Well the new Google tool “Content Experiments” just came out and replaces the old school “Website Optimizer”. One interesting thing is that the Multivariate or Multivariable split testing has gone away.
This certainly simplifies things, but is it for the best?
Personally I like to have options. Much of the time, simple A/B testing is the way to go anyway, because you need a lot of data to get results.
However, with optin pages, you CAN get a lot of data. Especially if you’re running a lot of traffic.
Heck, if you’re a getting decent flow of visitors, you can easily test a dozen attributes of your web page and set up a new test every week.
So, that’s too bad for Google users…
But, great news for Extreme Optimizer users, because now you’ll have an additional edge to test and optimize your way to success.
Multivariable split testing can be a powerful tool in your arsenal.
Not really sure why Google would intentionally take away features… Maybe it is because of the development effort since Google Content Experiments is now being integrated into Analytics.
(Don’t get me started on Analytics — it too has its pluses and minuses.)
Anyway, the only thing that really matters is that you’re creating conversion breakthroughs for your sites.
We’re here to help.
Click here to check out our Extreme Optimizer software and see all the powerful ways that you can use it to increase your conversions.
– Claude Johnson
Infinite Profit Solutions – Chief Architect
If you haven’t heard, Google is changing “Google Website Optimizer” into “Content Experiments”.
I personally feel that this is the best thing ever because a lot more people are going to move AWAY from using Google’s website optimization tools and move into alternative solutions.
When I talk to people about optimizing their web pages and split-testing tools, they often ask “What do you think of GWO (Google Website Optimizer)”?
Most of the time, my answer goes something like this:
It’s ok if you’re broke and can’t afford a real testing solution.
This means that you instantly lose 20% to 30% of the data you should be getting. In my opinion, DATA = MONEY. And to lose 20% of valuable data is completely unacceptable.
And now it’s going from “ok” to BAD.
In the grand scheme of things, I’m actually a big fan of many of Google’s tools. However, I simply do NOT understand these recent changes they’ve made to their testing software.
Why on earth would anyone use Google’s new “Content Experiments”?
Here’s 6 reasons why I don’t think anyone should touch it just yet:
1. They’ve killed multi-variable testing. What if you want to do a multi-variable test on your landing pages?
2. Content Experiments is now “buried” in Google Analytics. And most people are already intimidated by the overwhelming array of options offered by the Google Analytics tool. Now Google Analytics will be scarier than ever.
3. You can only test 5 variations at a time. What if you want to test more things?
4. Content Experiments won’t declare a “winner” for at least 2 weeks – and you can’t run a test beyond 3 months. What if you have an outright winner before 2 weeks is up? Or what if a site has relatively low amounts of traffic, and needs more than 3 months to acquire sufficient testing data?
5. You have you to create multiple webpages instead of just having 1 url. This takes more time to set up and will clunk up your server with additional pages.
6. Content Experiments is currently limited to a maximum of 12 tests at a time. Now, this doesn’t mean you can only run twelve tests per account. But if you have 12 tests running, you’ll have to wait for a test to finish before being able to set up another one. (For people just getting started, this is likely to be fine, but it may force more active and experiences split-testers and website optimizers to seek elsewhere for a more suitable solution).
IT’S NOT ALL BAD…
True, Google have stated they intend to improve the features of the “Content Experiments” tool over time – and that the current version is likely to be improved in the future. (However, only time will tell, since it’s forerunner, Google Website Optimizer, remains pretty much unchanged since its introduction more than 5 years ago).
Naturally, on this site we advocate and encourage our readers to use our very own proprietary split-testing software – Extreme Optimizer. And perhaps one of the most edifying pieces of news to come from Google Website Optimizer’s demise is that more people will be actively looking for alternative solutions to meet their needs.
JUST SO YOU KNOW…
Extreme Optimizer is so much easier to use. And the features eclipse Content Experiments by a few thousand light years.
Written by Matt Gallant, the mad marketing scientist. To learn more about Extreme Optimizer “the ultimate optimization software” -> click here. To hire Matt Gallant and the mad marketing crew and have them optimize your business RISK-FREE, please -> click here.
Optin pages are one of the very best places to learn split-testing. Why?
Because you can get tons of results daily (even if you have only a little bit of traffic). You can really test tons and tons of ideas quickly and get some good experience.
Remember, an optin page is at the “top of the funnel” where potential prospects enter your world. So you should ALWAYS be testing new ideas on your optin pages, and aiming to improve conversion and performance.
What we’re going to do here is deconstruct all of the key elements of the optin page. And once you’re able to identify these various elements, it makes it a lot easier to decide what you need to split test (and the order of which elements to test 1st, 2nd, 3rd etc.)
