Home > >

When testing angles, would it be a good idea to..... (7)


10-23-2015 05:35 AM #1 victor s (AMC Alumnus)
When testing angles, would it be a good idea to.....

Hi all,

I'd like to get your guys input on this.

When testing angles, would it be a good idea to keep the same format for all the banners?

For every angle that I come up with for my offers, I'm going to create 5 banners. But, I don't know if I should change the banners format. By format, I mean the color / font / CTA / titles and design/positioning. Or would it be best to leave all 5 banners the same and just change up the image?

Hope I'm making myself clear ;p If not, let me know and I'll elaborate.

Thanks!


10-23-2015 11:08 AM #2 caurmen (Administrator)

Good question!

I'd recommend changing the design once within those 5 banners, just in case your existing banner design doesn't work for some reason. (This is a reason to use slightly more banners - 8 or so - as it gives you more latitude in testing).

But it's not a disaster to use a single layout for all banners, provided you're consistent with it between angle tests. If ALL your angles generate impossible-to-profit CTRs, then run a second test with a new layout.

Hope that helps!


10-23-2015 12:19 PM #3 MarinaKimia (Member)

Hi victor s,

I hope this is what your asking about, if not just let me know, or PM me.... happy to help ...

So some of the experience we have had at Kimia.

In terms of font, color, CTA etc. it´s important to adapt the texts and CTAs to each format. The bigger the format the more you can show obviously, and you need to take advantage of that.

For example, when you use a small format, it´s important for the text and CTA to be visible and readable instantly and be short, of course. On the other hand, for larger formats, let´s say that you can "play" a bit more with the CTA and text, but the swift readability of the text is always something to keep in mind. Also, making the CTA stand our from the background colour (seems quite obvious but believe us its not always the case xD)

In terms of maintaining the images, it depends of the image really, one image can be good in many formats, depending on the angle. The important thing is to keep consistency in CTA button through the banners (if we´re talking about the same ad or campaign).

One image can be adapted to different format (again, depending on the image) what we would recommend is that you do go for different images per format, it brings a variety and dynamism to the creativity. This allows you to analyse data afterwards and see which image works best for you, then you can build on that for your banners... So basic market research right there

Cheers, good weekend! (and again, if theres anything Ive missed from your question, do let me know!)


10-23-2015 03:47 PM #4 victor s (AMC Alumnus)

Quote Originally Posted by caurmen View Post
Good question!

I'd recommend changing the design once within those 5 banners, just in case your existing banner design doesn't work for some reason. (This is a reason to use slightly more banners - 8 or so - as it gives you more latitude in testing).

But it's not a disaster to use a single layout for all banners, provided you're consistent with it between angle tests. If ALL your angles generate impossible-to-profit CTRs, then run a second test with a new layout.

Hope that helps!
Thanks for your input on this, Caurmen!

Quote Originally Posted by MarinaKimia View Post
Hi victor s,

I hope this is what your asking about, if not just let me know, or PM me.... happy to help ...

So some of the experience we have had at Kimia.

In terms of font, color, CTA etc. it´s important to adapt the texts and CTAs to each format. The bigger the format the more you can show obviously, and you need to take advantage of that.

For example, when you use a small format, it´s important for the text and CTA to be visible and readable instantly and be short, of course. On the other hand, for larger formats, let´s say that you can "play" a bit more with the CTA and text, but the swift readability of the text is always something to keep in mind. Also, making the CTA stand our from the background colour (seems quite obvious but believe us its not always the case xD)

In terms of maintaining the images, it depends of the image really, one image can be good in many formats, depending on the angle. The important thing is to keep consistency in CTA button through the banners (if we´re talking about the same ad or campaign).

One image can be adapted to different format (again, depending on the image) what we would recommend is that you do go for different images per format, it brings a variety and dynamism to the creativity. This allows you to analyze data afterwards and see which image works best for you, then you can build on that for your banners... So basic market research right there

Cheers, good weekend! (and again, if there's anything Ive missed from your question, do let me know!)
Thanks for your very detailed response, MarinaKimia! I'll keep this all in mind when creating my banners.


10-26-2015 09:28 AM #5 cbrughmans (Member)

When testing something, always change only one variable. Otherwise you won't know if the change in performance is caused by variable 1 or 2.

- if you change the banner size or visual, then keep the traffic source and landing page the same.
- if you change the traffic source, then keep the banner and landing page the same
- if you change the landing page, then keep the traffic source and banner the same

Keep testing as simple as possible so you don't lose overview, don't spend too much time & can start upscaling faster!


10-26-2015 10:11 AM #6 MarinaKimia (Member)

Quote Originally Posted by cbrughmans View Post
When testing something, always change only one variable. Otherwise you won't know if the change in performance is caused by variable 1 or 2.

- if you change the banner size or visual, then keep the traffic source and landing page the same.
- if you change the traffic source, then keep the banner and landing page the same
- if you change the landing page, then keep the traffic source and banner the same

Keep testing as simple as possible so you don't lose overview, don't spend too much time & can start upscaling faster!
Good advice for the testing process. Definately.


10-28-2015 08:48 AM #7 acepowermarketing (AMC Alumnus)

Quote Originally Posted by cbrughmans View Post
When testing something, always change only one variable. Otherwise you won't know if the change in performance is caused by variable 1 or 2.

- if you change the banner size or visual, then keep the traffic source and landing page the same.
- if you change the traffic source, then keep the banner and landing page the same
- if you change the landing page, then keep the traffic source and banner the same

Keep testing as simple as possible so you don't lose overview, don't spend too much time & can start upscaling faster!
this


Home > >