Lately I have been using Google blog search to find inspiration points for my new posts. It has been an excellent system for giving something back to the blogging community while getting a small amount of exposure in the process.

I quite like the interface from a browsing perspective. It is very easy to find clusters of stories on a single topic. Hopefully they will start incorporating this clustering into the search results. (blog and organic)

Regardless of how you run your tests it is important to get data on each CTA. This can save a good deal of time/money and lead to some interesting conclusions.

Example:
You are running A/B tests on a landing page with 4 call-to-actions. The element in your test is in close proximity to one of your CTAs. Under normal circumstances you may need to drive…say 10,000 visitors before making your conclusion on aggregate data. But, if you segment the conversion data to each CTA you may only need 5,000 visitors to make a conclusion based on the more refined data.

This would allow you to run more tests, faster, on a lower budget…and if you can’t draw an early conclusion you still have aggregate data to fall back on.

I often wonder how much time developers waste on a daily basis tracking non-actionable data. In the past I spent so much time keeping an eye on all kinds of metrics that didn’t have any impact on how I designed or promoted a website. Even today I track various “ego metrics” just for the sole purpose of making myself feel better about the work I’m doing in the trenches. However, I am much more time conscience when looking at this data. In a way it almost makes me feel guilty.

Examles of “Ego Metrics”

1. Google Pagerank - In a way I have to say visible pagerank is a good thing. Sure it is the root of a lot of common SEO missconceptions. At the same time how many webmasters are absolutely enfatuated with their pagerank? The collective time lost by people watching this simple 1-10 number has, at some level, decreased the online publishing competition.
2. Hourly/Daily Visitors - I do think it is important to keep track of the raw traffic coming to your website but you don’t need to track this at a gradual level. Be honest, if you drove in 1,000 visitors yesterday and 800 visitors today what action could you possibly take? It might be worth investigating if your overall traffic drops to a certain threshold but you don’t need to keep your finger over your traffic pulse.
3. Alexa Rank - Let’s put aside the subject of accuracy when it comes to Alexa rank and focus on the more important question. Why does it matter how much traffic you generate in relationship to all the other billions of sites out there? Knowing where you stand could act as an inspiration point to make you work harder but aside from that it’s not really an actionable data point. I find data from Compete.com much more usable as you can more easily compare sites side by side.
4. Backlinks - Just to make this clear, I’m not suggesting backlinks aren’t important. They are absolutely vital to your SEO campaign. I also believe this data is quite actionable. My point is, typing in “link:yoursite.com” into a search engine will not help you increase the number of links. So don’t check this on a daily basis. Cut down so you are only checking on a monthly basis and spend that extra time actually building links.
5. Adsense Daily Earnings - I used to spend so much time looking over earnings metrics and to be honest I didn’t really use that data to make major impacts. Sure you want to optimize to drive your eCPM up but you don’t need to track this on an hourly basis.

This isn’t an exhaustive list because I really don’t want to dwell on these metrics. They really don’t matter when it comes down to it. Forget about this stuff and spend your time developing content. Make sure the metrics you are spending time on are actionable.

John from PPCHero posted an excellent set of landing page tips centered around the idea of simplicity. His #1 tip “Keep call-to-actions above the fold” is common sense advice designers seem to break 60%+ of the time.

I especially liked his analogy comparing a landing page to a river. The rocks in the stream are compared to barriers which naturally divert visitors away from the end goal of becoming converted visitors. This reinforces an idea from one of my previous posts, simplify your landing page by removing your sidenav.

He also mentioned an idea that has been taking off in the design community over the last several years. Use white space to draw focus to key elements. I couldn’t agree more. A well designed element floating in a sea of beautiful white space will not get overlooked.

One area of simplicity that very commonly gets overlooked is the length of your ad copy. Why does it require 1,000 words for so many landing pages to get across the idea that you should “buy a widget”? I have seen studies of longtail ad copy performing pretty well. Rand Fishkin gave us a nice example of a “scroll forever” page during Pubcon 2007 in Las Vegas. Still, I can’t help but think that if you test long enough your ad copy will progressively get shorter and shorter.

David Lindop of Setfire Media offers 7 practical landing page tips. His tips are a mixture of ad copy and design. I’m not a copywriter so this post will focus primarily on the design aspect.

1. Add by subtraction: simplicity sells - Take a look at your landing page on an element by element basis and ask yourself “Which elements do not promote my call-to-action?” Then promptly remove those elements all together. Distractions are ROI killers. You don’t really need a side navigation that could very easily derail your potential customers.

2. Contrast Hierarchy - This is a design principle I’ve discovered over the past several years that I strongly believe in. Basically, take all the elements on the page and place them in an order of precidence. The most important element (call-to-action) should have the highest contrast while the less important elements should have the lowest contrast. I’m not suggesting that boilerplate content should be barely visible or that there should be noticable steps between the levels of precidence. I simply believe that your most important elements should have the highest contrast.