Let’s keep this part simple.
When we are testing an optin page, here is the order of things to test:
1. The headline
2. The offer
3. The format
4. The first paragraph
7. The pre-head
And here’s the process I suggest you use: test 1 element at a time and test 3 variations (including the original or the “control”).
So for example, let’s say I want to start testing the headline – I would create 2 new headlines and test them against the original one. Once that test has produced a ‘winner’, I would test the offer. Then, you can test the format… after that the first paragraph, then the pictures… then video and then the pre-head…
1. The title of the webpage – this is the text that’s on top of the webpage itself. Ideas to test:
A) enter the keyword, such as: “you want some insider info on keyword?”
B) try several benefits, “want to teach your dog new fun tricks tonight?”
C) state the offer: “get 25 free videos below…”
2. The prehead – this is the headline before the main headline. Ideas to test:
A) enter the keyword, such as: “you want some insider info on keyword?”
B) highlight the small amount of time it’s going to take to get the reward “in just 30 seconds, you can own…”
C) ask a qualifying question, “do you love guitar soloing?”
3. The headline – this is the heart and soul of any head. This sells them on reading more. Ideas to test:
A) state the offer “who else wants 15 free dog training videos?”
B) very important: test different titles of the report (or offer). Example: test “47 hot traffic tips” vs. “47 insider traffic tricks” etc…
C) challenge them: “do you have what it takes to get ________?” (this worked surprisingly well in one of the markets we’re in.)
Advanced tip: run a multivariable test just on the headline. In other words, you will split- test every single word of the headline. This can give you astounding results. (But be aware this is part of the ‘Vertical Testing’ phase, after you have tested lots of different ideas in the ‘Horizontal Testing’ phase).
4. Sub-head – this is another headline that follows up the main headline or it’s a headline inside the ad that “breaks up” the copy. Ideas to test:
A) state secondary benefits “you’ll feel closer than ever with your pooch.”
B) tell them what to do next: “simply answer the 5 questions below to get your insider report…”
C) tell them to “simply enter your name and email below now…”
5. The lead-in – what’s the concept behind the headline and the entire ad? Ideas to test: (this varies greatly from market to market so I’ll use the dog market as an example)
A) dog breeding secrets
B) how to give your dog a massage
C) dog grooming
D) dog tricks
6. The offer – this is perhaps the most important part. What are you giving them in return for them giving you their name and email? Ideas to test:
A) free reports
B) free videos
C) free software
D) free consultation
7. First paragraph – the first paragraph can make or break your flow. Ideas to test:
A) the “qualify-and-welcome” strategy “if you’re looking for better, faster ways to train your dog, then this is the right place for you.”
B) hit a pain button and offer salvation “are you sick and tired of eating small vegetables and watching half of your crop die or rot? Then, you’ll be thrilled with the report i’m about to share with you.”
C) none at all – on optin pages: less is usually more.
8. The format – how the page is set up. This is very important.
A) the clean-white design
B) pleasing color combos
9. Location of optin box
A) right below the headline
B) to the right of the headline
C) on the bottom of the page
D) to the left of the headline
E) on the top and on the bottom
F) next to video
G) below video
10. The fly-in pop up – this is a technology that allows a pop-up to kind of “slide in” after the visitor lands on your website.
A) comes on instantly
B) comes on after 5 seconds
C) comes on after 10 seconds
D) comes on after 20 seconds
E) none at all – on optin pages: less is usually more.
F) has video inside
G) makes a different offer (if your offer is videos on the main page – try offering a report).
11. The exit pop up – this is a pop-up that appears if they click “back” or try to close the browser. You offer them another chance to opt in.
A) none at all
B) has video inside
C) makes a different offer (if your offer is videos on the main page – try offering a report).
12. Salutation – this is the “dear xxxxx” part.
13. The signature – this is where you can put a graphic of your signature. Blue seems to work better in a lot of cases (but make sure you test this out for yourself). Also, you can test different “titles” such as “the only all-in-one online coach”.
14. Video – a video that tells them to opt-in…
15. Audio – an audio that tells them to opt-in…
16. Background color – the color of the background.
17. Pictures – pictures of you or your offer
18. P.S. – this can make a big difference as well.
19. Fonts – the type of font can actually make a big difference
20. Bolding, highlighting, italicizing – I don’t suggest you go crazy here, but these can make a difference sometimes.
TIP: We’ve done quite a few optin form tests and almost always, the shorter it is – the more it converts. However, be aware that Google doesn’t like short optin forms so much… especially if you don’t have links going to some content. Just remember Google likes sites with content and they like it if you have a blog and articles on the site too.
