How to Run Landing Page Tests On and Off Paid Social Platforms

Join us for an exclusive webinar featuring Mariate, Alexandra and Nima where we will unveil a comprehensive blueprint for crafting a successful paid media strategy focused on landing page testing.With escalating costs in paid advertising, understanding how to maximize each visitor’s experience is crucial for retention and conversion.

This session will dive into the methodologies for executing and analyzing landing page tests within paid social channels, offering a blend of theoretical knowledge and practical insights.

The Pearmill team will guide you through the nuances of setting up and managing landing page experiments on paid social platforms. You will learn about the critical rules to follow, the structure of effective tests, optimal conversion duration and budget allocation.

The session will also cover data analysis techniques and criteria for graduating landing pages.

In the second part of the webinar, Pearmill will explore the use of A/B testing platforms. Discover common pitfalls to avoid in A/B testing and gain insights into analyzing A/B tests results effectively.

1. How to run landing
page tests on and off
paid social platforms

May, 2024
Advanced Landing Page Strategy

2. The team
Mariate
Paid social lead
Alexandra
CRO lead
Nima
President

3. What are we
talking about
today?
Introduction to landing page testing
How to run landing page tests in
paid social channels
A. Rules we follow
B. Structure
C. Conversions, durations & budget
D. Testing budget simulation
E. Data analysis
F. LP graduation
2
1

4. How to run landing page tests
in an A/B testing platform
A. Setting up your LP test
B. What to look for when analyzing A/B test results
C. Common pitfalls when A/B testing
What are we
talking about
today?
3
Pros & cons of each4
Case studies5
How Pearmill can help you6
Q&A7

5. Introduction
to landing page
testing: background
Landing page testing is one of the biggest levers
we can pull to improve performance in the paid
social channels.
These tests can run on Paid Social
or on an A/B testing platform.
We increased CvR by +143% for our delivery app
client. We also increased purchases by 15% for one
of our healthcare B2B clients.
We will demonstrate how to do this on both
platforms, as well as explain the benefits of
each, depending on your circumstances.
1
2

6. How to run
landing page
tests in the paid
social channels

7. Rules we
follow
Use the campaign and ad set settings from
your best performing evergreen campaign.
Create a separate landing page 
testing campaign
A benchmark is the control — in other
words, it’s your current best performing
landing page. 
We'll compare all new landing pages to
this control.
Include a benchmark variable
(aka “a control”)
1
2

8. Example
structure
Campaign
Ad sets
Ads + URLs

9. Conversions, duration,
& budget
We determine budget based on
the number of variables we have,
and how many conversions we want
to attain per variable. 
In general, we don’t want to spend
more than 20-25% of the budget
on a testing campaign to keep
performance in the account stable.
Budget

We aim to get at least 20 conversions
per variable - and more if possible.
This gives you stronger directional data
on the best performer.
How many conversions
to aim for? The test should run at least 7
consecutive days, but we recommend
running for 15 days if possible.
This duration allows you to take into
account the good and bad days the
accounts have, and ensures we have
spend on every ad every day of the
week.
Testing duration

10. Use case:
An e-commerce client is using their
homepage as an evergreen landing page.
They are interested on testing:
A: A paid social-specific landing page
B: Driving traffic to their full catalogue
of products
Testing budget
simulation
Formula variables example
Budget = $x
Variables to test: 3 (2 New + Benchmark)
Conversions per variable = 20
CPA = $75
X = 3 * 20 * 75 = 4500
Budget (x) = N variables * N conversions * CPA

The numbers we need for our formula

# of variables we want to test
# of conversions we want to generate
per variable
CPA = Cost per action (average)
we optimize for in the ACQ ad set.

11. Data analysis:
rules we follow
Wait at least 72h after the test has concluded -
for all conversions to be properly processed
Wait & analyze
Metric priority when ranking ads:
Rank
1
2
Main CPA (cost per purchase in this case) 
Up-funnel metrics (cost per add-to-cart 
& cost per initiate-checkout in this case)
IPP (impressions per purchase) - this is a metric
we implemented to take into account differences
in CPM between different variables.

