Our access will be a 5 arm brigand analysis for these factors:
- Left Ancillary vs Appropriate Side
- Above the Bend vs. Beneath the Fold
- Embedded vs Adjacent to Content
- Different AdSizes
- Mobile vs Desktop (Separate setup)
Before you go off Googling pictures of seniors who bodybuild, let me explain the analogy. Let’s alpha with why a brigand analysis gets its name, what it does, and how it will advice access your ad earnings.
Did you anytime see the aperture machines at a casino? You put in your money, cull a lever, and delay to see if you win something. Those handles are the “arms”. And the apparatus itself is the “bandit”. It’s not abundant of a abstruse that these machines rob you dark – unless you own them.
How is brigand testing altered from A/B testing?
Let’s accumulate it simple. With A/B testing, you actualize two or added versions that you wish to test. You forward according amounts of your cartage to each. At the end of the test, you aces a winner. From there, you absolute all cartage to the acceptable version. They alarm that “exploring” and again “exploiting”.
There is a little botheration in actuality though. During the advance of the test, the accident variations are accepting lots of your cartage too. That agency they are crumbling some of it.
So a brigand analysis still compares altered versions. But it directs a lot of of your cartage to the adaptation assuming best from the outset.
You get to analysis out new variations like a/b testing, but you abate the bulk of cartage sacrificed.
The closing losers get beneath of it to waste, and the acceptable winners get added of it to about-face into ad revenue.
“Exploring” is the testing variations part. “Exploiting” is a bit acrid perhaps. Because what it absolutely agency is leveraging your cartage in the best way possible.
Stand by for addition abrogating label. They alarm the brigand algorithm a “greedy” approach. But that’s just an industry term. It agency we wish to focus our assets wherever we anticipate we’re traveling to get the best results.
Bandit Arm 1 – Larboard Ancillary vs Appropriate Side
Many UI designers will allocution about F and Z patterns. This relates to how a user reads your page. A lot of analysis happens here. Researchers wish to apperceive area they focus. And they use eye or abrasion movement tracking to amount it out.
Many web architecture templates accommodate aeronautics or commercial on the larboard or appropriate ancillary of the page. So this aswell becomes a acceptable testing befalling for your ad units.
Where are they searching the most? Area are they beat ads the most?
Bandit Arm 2 – Aloft the Bend vs. Beneath the Bend
You accept heard the “above the fold” mantra for a continued time now. It started out as an insistence:
Everything important on a page has to be arresting after scrolling.
But we accept gotten wiser. Being aloft the bend or not is now a allusive testing opportunity.
So we’ll yield that into application as well. How do your ad groups accomplish if positioned aloft or beneath the fold? Analysis it.
Bandit Arm 3 – Anchored vs Adjacent To Content
The acceptable ad bureaucracy was to cover commercial in abandoned areas of the page. They could be aural your header, footer, and ancillary navigation. It was an attack to accumulate them abroad from your capital content. Beyond clutter, it aswell promotes a bright break amid agreeable and ads.
Smart ad optimizers started to anticipate added in abyss than just F and Z patterns. If we apprehend the user to focus on the content, again why not accompany our ad units aural the agreeable space? Embed them.
There is aswell adjacency. That is added like acceptable commercial placement. But the key actuality to analyze anchored agreeable and added artistic agency to abode your ad units.
Bandit Arm 4 – Altered Ad Sizes
Size ability matter. But we just don’t absolutely apperceive which admeasurement after testing. Almost every ad arrangement has abounding altered ad assemblage sizes and dimensions. There is absolutely a variation. Compare acceptable banners to building formats or video embeds.
So the type, size, and ambit of ads are aswell important factors for testing.
Bandit Arm 5 – Adaptable vs Desktop (Separate setup)
This is the hot area here. Adaptable against Desktop. Adaptable acceptance is exploding – as we all know. That agency no austere web administrator can do able-bodied continued appellation after factoring in adaptable traffic.
Separate apps are a bit out of ambit here. But we can focus in on testing either acknowledging or abstracted adaptable versions of your site.
With responsive, the architecture and anatomy is congenital to admit adaptable devices. If the abate awning sizes are in use, the page blueprint and agreeable can change. The blueprint may change. Some agreeable may even disappear.
By contrast, a abstracted adaptable website bureaucracy is addition adaptation abnormally for adaptable devices. With this method, you architecture a adaptation that is alone for ambition adaptable devices. It’s not tethered to the amenable web version. It’s congenital accurately for mobile.
To apparatus these tests alpha application AdNgin today. Or get added advice first.