Now – it’s time to put this advice into action. Forget about the “theory” of testing and start testing. If you have never done testing before, I promise you’ll really enjoy it once you get started.
From August 1 2012, Google Website Optimizer (GWO) will no longer be available to its users. That’s right, Google are taking their free split-testing tool off the market.
Instead, they’ve made a new tool available inside their Google Analytics software, called “Content Experiments”. (It’s available now if you want to give it a try).
Clearly, Google has some very smart cookies working for them – so they’ve no doubt thought this through. But from a user perspective, it does beg the question whether it’s really the wisest move.
True, for those who already have Google Analytics installed, there’s now only one snippet of code to insert on a site for a testing experiment. (Compared to the multiple snippets that Google Website Optimizer requires). So that simplifies things.
Poring through the official Google statements about ending Google Website Optimizer, the emphasis is clearly on simplification as a whole. Their aim is to “simplify website testing”.
And sure, by cutting back on the kinds of tests you can perform on the new “Content Experiments” system – limiting it to A/B testing and eliminating multivariate testing functionality – the idea is evidently to make split-testing even more accessible to the masses, and to make it even easier to get started.
Not only that, their ‘Wizard’ provides step-by-step instructions on how to set up a test for the uninitiated.
But there are two clear challenges in all of this:
1) The existence of this new testing tool is not obvious inside of the Google Analytics interface. It’s not easy to find. So no matter how easy it is to use, it doesn’t really matter if the tool is hard to locate.
2) You’re now required to use Google Analytics if you want to perform a split test (which the old ‘Google Website Optimizer’ did not necessitate).
Google’s recently reported that more than 10 million websites have installed Google Analytics – so there’s clearly a wide user base who can be exposed to the merits of testing and optimizing their sites.
And that’s great for the state of the internet as a whole.
Yet only time will tell whether ‘burying’ this new testing tool inside of Google Analytics encourages more website owners to test and optimize their sites.
There’s also the strong possibility that more advanced users will search elsewhere for more advanced testing functionality than Content Experiments currently offers.
Unless Google adds some additional features like enabling multivariate testing – and broadens its limit of only 12 active tests at a time – in the next few months, heavyweight testers may need to expand their search for something a little more flexible and less limiting.
The Infinite Profit Solutions Team
Here’s a quick overview of the main steps we recommend for creating ads in your market that convert:
1. Split test Text Ads in the ‘search’ and ‘display’ networks of Adwords to find a high-converting ad.
2. Take the winning text ad and split-test image ads using the same copy, but different fonts.
3. Take the winning font image ad and split-test different pictures and images in Image Ads (while the font stays the same).
We usually roll out both Search Network and Display Network Text Ads right out of the gate, and test them out. (And if they’re not doing well, then we’ll try something else). But they’re usually identical when we first test them on the ‘Search’ and ‘Display’ networks of Google.
We like to start with Text Ads first and find copy that converts as the starting point. Again, it starts with ‘Horizontal Testing’, meaning we test really different ideas against one another.
Then once we’ve identified a winning concept or idea, we’ll move onto ‘Vertical Testing’. This is where we get more granular, only changing out certain words, and only testing that one word. (Some people change several things in the ad — we like to keep just one variable changing the whole time).
And here’s a few extra tips for creating your text ads too:
Keyword Inclusion: You definitely should try including the keywords you’re targeting in the ad. Preferably, try to have the keyword phrase at least twice in the ad (if possible, depending on how big the keyword phrase is).
And it’s very important to have the keyword in the HEADLINE of your ad. We’ll often test having the keyword in the headline and then the keyword somewhere in the body.
Just try to fill up the ad with the keywords…because in Google’s “search” network …Google will highlight the keyword if it matches the keyword the user types into the search engine. That means it will be bolded twice in the ad, and it’s much more likely to stand out from the rest of the ads on the Google search page. It’s just one more thing to drive eyes to your ad.
Ask Questions: Adding a question mark to your ad, especially in the headline, often works really well. For example, we’ll test asking questions in the headline, such as:
Toilet Train Your Cat? vs. Toilet Train Your Cat
…And it’s basically continuing a conversation that a person is already having in their head.
From our testing, asking a question often results in higher ‘Click Through Rates’.
Power Words: “Power words” are emotionally evocative words and phrases, and they can be both positive or negative.
Examples include: humiliate, slap, caress, kiss, awful etc… And when you include words like these in your ads they can sometimes help boost response as well. So they’re definitely worth testing.
