AdWords
2K members online now
2K members online now
Improve your AdWords performance and boost your ROI, CTR, and Quality Score
Guide Me
star_border
Reply

How to choose which landing pages work and which don't

[ Edited ]
Collaborator ✭ ✭ ✭
# 1
Collaborator ✭ ✭ ✭

Hello All,

 

At what parameters are you looking when you decide that a landing page definitely needs changing , except the conversion rate ?

 

For keywords it's not so hard to take this decision to delete them when they do not bring enough conversions or asist other conversions.

2 Expert replyverified_user
2 ACCEPTED SOLUTIONS

Accepted Solutions
Marked as Best Answer.
Solution
Accepted by topic author Adrian B
September 2015

Re: How to choose which landing pages work and which don't

Top Contributor
# 2
Top Contributor

Hey Adrian,

 

As per my understanding, the first impression is your ad copy which is visible to the user when he is looking to buy any product or service. So your ad copy correlates to the landing page and the user while looking at your ad copy will assume that whatever promises you are making in the ad must be there on your landing page to justify it.

 

Now once the user is on the landing page, it totally depends upon your page now whether the user is going to take any action in the form of filling up the form or making any purchase. I would say that the bounce rate, time spent, exit rate etc...would be few of the metrics to look at to decide whether the landing page is proving itself or not. Then based on the statistics you have, you can decide whether to change the landing page or not.

 

There's something called Content Experiments available in Google Analytics where you can experiment different version of the landing pages and choose what is working and what not.

 

https://support.google.com/analytics/answer/1745147

 

My thoughts!

Pankaj

View solution in original post

Marked as Best Answer.
Solution
Accepted by topic author Adrian B
September 2015

Re: How to choose which landing pages work and which don't

[ Edited ]
Participant ✭ ☆ ☆
# 6
Participant ✭ ☆ ☆

I have done some content experiments,

(or in the past it was website optimizer)

but the outcome was inconclusive.

To react to the last part of your post:

What kind of experiment would you prefer :

Adwords experiment as you describe :
Two the same adgroups with only one difference :
Control Adgroup  : destignation URL to landing page A
Experiment Adgroup :  destignation URL to landing page B


Or a content experiment with landingpage A and B

Difference :
Adwords experiments :
- Only visitors via Adwords Ad, not visitors via referrals en organic search
- Same Visitors in experiment can see different pages. With click 1 on ad landingpage A, with click 2 on second A landingpage B etc.


Content experiment :
- All visitors
- A visitor will see always the same page if the visitor does not delete cookies. The first page he sees, will per example page B, than he will see page B in second and third visit.

The experiment tool will tell you if result is significant.

** I´am learning Adwords and find it very interesting.
** Ik ben Adwords aan het leren en vind het erg boeiend.

View solution in original post

Marked as Best Answer.
Solution
Accepted by topic author Adrian B
September 2015

Re: How to choose which landing pages work and which don't

Top Contributor
# 2
Top Contributor

Hey Adrian,

 

As per my understanding, the first impression is your ad copy which is visible to the user when he is looking to buy any product or service. So your ad copy correlates to the landing page and the user while looking at your ad copy will assume that whatever promises you are making in the ad must be there on your landing page to justify it.

 

Now once the user is on the landing page, it totally depends upon your page now whether the user is going to take any action in the form of filling up the form or making any purchase. I would say that the bounce rate, time spent, exit rate etc...would be few of the metrics to look at to decide whether the landing page is proving itself or not. Then based on the statistics you have, you can decide whether to change the landing page or not.

 

There's something called Content Experiments available in Google Analytics where you can experiment different version of the landing pages and choose what is working and what not.

 

https://support.google.com/analytics/answer/1745147

 

My thoughts!

Pankaj

Re: How to choose which landing pages work and which don't

[ Edited ]
Participant ✭ ☆ ☆
# 3
Participant ✭ ☆ ☆

THE SEXIEST METRIC EVER
http://www.youtube.com/watch?v=ppgfjo6IIf4

 

For the rest all the engaging metrics I have. Pankaj did mention some.

** I´am learning Adwords and find it very interesting.
** Ik ben Adwords aan het leren en vind het erg boeiend.

Re: How to choose which landing pages work and which don't

[ Edited ]
Follower ✭ ✭ ☆
# 4
Follower ✭ ✭ ☆

Apart from the conversion rate I would consider page load time and bounce rate.

Re: How to choose which landing pages work and which don't

Collaborator ✭ ✭ ✭
# 5
Collaborator ✭ ✭ ✭

Has anyone used Content experiments or took a radical decision about a landing page because of factors x or y ? Can you say that you decided that a landing page needs to be changed when bounce rate was at a certain value ?