12. Landing page
graduation
Promoting the landing page winner to your
evergreen campaigns can cause disruption
and send everything back into learning
mode if not done correctly.
Use a URL in your ads that allows you to update
your designs directly (recommended).
Most advertisers change the links in all of their ads.
However, if you use a landing page in your ads that
your CRO team can update directly with your
design winners, then you won’t need to go through
the process of changing the links in your ads.
1
If you need to change ad links, re-launch
your ads or ad sets
Never change the links directly in ads that are
running. If you change the links and something goes
wrong, you won’t be able to revert the changes.
If you launch new ads, you can always go back
to the previous ads if needed.
2

13. How to run LP
tests in A/B
testing platforms

14. Setting up your
landing page test

Define the objective Create variations inside
the A/B testing platform
Set the parametersState the goal of the test
and the metric you want
to improve Depending on the objective
and/or limitations, choose
between setting up an A/B
test, MVT, or Split URL test Determine the test duration,
audience, & variant weight.
Select the metrics you want
to optimize for
32

15. https://cxl.com/ab-test-calculator/ State the goal of the test and the metric
you want to improve:
Use a sample size calculator to determine
the test duration based on your traffic and
desired level of confidence.
Example: CXL calculator; Optimizely sample
size calculator
Align your test goals with overall business
objectives
Avoid running concurrent tests on the same
audience to prevent interference and ensure
accurate results
Setting up your
A/B test
Define the objective1

16. Depending on the objective and/or
limitations, choose between setting up
an A/B test, MVT or Split URL test.
A/B test
Test one variable with two
or more variations
Multivariate test (MVT)
Test multiple variables simultaneously
Split URL test
Test entirely different pages
or layouts, sections
Create variations 2
Setting up your
A/B test

17. Examples: all visitors, specific
segments, or custom audiences
Examples: conversion rate,
engagement metrics
Default: equal distribution among
variations
Select target audience for your test
Select the primary and secondary metrics
Set the variant weight (traffic
allocation) for each variation
Set the parameters 3
Setting up your
A/B test

18. Isolate the variables to clearly understand
their impact on the desired metrics
Changing too many
variables at once
Determine the test duration by
doing a pre-test analysis
Having too short a test duration
Common
pitfalls when 
A/B testing

19. Not having a clear
hypothesis
Use problem statements to build
a hypothesis
Example: If [CAUSE], then
[EFFECT]  because [RATIONALE]
Structure: Problem - Solution - Result
If we simplify the checkout process by
reducing the number of form fields and
steps based on the user behavior insights
from the scroll-mapping data, then the cart
abandonment rate will decrease because
the data suggests the users are dropping
off at specific points in the checkout flow
due to complexity and friction.
Common
pitfalls when 
A/B testing

20. What to look for when
analyzing A/B test results
Statistical significance
Use statistical methods to determine
whether observed differences between
the control and variant(s) are likely due
to chance or a genuine effect Impact on key metrics
Analyze how test results impact
important metrics such as CvRs,
revenue, and user engagement Confidence levels
Assess the reliability of the test results
by considering the confidence level,
which indicates the probability that
the observed difference is real Long-term strategies
Consider how test insights can inform
long-term optimization strategies and
future test ideas

21. Pros & cons of running LP tests
on and off Paid Social
Paid Social
No additional costs
Inability to run other tests at the same time
Doesn’t require additional knowledge
Running these tests can cause disruption
on the account
Reaching significance is harder with Paid Social
channels, and we tend to use directional data
instead
Cons
Pros
A/B testing platform
Other tests can be running in the platform
at the same time
Additional costs
Zero disruption on the account as it runs off Paid Social
Reaching significance is easier in VWO as the
test can run for a longer period and for a larger
portion of the traffic
It does require additional knowledge
Cons
Pros

22. Case studies

23. Delivery app
The variation registered a significant
uplift (+145.24%) in leads compared
to the control 
The integrated sign-up form made
it easier for users to register without
having to navigate to another page,
which was especially beneficial on
mobile devices

24. Healthcare 

B2C client
The variation registered an
uplift in purchases (+15%) and
generated significant uplifts on
all metrics inside the funnel

25. Pearmill 

support

26. How Pearmill can help
Full Stack Services Paid Social + CRO Services
6 Growth Marketing – Paid Search & SociaP
6 Ad Creative Production & Optimizatio=
6 Conversion Rate Optimizatio=
6 Growth Analytics & Attribution
6 Paid Social Managemenb
6 Strategy & Executio^
6 Conversion Rate Optimizatio^
6 CRO Strategn
6 Landing Page Design & Development

27. Q&A

28. Thank

you!