Build up your own “swipe file” of power words, so you have plenty at hand when creating new ads to split test.
Always be searching for power words. I can’t say this often enough. Just ONE WORD can change everything. It all about whatever makes the concept more painful, more powerful, more delicious.
NOTE: John Carlton has a pretty awesome Power Word List if you want to look up his stuff.
Benefits and Calls-To-Action: Adding the benefits the user stands to gain is often a good idea. And you can add yet another dimension to your ads by making “line 2” about the benefit, while “line 3” has some kind of “call-to-action” that motivates users to click the ad.
Here’s an example:
Headline: Toilet Train Your Cat?
Line 2: Keep Your Floors Fresh & Clean
Line 3: Try This Foolproof System
Display URL: www.TrainYourCat.com (this is not our domain – purely for demonstration purposes)
Simply aim to have a “benefit” on “line 2” and a “call-to-action” on the “line 3”.
Sentence Case vs. First Letter Caps: Also worth testing is whether you capitalize the first letter of each word in your ad, or just use regular sentence caps.
First Letter Caps: Toilet Train Your Cat?
Sentence case: Toilet train your cat?
Of course, this is definitely a “Vertical Test”. So you’ll only want to try it after experimenting with plenty of ideas in the “Horizontal Testing” phase. (see our article on ‘Horizontal Testing vs. Vertical Testing’ for more on this).
Letting Line 2 “Bleed” Into Line 3: Most of the time your text ads will appear on 4 separate lines. And when this is the case, it’s often good practice to make each line its own self-contained idea.
However, if you get into the top positions, sometimes Adwords will stick lines “2” and “3” together onto ONE line. And when this happens, having “line 2” run into “line 3” will create this one very “long” ad, that stands out from others. And it logically flows. Certainly worth testing.
How Many Text Ads Should You Test?
There’s not really a “sweet spot” number of Text Ads to begin testing first before moving on to the image ads phase. You really just want to keep testing until you come up with text that you find has a good CTR (Click-Through Rate) that converts for you too.
That being said, there is a limit to the number of text ads you’ll want to test against one another at any given time.
We usually recommend testing 3 or 4 ads between themselves. We don’t usually try to test any more than that at a time, because we want to gather as much data as I can. (If you do more than 3 or 4, it just takes longer to get results that are worth using).
Once you have a really well-converting text ad, it’s time to move that over to the image ads. Use the exact same text copy in these image ads.
It couldn’t be easier. You simply use the same copy you’ve determined from the text ad, and turn it into an IMAGE AD for the Adwords Display network.
Extra tip #1: We find that the ad sizes that work the best for us are 300×250 pixels and 336×280 pixels. They seem to get the most traffic and the most results for our campaigns.
Extra tip #2: Split your ad groups into each image size. (One ad group per image size). If you split them all up in this way, you can really get an idea of how each size performs in your tests.
Test Different Fonts Against the SAME Copy
The font you use can have a significant effect on the performance of your image ad. And this is something very few people realize.
Again, fonts are really important. So try testing different fonts. We usually always start with ARIAL. And then we test against crazier fonts. And sometimes they’re just wild fonts. Sometimes they’ll work better than others. And if a wild font works, we’ll try another wild font up against it. We’re just constantly testing the fonts.
Our favorite resource for finding fonts to test is DaFont.com (http://www.dafont.com/) – you’ll find loads there. We also underline the text and keep the text color blue – so it looks like a link.
NOTE: Different ad sizes will require different font sizes. But for the 300×250 – because that’s our ‘standard’ size ad – these ads have the size of 18-point font size. (You can make the font size a little bit bigger, but it really depends on how much copy you have for your ad. The font size will dependent on how much “stuff” you’ve got to say.)
Next, we take that same copy, with the best converting font, and we start testing different pictures across that copy.
Pictures are really important. So we’ll go to somewhere like iStockPhoto.com and buy 20 pictures. Always make sure your pictures are cleared for copyright — which is why we use iStockPhoto images. And then make image ads with the exact same copy, but different pictures.
And for the pictures themselves, faces work well. People like to see faces.
In our experience, we notice that the more simplistic the ad – even sometimes slightly ‘ugly’ – the better it performs. But we still do test really crazy fonts, and really crazy pictures, and weird colors. And again, it seems to work really well when it’s really simplistic.
We’ll just have a white background, a picture, the headline across the top of the picture, and then the body copy of the ad below the picture. In fact, we rarely change the background from white. For some reason, it just does well for us.
And, of course, you can add buttons (some sort of ‘call-to-action’ button), or maybe even some sort of undulating arrows — that always helps a little (but not too much animation…just a slight amount of animation helps).