 

I mean all the debates here in the forum are about improving a landing page, not about deciding between 2 or 3 landing pages which to keep.

 

What can we do except Content experiments, can we create 2 ad groups with the same keywords and ads but change the ads to a different landing page in the second ad group ? I see no easy way to decide like :

If metric 1 is above below x , change landing page.

Marked as Best Answer.
Solution
Accepted by topic author Adrian B
September 2015

Re: How to choose which landing pages work and which don't

[ Edited ]
Participant ✭ ☆ ☆
# 6
Participant ✭ ☆ ☆

I have done some content experiments,

(or in the past it was website optimizer)

but the outcome was inconclusive.

To react to the last part of your post:

What kind of experiment would you prefer :

Adwords experiment as you describe :
Two the same adgroups with only one difference :
Control Adgroup  : destignation URL to landing page A
Experiment Adgroup :  destignation URL to landing page B


Or a content experiment with landingpage A and B

Difference :
Adwords experiments :
- Only visitors via Adwords Ad, not visitors via referrals en organic search
- Same Visitors in experiment can see different pages. With click 1 on ad landingpage A, with click 2 on second A landingpage B etc.


Content experiment :
- All visitors
- A visitor will see always the same page if the visitor does not delete cookies. The first page he sees, will per example page B, than he will see page B in second and third visit.

The experiment tool will tell you if result is significant.

** I´am learning Adwords and find it very interesting.
** Ik ben Adwords aan het leren en vind het erg boeiend.

Re: How to choose which landing pages work and which don't

Top Contributor
# 7
Top Contributor
You can consider bounce rate or time spend on a landing page as two crucial factors, but at the same time always consider the ‘type’ of landing page you have.

A specifically created PPC landing page with almost all the information about the product or service on it can have a higher bounce rate and lower time spend statistics, as most of the visitors don’t have to move to another page to make a decision.

So it is always important to analyse what kind of landing page you have.

Just a heads up!

Deepak

Re: How to choose which landing pages work and which don't

Collaborator ✭ ✭ ✭
# 8
Collaborator ✭ ✭ ✭

The content expriment page provides an example and a conclusion "When you see which page drives the most conversions, you can make that one the live page for all visitors."

 

Assuming we are only discussing this about a simple Search Campaign ,and landing pages about Leads not sales (which can be measured easy) ,  the decision steps to take in order to choose a bad landing page would be  :

 

Make a table of 4 columns and as many rows as the landing pages are.

 

1) On the first column put the conversion rates ( the lower they are , the worse is the landing page)

 

2)  Put in the second column bounce rates

 

3) Put in the third column average time on page.

 

4) In the blank column no 4 , multiply time on page by the conversion rate and divide it by the bounce rate . A higher bounce rate is bad and it will result in a smaller number, also a small time on page is bad and it will result in a smaller number. The candidate with the smallest number is the worst landing page.

 

(avg. TimeOnPage x Conv. Rate ) / Bounce rate = indicator of efficiency ?

 

I just made this up, everyone's input is welcome.

 

Re: How to choose which landing pages work and which don't

Participant ✭ ☆ ☆
# 9
Participant ✭ ☆ ☆

Hi Adrian,

Bounce rates do not lead to smaller time on page, because it is not counted when there is a bounce. Time on page is counted if Google analytics knows a start time and could measure a second activity on the website. Because Google analtyics could measure a second activity, it is not a bounce. visits with Bounces are left out with this metric.

 

Time on page is calculated as the difference between the initiation of successive interactions that trigger the tracking pixel. (like pageview or event or ecommerce transaction)


I don´t see the purpose off what you are doing.
If in an experiment you see that conversions are better in one page. Why then worry about bouncerate or time on site? Maybe you do not have to do that.

As described above by Deepak, not always is bouncerate a indicator. You must do web analyses. See all in perspective. to what conclusion does all metrics tell you. Try to see whole picture. Maybe something else is happening.

And I would think that the ratio conversions / visitors would be more important or ratio conversions / clicks.

But that does not say all. If you have 2 products/services 1 with profit 1000 euro per sale, and one with profit 10 euro per sale, than ratio is different for what is positive.

Or maybe more important is ratio profit / visitors and profit / clicks.

** I´am learning Adwords and find it very interesting.
** Ik ben Adwords aan het leren en vind het erg boeiend.

Re: How to choose which landing pages work and which don't

[ Edited ]
Collaborator ✭ ✭ ✭
# 10
Collaborator ✭ ✭ ✭

I want to work as little as possible with hunches and maybes , even if these can help.

 

For example if we have 3 or 4 scenarios of analyzing a landing page ( from the conversion point, from the bounce point of view, etc etc )    I want to establish the formula / pseudo code / flow of operations, needed to provide a simple end result : replace Landing Page because it does not generate what you want .