Skip to content
Academy
Marketing Academy · Field Work●Conversion Rate Optimization
All projects·From: CRO 101
CoreBuild the Asset· 45 minutes

Build the Roadmap: Scoring and Sequencing Six Test Ideas

Nykaa (FSN E-Commerce Ventures)

Objective: Given six raw test ideas pitched by different teams, score each on impact and effort, plot them against the lesson's priority matrix, and write one fully formatted hypothesis for the top pick.

You're the CRO lead at Nykaa. Five teams have each pitched one test idea for next quarter, and the product team added a sixth. You have budget for 3.

Score every idea, plot it, and only then write a hypothesis, not the other way around.

Before you start

What you'll need

Free path (everything below is enough to finish)

Google Sheets
FreeScore, plot, and rank the 6 test ideas

Free, sortable, easy to share with stakeholders

Paid upgrades (optional, faster/deeper)

Optimizely(optional)
PaidLaunch the top-ranked test once the roadmap is approved

Enterprise-grade experimentation platform for running the resulting tests

The process

2 steps

Step 01 of 02

Testing Priority Matrix: impact vs effort

Not every test idea is worth running. Prioritize using impact, confidence, and ease, focusing first on high-impact, low-effort changes.

Six ideas: (1) Rewrite the checkout CTA copy, (2) Full homepage redesign, (3) Shorten the signup form from 6 fields to 3, (4) Change the 'Add to Bag' button color, (5) Add a size-fit quiz to product pages, (6) Rebuild the entire mobile checkout flow. Score each 0-1 on effort and impact, then identify the 3 quick wins.

Google Sheets— Build a 4-column sheet: Idea, Effort (0-1), Impact (0-1), Quadrant.

Procedure

  1. Score each of the 6 ideas on effort (0=low, 1=high) using team estimates
  2. Score each on impact (0=low, 1=high) using the lesson's benchmark data (form length, CTA clarity)
  3. Plot each into a quadrant: Quick Wins (low effort, high impact) vs Plan Carefully (high effort, high impact)
  4. Select the top 3 from Quick Wins first, only pulling from Plan Carefully if fewer than 3 quick wins exist
Sample output
Idea                         Effort  Impact  Quadrant
Shorten signup form 6->3     0.3     0.85    Quick Win
Checkout CTA copy rewrite    0.2     0.7     Quick Win
Size-fit quiz                0.5     0.75    Plan Carefully
Add to Bag button color      0.15    0.25    Skip/Deprioritize
Homepage redesign            0.9     0.8     Plan Carefully
Mobile checkout rebuild      0.95    0.85    Plan Carefully

Healthy

The 3 selected tests are all Quick Wins or the highest-impact Plan Carefully item, never the button-color test.

Unhealthy

Picking the homepage redesign first because a stakeholder is excited about it, ignoring the effort score.

What this means

Impact and effort scoring exists specifically to overrule internal politics and hunches with a repeatable ranking.

So what do I do about it?

SymptomActionEffort
The roadmap defaults to whichever team pitched loudestScore every idea on the same 2 axes before ranking30 min
YouYou can do this yourself, no engineering access required.

Step 02 of 02

Hypothesis format: If X then Y because Z

Write one hypothesis per test using the format: 'If we [change X], then [metric Y] will improve because [reason Z].' Change only one element per test.

Your #1 ranked test is shortening the signup form from 6 fields to 3. Write the hypothesis in the required format, and name the one variable this test isolates.

Google Sheets— Add a Hypothesis column next to the #1 ranked row.

Procedure

  1. State the single change: removing 3 non-essential fields
  2. State the metric: form completion rate
  3. State the reasoning: fewer fields lowers the effort required to convert
  4. Confirm no other element (button color, copy, layout) changes in the same test
Sample output
Hypothesis: If we reduce the signup form from 6 fields to 3 (name, email, password), then form completion rate will improve, because each additional field adds friction that causes drop-off before submission.
Isolated variable: field count only. Button copy, color, and layout stay identical to control.

Healthy

One variable changes; the hypothesis names a measurable metric and a clear reason.

Unhealthy

A hypothesis that changes the form length AND the button color AND the headline in the same test, making the result unattributable.

What this means

A hypothesis that isolates one variable is the only kind that produces a usable, attributable result.

So what do I do about it?

SymptomActionEffort
The test brief lists 3 simultaneous changesSplit into 3 sequential single-variable tests instead30 min
YouYou can do this yourself, no engineering access required.

Final deliverable

A ranked 6-item test roadmap with quadrant placement for each idea, plus one fully formatted, single-variable hypothesis for the #1 test.

See a reference example
Sample output
Lenskart Q3 test roadmap (excerpt)

#1 QUICK WIN: Remove redundant prescription-upload step (effort 0.25, impact 0.8)
Hypothesis: If we combine the 2-step prescription upload into 1 step, then checkout completion will improve, because removing a step reduces the chance of mid-flow abandonment.

#2 QUICK WIN: Add a 'lens type' filter to the PDP (effort 0.3, impact 0.7)
#3 PLAN CAREFULLY: Rebuild the frame try-on AR flow (effort 0.8, impact 0.85)

Success criteria

You're done when you can:

  • All 6 ideas are scored and correctly placed into a quadrant
  • The top 3 selections follow the quadrant ranking, not intuition
  • The hypothesis isolates exactly one variable and names a measurable metric