How Many Image Ads Should You Test?
Image ad testing is a constant process. We’ll just constantly test.
For example, we’ll drop in five 336×280 pixel image ads and see which works the best. And keep doing this over and over.
We’ll generally split-test about five ads at a time against each other. That’s a good number: 3 to 5 image ads at a time.
Additional Tips To Create Effective Image Ads:
Colored Borders: You can also add bold colored borders around the image ad to make it more attention-getting. A lot of red borders. That often helps. Red is the color we go to first for a test. We use a very bright and attention-getting red (NOT that brownish-red that a lot of people use for fonts in sales letters). We use the uglier bright red.
Surrounding Content: Always think about how that image ad is going to look “surrounded” by other content. You can go into the Display Network, and see where the ads are placed…then actually go to the site and look at your ad.
Figure out if the ad is “blending in” or “really stands out”. And sometimes the crazy fonts will totally stand out. And that’s what you want.
Crazy Designs: Sometimes we’ll go in and literally make these absurd ads, drawing things like happy faces over the pictures and things like that …and just ‘dumb stuff’ – and sometime those crazy designs will help boost conversion too.
Ultimately, there’s one key reason why our businesses have had done well with Adwords. And that’s because we NEVER STOP TESTING — everything from the ads to the landing page to the sales letter, to the order page and beyond. It’s just ALWAYS being tested.
Do you know my favorite part about Google Adwords? It’s the SPLIT TESTING capabilities.
It’s a great place to test headlines, specific words, URLs and identify what works in your market. And best of all, it’s quick. You can gather valuable feedback in a matter of hours.
Not only that, but Google actually rewards you for improving your ads too. The Google formula for the ranking is:
(Your bid) X (Click Through Rate).
(Your click-through-rate is the % of people that click on your ads. For example, if 20 people out of 1000 click on your ad – your CTR is 2%.)
With Google Adwords, you can literally set up hundreds of tests each week and get massive amounts of information back. Then, you transfer the data back into the other parts of your business:
– the optin page
– the sales letter
– the offer
– the email sequences
– back end product creation
Simply put, Adwords is one of the fastest place to test ideas.
There are 5 core drivers to every successful Adwords Ad. These are:
Often, you’ll find 2, 3, 4 or even all 5 of these drivers inside of just one ad.
Let’s go deeper into each one, so you can see for yourself how you can use them in your ads.
There are a total of 9 different emotions that you’ll want to test in your Adwords Ads:
And here’s an example of how to use each of these emotions in the internet marketing industry, for example:
Shame: “I’m ashamed I’ve spent so much money on courses, but I haven’t done anything.”
Guilt: “I feel guilty that I’m spending my family’s lifesavings and I’m not making any money.”
Apathy: “I feel it’s HOPELESS to make any money on the internet.”
Sadness: “I’m depressed about my financial situation and my lack of success.”
Fear: “I’m terrified I’m going to be a broke loser and work a job I hate the rest of my life.”
Desire: “There’s nothing I want more than to achieve success on the internet!”
Anger: “I’m so freaking pissed off about having lost so much time, energy and money and I still don’t have a success yet!”
Pride: “Yes I’m a success! I can finally buy a fancy car and show the world who I am.”
These are essentially the ‘emotional angles’ you will want to experiment with when crafting and testing your ads. You simply need to adapt them, so that they ‘fit’ the market you’re operating in.
A ‘benefit’ is simply what someone stands to gain from using your products or services. It can be the ultimate end result they desire, or one of the stepping stone triumphs they will enjoy along the way, as part of the overall experience.
Let’s say we were in the dating market. Here are some potential benefits that might appeal to men:
– More dates
– More dates with beautiful women
– More physical intimacy
– More respect from other guys
– Finding the woman of their dreams
– Someone to share their life with
– Getting married
– Having a family
In contrast to a benefit, a “feature” is simply the date, facts, figures and specifications that your product or service has.
For example, the features of a Porche 911 could be:
– It goes from 0 to 60 in 4 seconds
– It has 550 brake horsepower
– It’s jet black
– It has leather seats
Bear in mind, your prospects only care about a handful of these. They don’t care about all of them. And the only way, you’re going to find out which ones they care about is to TEST, TEST AND TEST.
Different kinds of “promises” you can test include:
– Guarantees: money back, double your money back, 30-day, 60-day, 90-day, 1-year, etc.
– Free shipping: If you’re selling a physical product, this might be worth testing to see if it boosts your sales conversion rates.
