Skip to content
Academy
Marketing Academy · Field Work●Human Psychology
All projects·From: The Peak-End Rule
CoreSimulation· 50 minutes

The CX Redesign Budget: Fix the Middle, the Peak, or the End?

Swiggy

Objective: Given a limited quarterly CX budget and a dashboard of ticket volume vs. NPS drivers, make three sequential resource-allocation decisions that either correctly apply duration neglect and the peak-end rule or waste the budget chasing ticket volume instead of remembered experience.

You're the CX lead at Swiggy with one quarter's budget to improve the delivery experience. Support ticket volume says one thing; NPS driver analysis says another. You can't fix everything this quarter.

Across three checkpoints, decide where the team's limited sprint capacity goes, and see how each choice plays out over the quarter.

Before you start

What you'll need

Free path (everything below is enough to finish)

Google Sheets
FreeModel the budget allocation and track NPS/ticket-volume deltas across the three checkpoints

Free, sufficient for a single-quarter budget-tracking exercise

Notion
FreemiumDocument the rationale behind each budget decision as a lightweight decision log

Free, keeps the 'why' attached to each choice for the quarterly review

Paid upgrades (optional, faster/deeper)

Mixpanel(optional)
FreemiumTrack real cohort-level NPS and repeat-order rate changes tied to each shipped fix

Gives statistically real before/after cohort comparisons instead of aggregate quarterly deltas

No access? Google Sheets with manually pulled weekly NPS/ticket exports covers a single-quarter pilot

The process

Simulation

Week 1: Diagnosing the journey

Week 1

Support ticket volume is dominated by 'my order was late' complaints (61% of tickets). But a separate NPS driver analysis shows the rider hand-off moment, not average delivery time, correlates most strongly with which customers churn after a bad experience.

Sample output
Ticket volume: late delivery 61%, wrong item 22%, rider hand-off complaints 9%, other 8%.
NPS driver regression: rider hand-off quality explains 3.2x more variance in churn than average delivery time.
Spend to date:₹0 of ₹40L quarterly CX budget
Budget remaining:₹40L

Where does the team's first sprint go?

Final deliverable

A three-checkpoint decision log showing where the quarterly CX budget was allocated, the outcome of each choice, and a final recommendation on which metric (NPS/repeat-rate vs. ticket volume) should drive next quarter's CX budget.

See a reference example
Sample output
Zomato CX budget simulation, decision log (excerpt)

Week 1: Funded the rider hand-off redesign over the top ticket-volume complaint
  Result: NPS driver improved, ticket volume on hand-off complaints dropped
Week 4: Funded a personalized delivery-complete notification with an auto-credit for late orders
  Result: Post-delivery survey response rate rose, comments shifted positive even on late orders
Week 8: Recommended scaling both fixes company-wide, reporting NPS/repeat-rate as the primary CX metric
  Result: Board approved scaling; ticket volume reframed as an operational, not a churn, metric

Success criteria

You're done when you can:

  • All three decision points are resolved with an option selected and its outcome recorded
  • The final recommendation correctly identifies NPS/repeat-rate, not raw ticket volume, as the metric that tracked the fixes