I love to snow ski but here on Long Island, getting to the mountains isn't the easiest of trips. This tends to leave large gap in ability between those who ski once or twice a year and those who are lucky and dedicated enough to spend more time on the moutain.
The result is an interesting phenomenon where a large number of beginner or intermediate snow skiers inevitably steer the conversation toward how frequently they or you fall while on the mountain. It's their litmus test to decide how good a skier somebody is. Someone who doesn't fall much is a good skier, and vice versa.
I'm a good skier. I may not be the best but I can hang with anybody and ski anything (shy of what you see in the movies - my cliff-hucking days are long behind me, if they ever existed at all), and I fall often. In fact, the best skiers I know are those that fall the most often, not the least. So when this topic comes up, my response is always the same, "if you're not falling, you're not learning."
The point is that it's easy to do the same things every day and get better at those things over time. But while you're trying not to fail, others are leap-frogging you because they're pushing their limits, trying new things, falling, getting back up, trying something else, and improving exponentially with every iteration. You may end up with a few bumps and bruises along the way, and it might hurt a little in the short-term, but the long-term rewards are well worth the short-term sacrifices.
Which brings us to our topic: Testing.
It's tedious. It's time consuming. It can be frustratingly slow, and it can cost you in the short-term. Yet it's the most important thing you can do to drive consistent, incremental improvements in every aspect of your business.
If you're not familiar with the concept of A/B testing, it's been used in the advertising industry for years. The basic premise is to produce two pieces of nearly identical content with a single change to one version that you believe will result in a better performing ad, and then measure the results. In advertising, frequently that means trying two different headlines, calls-to-action, images, or placement of any or all of the above.
While a lot of SMBs aren't familiar with this concept, even fewer are aware of how widely you can apply A/B testing throughout your business.
The underlying goal of A/B testing is constant, incremental change that increases organizational knowledge and provides continuous improvement. This is achieved by continually testing your assumptions and challenging the status quo; making changes where necessary and refraining from changing something that already works well.
Ok, that's a lot of conceptual talk, how can you apply this to your business or your job?
The most important thing to understand is that this concept works best when it's ingrained in the culture of your organization. Testing shouldn't just be something that happens every once in a while when you decide you want to try something out. It's something everyone in your organization should understand the value of and be empowered to implement.
For most businesses, advertising and marketing is an easy first step on the road to becoming a testing maniac. Digital advertising makes it very easy to A/B test, and most (if not all) modern digital ad platforms give you the ability to easily test multiple versions of the same ad. For example, Facebook advertising allows you to build a single ad and test up to five different images at a time, or to simply copy an existing ad, make a simple change, and run the two side-by-side in the same ad group for as long as you need to determine a winner.
In your ads, test everything you can: Headlines, calls-to-action, images, descriptive text, display URLs, etc. Just remember that you need to measure your results against a goal. If that goal is traffic to your website, most clicks wins (although it should be noted that this is the most basic measure of performance and probably one of the worst). A better option is to set your goal as quality traffic to your website and then determine what that means to you. Maybe it's a user that visits four or more pages because you make money by selling ad-space on your site.
If you want users to fill out a form, join your mailing list, or buy something from you, then your goal is conversions (a four-page visit could also be considered a conversion, but that's just semantics). Conversions have gotten much easier to measure performance against, and again, all modern digital ad platforms have built in conversion rate measurement and testing options. This metric is far better because you can build an ad that drives a lot of traffic at a very small cost; but if it's not well-targeted traffic, they'll leave your site as soon as they arrive. Well-targeted traffic might cost you a bit more per click, but the resulting increases in conversion rates should be worth it.
On the other side, quality landing pages on your website can do everything from lowering your cost-per-click (CPC) to increasing conversion rates, so it's important to test those too. Maybe you want to test an intro video against a static image and some text, or requiring more information to download your whitepaper to see if you can capture more information about your prospects without lowering your conversion rates. Here again, call-to-action and headline placement and copy can also be tested and optimized.
The same goes for email. One of the reasons I've been pushing my clients to MailChimp from Constant Contact is the extensive built-in testing functionality. Test subject lines, images, image placement, calls-to-action, copy, and design to find the optimal way to present your information. If your mailing list is large enough to support it, the best way to do this is to send your test emails to a small subset of your list, then send the winner to everybody else.
It's important to understand how powerful this can be. If you have 15,000 subscribers and you can increase click-through rates by just 1% by writing a more compelling subject line, 150 more people just performed the action you wanted them to. Then, if you've optimized your landing page and again convert just 1% more of your visitors, that's 2 people that buy from you that never would have before (not to mention converting the additional 1% of those that would have clicked through your email anyway if you're following the math correctly ;-).
This tactic isn't just limited to digital advertising. Print ads can just as easily be tested. If the publication you're advertising with won't allow you to run two different ads in the same issue (most won't), get a better rate by signing a longer term contract and test them one after the other (being sure to account for any differences in season, of course). It's not as immediate a result as digital, but "get rich quick" is a myth and this stuff takes time and effort to get right.
Where this really gets fun (yes, I'm a huge dork, I actually love this stuff!) is when you start experimenting and testing stuff that's a little more outside the box. Do you have trouble getting through your email every day? Test a new approach to managing your inbox for a week and measure your productivity. Manage a retail store? Test product placement and see if you can find ways to draw customers to higher margin items or impulse purchases at the register.
Run a manufacturing facility? Study your process to identify bottlenecks then design and test alternative solutions. Find that you consisitently don't hire the best candidates? Test changes to your screening and interviewing process and see which employees last longer or produce better results.
You can apply this anywhere efficiencies can be gained or results can be improved. If you're having a hard time figuring out where you can apply this to your business and you need ideas, feel free to reach out to me anytime.
Last updated on