– Numbers: This could relate to the number of tips you have to share, how many products you have left in stock, etc.
– Minutes, Hours, Weeks, Days and Months: Time until a special offer ends, how quickly a problem can be solved, etc.
When I say “style”, we’re really talking about the overall tone and delivery of your ad.
Here are some examples of styles:
– Funny: if appropriate for your market, experiment with injecting your ads with a bit of humor.
– Professional: if you offer professional services such as legal, tax or accounting advice, then you may choose to stick to this particular style.
– Absurd: This includes off-the-wall comments, that people really don’t expect. They can sometimes help build curiosity and motivate readers to want to know more.
– Intense: This style drills right to the central facts of a situation, and states exactly what is going on, regardless of how emotionally challenging it might be of the reader. The aim is to bring the issue right to the forefront of a prospect’s mind, so they are even more motivated to do something about it.
– Advocate: As an advocate, you’re fighting in the same corner as the reader of your ad, willing them to succeed. And the language of your ad should reflect this.
– Investigator: This is where you adopt the role of the “researcher”, so your ad would reflect this impartial, investigative approach.
When first creating and testing within Adwords, we recommend you test very different ideas based on the 5 “core drivers” listed above.
This all relates to the concepts of ‘Horizontal’ and ‘Vertical’ testing we have discussed elsewhere. (Just to recap: Horizontal Testing is when you test wildly different ideas. Vertical Testing is when you test basic variations of the same concept e.g. changing out just a word or phrase).
Get really creative in crafting a bunch of different ads to test, using the guidelines above. Test vastly different ads against each other. Then when you find a combination that works really well, starting testing individual words within those ads too, to improve your conversions even more.
Here’s a real quick look at the “7 Pillars of Effective Split Testing”. These are some powerful guidelines to help you split-test effectively in your business:
1. There Are No Rules – What works in one market might alienate people in another market. Even in the same market, we’ve seen it many times where something that worked with the “beginner” section of the market then failed with the “advanced” section of the market.
2. Test Wild Things – Too many times, internet marketers are playing the “copy-cat” game. They steal each other’s headlines and concepts and they make marginal improvements, if any. Ingrain this into your mind: BREAKTHROUGH RESULTS come from testing BREAKTHROUGH IDEAS. Try adapting and testing headlines from OTHER markets, and not just the market you’re operating in.
3. Go DEEP – It’s NOT the element (such as audio or video) that creates an improvement – it’s what you put into the element.
For example, I can’t tell you that audio works or doesn’t work to improve results. It all depends on WHAT I do with that audio. So, that’s why I always split test 3 things or more for each element.
Here’s what I mean by that:
Many people are tempted to ‘jump the gun’ after testing something like video or pictures and getting negative results. They’ll go to a forum and say, “Video doesn’t work” or “Pictures dropped response.” That’s inaccurate – what is really going on is: THAT specific video or THAT specific picture lowered conversion rates.
4. Think HARD Before You Do Anything – The key to great testing is putting in great ideas right at the beginning. If you are lazy with the elements that you are testing, then don’t expect to get great results. It really is that simple.
Set aside plenty of time each week to plan your split tests, so that you’re focusing on the factors that matter.
5. Analyze Your Results – Ask yourself “Why is this working?”, or “Why is this scaring them away?” There’s always a reason ‘why’, and it’s good to figure that out if you can.
6. Keep A Record Of ALL Your Testing Results – This data can be very valuable to you over the long haul. Document your tests. Review them over time.
7. Share Testing Data With Other Marketers – Don’t hide your split testing discoveries. Instead, share them with your mastermind of fellow marketers (especially if they’re in different markets to your own). You’ll be helping them out, and they will be more inclined to share their discoveries with you too.
In today’s hyper-wired world of satellites, and internet: speed is king. I once heard it said that 1 year on the internet is equal to 7 years offline. Nowadays, I think it’s more like 10 years or more.
And the harsh reality is that, as the online space evolves with increasing rapidity, what works well on your website today may not be very effective in a few months’ time.
That’s why we’ve developed a set of “Testing Results Protocols” in our businesses – to help speed up our split-tests and make improvements more quickly.
First of all, I couldn’t stand waiting for “statistical validity”. (In a minute, I’ll explain why I think that concept just slows down the growth of your business to a crawl. I’ll also give you quite a few advanced shortcuts you can use to help speed up your testing results too).
Plus, I’ll show you some new upgrades in our split-testing software that makes setting up these tests faster and quicker too.
