How to Double your Conversions in 2017

by Guest Contributors
  • You don’t really know what works until you test and use the results of the tests as input for the next iteration.
How to Double your Conversions in 2017
FM
Join our Telegram channel

Yes, we should all really go to the gym more. And eat healthy. Spend more time with the family. New Year’s resolutions aplenty, but I’d really like you to add another one to the list. This one is to do with business, and to be more precise, with doubling your business.

[gptAdvertisement]

That sounds overly optimistic, but it really isn’t. If you make this year about testing and optimizing, doubling or tripling your business in 2017 is doable. A lot of brokers rely on the opinion of the highest paid person. This image looks nice, or that headline fits the brand. But you don’t really know what works, until you test and use the results of the test as input for the next iteration, ad infinitum.

The Importance of Testing

Testing and optimizing is a way of doing business more than it is some task you take care of though. It touches not just what colour button you put on the landing page or the adwords ads you have in rotation. It can be as broad as testing a new platform or a new sales process. Rather than doing what seems right, the mantra for every part of the organisation needs to be ‘good suggestion, let’s test this’.

To unlock the Asian market, register now to the iFX EXPO in Hong Kong.

When you test something in parallel against something else, something magical happens. You have information. You know based on hard facts what works and what doesn’t. You can do more of what works and less of what doesn’t.

But how you test is just as important as testing in the first place. You need to make sure the only variables that change are the things you want to test. So, first testing one process and later on testing another, won’t work. They need to be tested at the same time, with the same audience. What’s more, you need to look at the appropriate KPI for that test and not one that can be influenced by other elements.

To make this less abstract, imagine you are testing two pages. You’ll want to send similarly sourced traffic, in the same quantity, to both at the same time rather than testing one landing page one week and the other another.

Use the Right KPI for the Right Test

You shouldn’t use adwords traffic for one landing page and display traffic to another, they need to be the same type of visitors. You should also use the right KPI for the right test. So instead of looking at the amount of visitors to that page as the KPI, you will want to look at conversion rate.

And if you are testing landing pages, you shouldn’t add another variable into the test by using different banner creative for both pages, since this will also have an influence on landing page performance. Lastly, you will need to get enough actions in the sample to make a statistically valid conclusion. So you do not make a decision based on 3 conversions, instead you wait with further changes until you have at least 40 in order to ensure the test results hold up.

People in advertising are used to looking at numbers to make decisions, especially when the focus is online. A lot of the time though, basic testing procedure is not followed because it takes too long and requires keeping other variables constant. Outside of online advertising, testing is sparsely done, not even in the wrong way.

And that’s a shame because a lot of the gain to be had lies outside of the front-end of Marketing . So besides testing online creative material or webpages, brokers should test sign up flows, sales procedures such as when to call and what to say, email flows, the platform layout and its prompts and workflows, pricing and products and so on.

Anyway, it’s not even about specific things to test. It should be in your company DNA to test and optimize. When you start testing everything, and making decisions based on the awareness, Acquisition , conversion and retention outcomes instead of the highest paid persons’ feelings, you will continuously improve. In that light, doubling your conversions seems pretty doable, even easy.

This article was written by Bart Burggraaf, Partner at MediaGroup Worldwide

Yes, we should all really go to the gym more. And eat healthy. Spend more time with the family. New Year’s resolutions aplenty, but I’d really like you to add another one to the list. This one is to do with business, and to be more precise, with doubling your business.

[gptAdvertisement]

That sounds overly optimistic, but it really isn’t. If you make this year about testing and optimizing, doubling or tripling your business in 2017 is doable. A lot of brokers rely on the opinion of the highest paid person. This image looks nice, or that headline fits the brand. But you don’t really know what works, until you test and use the results of the test as input for the next iteration, ad infinitum.

The Importance of Testing

Testing and optimizing is a way of doing business more than it is some task you take care of though. It touches not just what colour button you put on the landing page or the adwords ads you have in rotation. It can be as broad as testing a new platform or a new sales process. Rather than doing what seems right, the mantra for every part of the organisation needs to be ‘good suggestion, let’s test this’.

To unlock the Asian market, register now to the iFX EXPO in Hong Kong.

When you test something in parallel against something else, something magical happens. You have information. You know based on hard facts what works and what doesn’t. You can do more of what works and less of what doesn’t.

But how you test is just as important as testing in the first place. You need to make sure the only variables that change are the things you want to test. So, first testing one process and later on testing another, won’t work. They need to be tested at the same time, with the same audience. What’s more, you need to look at the appropriate KPI for that test and not one that can be influenced by other elements.

To make this less abstract, imagine you are testing two pages. You’ll want to send similarly sourced traffic, in the same quantity, to both at the same time rather than testing one landing page one week and the other another.

Use the Right KPI for the Right Test

You shouldn’t use adwords traffic for one landing page and display traffic to another, they need to be the same type of visitors. You should also use the right KPI for the right test. So instead of looking at the amount of visitors to that page as the KPI, you will want to look at conversion rate.

And if you are testing landing pages, you shouldn’t add another variable into the test by using different banner creative for both pages, since this will also have an influence on landing page performance. Lastly, you will need to get enough actions in the sample to make a statistically valid conclusion. So you do not make a decision based on 3 conversions, instead you wait with further changes until you have at least 40 in order to ensure the test results hold up.

People in advertising are used to looking at numbers to make decisions, especially when the focus is online. A lot of the time though, basic testing procedure is not followed because it takes too long and requires keeping other variables constant. Outside of online advertising, testing is sparsely done, not even in the wrong way.

And that’s a shame because a lot of the gain to be had lies outside of the front-end of Marketing . So besides testing online creative material or webpages, brokers should test sign up flows, sales procedures such as when to call and what to say, email flows, the platform layout and its prompts and workflows, pricing and products and so on.

Anyway, it’s not even about specific things to test. It should be in your company DNA to test and optimize. When you start testing everything, and making decisions based on the awareness, Acquisition , conversion and retention outcomes instead of the highest paid persons’ feelings, you will continuously improve. In that light, doubling your conversions seems pretty doable, even easy.

This article was written by Bart Burggraaf, Partner at MediaGroup Worldwide

!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|} !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}