.

Why it decides:
✅ Ownership is specific
✅ Boundaries are named
✅ The boundary is a real constraint
✅ Adoption is owned
✅ Trade-off is named

DEBUGGING STORY — 7-12 LPA VS. BELOW:

Scenario: "Tell me about a time you found a
problem before a stakeholder did."

BELOW 7 LPA SIGNAL:
"One time the numbers looked off, so I
re-ran the query and they were right."
What the interviewer thinks:
- "No root cause."
- "No prevention."
- "They waited for the number to look off."
- Signal: below 7 LPA.

7-12 LPA SIGNAL:
"I built a reconciliation check that
compares the report total to the finance
total daily. It caught a 4% divergence one
morning before anyone saw the report.

The root cause: a product rename created a
duplicate row in the product dimension. The
join on product key duplicated revenue for
every renamed product — about 2,000 rows.

The fix: I deduplicated the dimension with a
surrogate key and re-ran the affected
months. The revenue was overstated by $40K
across two months.

The prevention: the reconciliation check
was already there — that's how I caught it.
I added a uniqueness test on the dimension
key, so the pipeline fails before the join
runs, not after.

The trade-off: the uniqueness test adds 15
seconds to the pipeline. I chose that over
faster runs because a $40K misstatement is
not worth saving 15 seconds.

The lesson: I now test every join key for
uniqueness before I trust it. A group-by
count takes 30 seconds and would have caught
this before it shipped."

What the interviewer thinks:
- "They built the check that caught it." ✅
- "The root cause is a real duplicate." ✅
- "The fix is specific." ✅
- "The prevention targets the cause." ✅
- "The trade-off is quantified." ✅
- Signal: confirmed 7–12 LPA.

Why it decides:
✅ Prevention existed before the bug
✅ Root cause is specific
✅ Fix is specific
✅ Prevention targets the cause
✅ Trade-off is quantified

PORTFOLIO REVIEW — 7-12 LPA VS. BELOW:

Scenario: "Review this portfolio. The
candidate lists one project."

BELOW 7 LPA PORTFOLIO:
- A Tableau screenshot
- No README
- No decisions
- No impact
- The title is "Sales Dashboard"

What the interviewer thinks:
- "This could be a tutorial copy."
- "No decisions, no impact."
- "I can't tell if they built it or
  screenshotted it."
- Signal: below 7 LPA.

7-12 LPA PORTFOLIO:
- A live dashboard link, not just a
  screenshot
- A README with: the decision it drives
  (weekly sales reviews), the impact
  (6h → 20min; monthly → weekly), the data
  model (star schema with region and month
  dimensions), the trade-off (ETL cost vs.
  query speed), and a reconciliation check
  that compares the total to finance
- A "what I'd do differently" note: "I'd
  add row-level security so each regional
  lead sees only their region; I left it out
  because it needs a user table I didn't
  have."
- A "how I measured it" section: the timing
  method, the reconciliation threshold, and
  the cadence change

What the interviewer thinks:
- "There's a live artifact, not just a
  picture." ✅
- "The README names the decision and the
  impact." ✅
- "The data model is shown." ✅
- "The trade-off is named." ✅
- "There's a reconciliation check." ✅
- Signal: confirmed 7–12 LPA.

Why it decides:
✅ Live artifact, not a screenshot
✅ README with decision and impact
✅ Data model visible
✅ Trade-off named
✅ Reconciliation check included

7–12 LPA Checklist:

FIRST 90 SECONDS:
☐ Name a decision that changed, not a
  task completed
☐ Name an impact with a number:
  6h → 20min, 31% → 42%
☐ Name a trade-off with a cost:
  ETL time, 20 seconds, a week of training
☐ Avoid "I built a dashboard" without
  what changed

WHAT DECIDES:
☐ Show end-to-end ownership: model,
  queries, refresh, training
☐ Name a failure and what changed
☐ Name a prevention, not just a fix
☐ Name the limit: what you can't
  attribute
☐ Name the next step with more time

WHAT CONFIRMS:
☐ A live artifact, not a screenshot
☐ A reconciliation or sanity check
☐ A measured before/after, not an
  estimate
☐ A behavior change: monthly → weekly,
  random → ranked
☐ A stakeholder who would vouch for it

RED FLAGS:
❌ "I built a dashboard"
❌ No before/after
❌ No impact metric
❌ No trade-offs
❌ No failure
❌ No prevention
❌ "My manager said it was good"

Why this works:
- The first 90 seconds set the signal
- The rest of the interview confirms or
  overturns it
- Impact with a number, judgment in the
  trade-off, and end-to-end ownership are
  the three things that can't be faked

30-DAY PRACTICE PLAN (FOR CANDIDATES):

Week 1 — Find the decision
☐ Pick 3 projects you've built
☐ For each, write: what decision did it
  change?
☐ Write the before/after: time, cadence,
  errors caught
☐ Record yourself explaining each in
  90 seconds
☐ Cut every "I built" sentence that
  doesn't name a change

Week 2 — Add the trade-off
☐ For each project, write one technical
  decision and its cost
☐ Write the alternative you rejected and
  why
☐ Write the limit: what can't be
  attributed
☐ Re-record the explanation with the
  trade-off included

Week 3 — Ownership and prevention
☐ For each project, list what you owned
  end-to-end
☐ Name one failure and what changed
☐ Name one prevention: a check, a test,
  a reconciliation
☐ Add these to your portfolio README

Week 4 — Mock interviews
☐ Run 2 mock interviews with a peer
☐ Ask them to challenge: "What decision
  changed?" and "How do you know?"
☐ Re-run with stronger impact and
  evidence
☐ Compare the two recordings

Signs it's working:
✅ You name a decision, not a task
✅ You have a before/after number
✅ You name a trade-off with its cost
✅ You name a prevention
✅ You name the limit honestly
✅ The interviewer asks "what else?"
  instead of "what did you build?"