But first let’s look at…
The 3 Levels of Testing Results
1. Donuts (zeros)
2. Incremental Improvements (also known as “whispers”)
3. Breakthroughs (also known as “screams”)
Let’s do some math together:
0 X 10 = zero
0 X 100 = zero
0 X 10,000 = zero
0 X a million = ZERO
In other words, you can’t multiply donuts (zeros)!
Here’s a rule of thumb: If you set up a sales letter test and 300 people have seen your letter and no one has bought – STOP THE TEST IMMEDIATELY.
Basically, it means that your sales letter is not working, and it’s time to move on. (Personally, I would even stop the test after 150 people saw the letter because I want to find a winner as quickly as possible.)
This is not the time to get despondent either. It’s actually great news. Because you want to find out what’s NOT working as quickly as possible. That way you can move on to the next test and find the next big breakthrough.
NOTE: There are exceptions to this rule… For example, when you’re selling expensive equipment, courses or packages. If you’re selling a $10,000 piece of construction equipment – then maybe a 0.3% conversion rate is acceptable.
However, if you’re selling anything below $100, as a general rule, don’t accept conversion rates below 0.3%.
Rule #2 is similar to rule #1. The bottom line is: you need to gather information and results quickly.
Why? Because you’re wasting time if you’re not aggressively looking for the ‘big breakthroughs’ in your business.
Let’s take some advice from the world’s greatest investor, Warren Buffet: Mr. Buffet makes investments based on ‘opportunity cost’. This means he doesn’t look at an opportunity and think “I can make 20% return if I invest 1 million dollars in that company.”
Instead, he looks at all of the other opportunities he’s “missing out on” by NOT investing the million dollars somewhere else.
And here’s how this applies to your split-testing strategy:
Scenario #1: The “normal” way
Let’s imagine you set up a split test. And the 2 factors are initially producing very similar results. They are “nose-to-nose”. But after 60 days you finally identify a clear winner that gives you a 10% improvement in your conversion rate. That’s not bad.
However: what did you LOSE by waiting so long for statistical validity to kick in?
Scenario #2: The “faster” way
You set up the same test as in ‘scenario 1’. However, you abandon the test after 1 week because it wasn’t a ‘breakthrough’ result – in other words, the results were too close and the test failed to produce a clear winner.
Next, you set up a new test, and you get a 25% improvement in your conversion rate after only 2 weeks. And then you set up another test and you get another 15% improvement after 2 weeks.
So with this strategy, you achieved a 40%+ improvement in 5 weeks, instead of the 10% in 8 weeks.
The point to get across here is this: by abandoning tests that are “nose-to-nose” quickly, you’ll find the ‘big breakthroughs’ faster.
This is a concept I stumbled upon by being impatient. Basically you want to look for big results fast.
The Big Breakthrough Formula:
You have a ‘Big Breakthrough’ when…
The Winner – Square root of the winner > the losing element
The big breakthrough formula is simple… You subtract the “square root of the winner” from “the winner”, and the number you’re left with should be BIGGER than “the loser”.
This will become a lot more clear in the example below. It sounds more complicated than it is…
Example of the Big Breakthrough Formula in action:
Let’s say you’re doing A-B split testing, and element “A” has 20 sales and “B” has 12 sales.
First, calculate the square root of “20” – which is 4.47. Now, simply plug-in the rest of the numbers:
20 (winner) – 4.47 (square root of 20) = 15.53
The next question is: Is 15.53 a bigger number than 12 (the number of sales of the “losing” element)?
In this case, the answer is ‘yes’, so we have a ‘big breakthrough’ result, and can move on to testing something else against this winning element.
Frankly, is this method perfect? Of course not. I would even say, by picking winners quickly – you’ll be wrong 10% of the time or so.
In the grand scheme of things, is this important though?
I’ll take SPEED over accuracy any day of the week. Even if I am wrong 20% of the time, the increase in my conversion rates will quickly make up for it.
The reality is, the results will self-correct over time as you test more variables. For example, let’s say I end a split test early and I was wrong about the winner… I’m going to set up a NEW test – and hopefully I’ll crush the “wrong winner”. And I’ll set up another test and beat the champ again… and again… and again…
To recap: focus on achieving BIG breakthroughs with your testing. Don’t be afraid to end a test early. You do NOT need 200 results to have a valid test. I think that’s just insane (especially on a sales letter).
There are 2 main phases to testing:
1. Horizontal Testing
2. Vertical Testing
Most marketers aren’t aware of the differences and mainly use vertical testing in their marketing.
So what’s the difference?
