Gambling on hunches when deciding which tactics and channels will help you achieve your goals is a flimsy strategy that could land you on the wrong path.<\/p>\n
The same mistakes are made far too often. Marketers try a channel once, get a disappointing result, and decide the channel doesn\u2019t meet their needs. Or after the channel fails to produce, they continue to plug away at it forever, assuming things will eventually work out.<\/p>\n
Good channels don\u2019t always reveal themselves right away, and bad channels don\u2019t turn into good ones after meeting a nonexistent spending threshold. To avoid early abandonment and wasted budgets resulting from a prematurely folded hand, employ small marketing tests to evaluate channels across multiple audiences and creative approaches before the next round of betting.<\/p>\n
The most successful marketers analyze new marketing channels according to not only their current performance but also their growth potential.<\/p>\n
si_guide_block id=\"34016\" title=\"Download\u00a0our Premium Resource \u2013 Evaluating Content Marketing ROI\" description=\"This guide is aimed at helping you improve your confidence in the value of content marketing by stepping you through a range of techniques to help marketers evaluate and prove their content effectiveness.\"\/]<\/p>\n
By looking at what a channel could become rather than what it is doing right now, you can base decisions on validated strategies instead of short-term numbers.<\/p>\n
In that sense, marketing and poker have even more in common. In both games, you'll lose if you never bet, but you'll also lose if you go all in on the wrong opportunity. The trick is to place small \"test\" bets, analyze the results, and push forward on the most promising plays.<\/p>\n
Let\u2019s say you\u2019re in the middle of testing one creative asset on one audience and things are going well. Because that small bet is indicating an opportunity for bigger gains, you want to double down on that channel and add new audiences and creative content.<\/p>\n
On the other hand, if one channel continues to disappoint despite using multiple creative assets, targeting multiple audiences, and A\/B-testing multiple combinations, drop the channel and move on. ROI will not eventually show up in a channel that exhibits no potential. Nonetheless, don't look at your experiment as a failure \u2014 you've\u00a0gained deeper insights\u00a0about your audience<\/a>.<\/p>\n Evaluate the success of each channel based on the opportunity it represents. Does this channel have growth potential for the next month or the next year? Did the test succeed because the content referenced a current event, or will evergreen appeal replicate the success across more tests?<\/p>\n Short-term wins matter less than long-term opportunities, so keep analysis periods long (at least one month) before deciding what to do next.<\/p>\n We tried Facebook ads in the past, for example, and gave up because they weren\u2019t producing the ROI we wanted. Our director of marketing, however, still saw potential for us on the platform. He spread $1,000 across five content tests over three months to see whether we had missed a winning combination.<\/p>\n As it turned out, he was right. Today, our Facebook strategy provides a dependable pillar of growth. We target a combination of lookalike audience and entrepreneurial interest, sticking to Facebook lead ads over normal ads. This strategy has reduced our lead acquisition costs by 80 percent. None of that would have been possible if our marketing director hadn't tested different content before we moved on.<\/p>\n Testing the success of a marketing channel isn\u2019t easy. Uncontrollable social factors can change short-term messaging success, but you don\u2019t have to wait a year before rendering a verdict on whether to invest in a channel. Follow these four steps to weed out bad investments and identify channels that will provide the greatest ROI.<\/p>\n Establish a measurable goal, and create a benchmark against which to measure your progress. Collect current data and measure regularly throughout the test to see whether your efforts are changing the numbers and, if so, which changes are the most impactful.<\/p>\n At Ladder, we set a goal to increase conversion rates in the below-the-fold section of our website. We benchmarked our performance at 0.75 percent and then decided to shoot for a 1 percent rate. Our designer moving our client logos and social proof section below the fold resulted in a massive uplift of 1.25 percent CVR and even boosted conversion rates in our hero section.<\/p>\n Without a hypothesis, you risk measuring the performance of your test against the wrong metrics. Set your goal, and then develop a hypothesis of what you expect your test to achieve and how it will achieve it.<\/p>\n The best hypotheses start out broad and get narrower. Begin with a general idea, establish how the test will operate, and then make an educated guess about what will happen in the form of an if-then statement.<\/p>\n Say you want to try to improve conversions from ads on Twitter by using a blog post as the content. Your baseline Twitter conversions are at 1.2 percent, and you estimate that your improvement will push that number to 1.5 percent. With those facts, your hypothesis becomes, \u201cIf I use this blog post in my Twitter ads, my conversion rate will improve to 1.5 percent because the content in the blog encourages readers to buy our product.\u201d<\/p>\n Report on performance weekly to keep track of tests and to create data points to evaluate once the test ends. Don\u2019t overreact to one good week or one awful one. Let the full test play out, and then evaluate the total results at the conclusion of the experiment.<\/p>\n At the end, ask questions about the impact of the tests, the accuracy of the hypothesis, and cause of the results. If you set out to increase Twitter conversions to 1.5 percent but conversions fell below 1 percent, don\u2019t get frustrated. Learn from the experience and\u00a0design a new test with a new hypothesis.<\/p>\n According to a survey by Epicor,\u00a0more than 46%\u00a0of CFOs admit to relying on instinct<\/a> when making important business decisions even when available data could help them because it's too tedious to access. Don\u2019t make such avoidable mistakes. Listen to your data, and let the results of your experiments guide your next move.<\/p>\n With the experiment complete and the results tabulated, look around the industry to see how other companies are doing. According to our own research,\u00a057% of marketing experiments fail<\/a>, so even if you didn\u2019t get the results you wanted, your data still provides valuable information on your position in the market.<\/p>\n To illustrate: If you set out to improve your cost per click on Facebook, how did you do against the\u00a0global average of 48 cents<\/a>? If you planned to get more exit-intent email captures, are you above or below the average? Look for these numbers, as well as industry-specific figures, to see where you have room for improvement.<\/p>\n Small tests turn hunches into data and data into viable strategies. Follow these tips to identify and analyze the potential of new marketing opportunities.<\/p>\nBalancing out<\/h3>\n
How to decide whether you should go all in on a channel<\/h3>\n
1. Know what you're trying to change<\/h4>\n
2. Build a hypothesis that centers your tests<\/h4>\n
3. Report on everything all the time<\/h4>\n
4. Compare against the industry<\/h4>\n