3. Warm/Cold Colors - I find this to be an very effective technique when it comes to drawing attention to your call-to-actions. It’s simple, if your design uses cool colors (blues/greens) then you should use warm colors in your money elements (reds/oranges). Likewise, if you have a warm design your important buttons should use cool colors.

4. Multiple call-to-actions - I wouldn’t go overboard cluttering the page with buttons and forms. Still, if you have a long sales pitch it is probably a good idea to embed appropriate call-to-actions at breaking points.

5. Testimonials - I mentioned earlier that it is a good idea to drop your sidenav. Testimonials work as excellent navigation alternatives. They keep users focused, they can add nice contrast in the design and most importantly they add credibility.

6. Keep it manageable - This principle applies to quite a few potential elements. In #5 I suggested replacing your nav with testimonials. One reason for this is you don’t usually want your text to span the entire width of the page. It’s true that a high volume of text can be read faster with wider columns but you run the risk of making the content appear daunting. Blocks of content should not be too wide or tall.

You also want to make sure your forms are manageable. Nobody wants to fill out 25 fields of personal data just to read your white paper.

Just keep things simple and continually ask yourself “Is this too much?” or “Could I get the same idea across in fewer words or by using simpler diagram?”

7. Use bullet points - Ok, so this idea could have been packaged into #6 but I think it is important enough to warrent its own bullet point. If you bulletize your content it makes it so much easier for users to scan through to pick up the info they want/need and ignore the rest. Be honest, have you read all of my bullets up until now?

8. Split test macro - I know you want to launch your landing page and immediately start testing every small detail but this can be a big mistake. Always A/B test large elements first before you start worrying about sub-elements. I highly suggest you start your testing with entirely different landing page versions.

9. Multivariate test micro - Once you have tested the large elements to get a general idea of what works and what doesn’t only then is it time to get your hands dirty. Feel free to test any micro element even if you don’t think an alternative will yield better results. You might be surprised at what you find.

10. Budget your testing - If you have a small advertising budget and you drive a low volume of traffic, it absolutely doesn’t make sense to run complicated MV tests. Just as a general rule of thumb, I would like to get a minimum of 10 conversions for every square in your MVT matrix. So if you have 2 headers and 2 call-to-actions that makes for 4 possible combinations and you want at least 40 conversions before moving to your next test iteration. The type of conversions you have impacts this. If you are selling high end products and driving a .05% conversion rate then you might want to be a little more liberal than a simple subscription conversion that is hitting at about 10%.

11. Test in iterations - You’re not going to hit the bulls eye on your first series of tests. Learn from your data and systematically make improvements until it becomes no longer cost effective to do so.

Ian Lurie suggests that it is imperative to incorporate your creatives into your MVT matrix. I do believe this can be a good idea if you are running a high budget PPC campaign. The only problem I see with this is I usually like to build a relatively high number of landing page specific creatives. It could be a huge pain to incorporate all your creatives into your multivariate test. You may also run into bidding problems going this route.

For example, if you are paying $.25 per click for creative A and $.10 per click for creative B and creative A results in 5% higher conversions…is it really worth it to push creative A and drop creative B? Maybe the keywords for creative A drive 100x the search volume as creative B. I still don’t see that as justification to drop creative B. Instead this would be an indicator that you should try to optimize creative A to see if you can create a cheaper or more cost effective variation.

Now given my above example what would you do if you are targeting creative A and creative B without plugging them into your MVT matrix? Well you would probably make the exact same conclusion even if you think creative A might be driving 5% fewer conversions because cost and volume are factors you shouldn’t overlook. In PPC it is good to maintain a high volume of landing page specific creatives targetting a wide array of long tail keywords.

I’m not suggesting Ian is wrong by any means. The more data you have the better you are equiped to make the best decisions when it comes to optimizing your landing pages. I just think you need to factor in the setup time and the size of your campaign before feeling obligated to use creatives as test elements.

A/B testing in terms of web development is traditionally thought of as a function of a landing pages. This is largly because campaigns can run for weeks, months or even years (time is on your side) and most landing pages should have straight forward, measurable call-to-actions.

On the flip side, email campaigns are very rarely thought of as A/B testing material. Most often you blast out an email campaign to a list one time. You can’t refine your email and resend it to the same list to see if it will be more effective…it just doesn’t work that way. If you have a large list and your running a campaign that isn’t time sensitive you can run a proper A/B test by sending your emails in a series rather than all at once but this seems to be a rarity.

Kristina Knight from Biz Report suggests that when you launch an email campaign you should use A/B testing as a means to progressively learn how to write effective ad copy. For example, you might find a term or phrase to be marginally more effective than a synonym. I generally agree with this approach. It’s great to learn as much as you can from your past work in order to enhance your future work. Still, I think it is best to run email marketing A/B tests in a series in order to get some short and long term value from your testing efforts.

This page is a simple collection of A/B testing articles from various sources.

1. 11 Adwords Landing Page Design Tips
2. MVT With Ad Creatives
3. Email Campaigns in A/B Tests
4. Simplicity Sells
5. Independent CTA Tracking