Vertical testing is when you go DEEPER with 1 idea. Here’s an example of a vertical test:
Version 1: The 3 Secrets to Great Titanium Clubs
Version 2: The 3 Factors to Great Titanium Clubs
Version 3: The 3 Elements to Great Titanium Clubs
So the core sentence is: The 3 _________ to Great Titanium Clubs.
The vertical element that you’re testing is: secrets vs. factors vs. elements.
On the other hand, horizontal testing is different.
Here’s an example of a horizontal test:
Version 1: How to Smash Balls 300 Yards
Version 2: What Every Golfer Should Know About Great Golf Clubs
Version 3: The 3 Elements to Great Titanium Clubs
So, here we’re testing totally different IDEAS (horizontal testing) versus different VARIATIONS of the same idea (vertical testing). The ideas that you’re testing are: “how to smash balls far” vs. “what golfers should know about clubs” vs “what makes great titanium clubs”. What you’re doing at this phase is looking for the “magic button”.
Now, let’s say that “What Every Golfer Should Know About Great Golf Clubs” is the winner… Then, you can start doing vertical testing around that winning headline. For example:
Version 1: What Every Golfer Should Know About Great Golf Clubs
Version 2: What You Should Know About Great Golf Clubs
Version 3: What Great Golfers Know About Great Golf Clubs
So in this vertical test you are testing: “what every golfer should know” vs. “what you should know” vs. “what great golfers know”.
Are you totally clear now about the difference between ‘Horizontal Testing’ and ‘Vertical Testing’? As a quick recap, Horizontal testing involves testing completely different ideas, while “Vertical Testing” involves testing different variations of the SAME idea.
So, the big question now is:
For the first 3 months to 6 months, 80% of your split testing testing should be horizontal. Why? Because you’re looking for the marketing building blocks that you’re going to build your business upon.
You’re searching for the “The Golden Idea” that works best in your market. And the fastest way to find “Golden Ideas” is to test a lot of different ideas.
Then, once you’ve found a great idea that’s converting well, it’s time to move on to Vertical Testing – testing every possible variation of that idea, to make it even stronger.
At this stage, you should reverse the testing ratio to 80% vertical testing and 20% horizontal. So you’re still testing new ideas (because there will always be a better idea somewhere). But you’re mainly focusing on running and refining the winning idea you already have.
99% of marketers think of Adwords as one of the best places to get traffic. And they’re mostly right.
However, in my opinion the ULTIMATE VALUE of Adwords is the speed and ease of its split testing power.
Quite simply, Google Adwords allows you to test lots of ideas and concepts in your market, so you can really understand what your target audience is looking for.
Here’s 2 analogies…
Imagine if I told you there was $100,000,000 buried “somewhere” on planet earth. Would you start investing your time, money and resources just digging “anywhere”?
Or would you first identify the exact location of the cash before beginning to dig?
You’d want to know the location, right?
It’s the same thing with oil companies. They do their research and FIND the oil patch before setting up the oil rig.
What’s funny is, most marketers invest hundreds of hours digging for the million dollar treasure without KNOWING if there’s any hope of striking gold. What I mean is, they create sales letters, order pages, dozens of autoresponder emails BEFORE knowing if their business idea is going to work.
Let me be the first to admit that I’ve been guilty of this in the past too. But I approach things differently now. Here’s how I create a new marketing process today…
What is organic marketing?
It’s the natural process of creating and adapting the sales process around what your market is telling you day by day. In other words, you listen to what your visitors, prospects and customers are telling you. Then you use that feedback to improve your sales process.
Again, it goes back to the attitude that “I don’t know anything. And I’m going to let the market tell me what they want” vs. the ego-centric attitude of “I know what they want, but really I’m taking some wild guesses and hoping it works.”
So, what I suggest is this: DON’T spend countless hours building what you think is the “perfect marketing process”. Instead, start with almost nothing: 1 or 2 autoresponder messages, a simple sales letter, and a basic opt-in page.
Then start testing in Google Adwords. Set up at least one new split test every single day. And use the results to attune your marketing and sales process with what your market is looking for.
Just so you know, Timothy Ferriss – the best selling author of the blockbuster “The 4 Hour Workweek” (which I highly recommend) – found his “magic button” book title through Adwords testing. He probably tested several different numbers: 4 hours vs. 10 hours vs. 20 hours vs. 40 hours, as well as other titles too.
And he let the results guide him on what to name his best-selling book. Now, imagine if he had done it the other way around. He could have written a book, given it a title that he thought was “hot”, spent $10,000 printing them and putting it out there, only to discover nobody resonated with the title. What an unnecessary waste of precious time and money.
Instead, follow his example, and use Adwords to help you find the hottest triggers, concepts, titles, and headlines for your target audience.