Hey,

Last month, we ran a user survey (by the way, congrats to the winner of the $500 gift card!). One of the questions we asked was how your performance is evaluated.

Well, 63% mentioned customer migrations as part of their evaluations, 58% the uplift they generate, and only 48% mentioned channel metrics as part of their professional evaluation.

The numbers are fairly similar to those from the survey that ran at the beginning of the year, showing that most of your peers are looking beyond vanity metrics, such as opens and clicks, and into true impact on customer behavior.

This month’s magazine is dedicated exactly to that. Helping you measure what matters: the impact you have on your customers’ behavior.

So, buckle up, and if you have comments or feedback, reply to let me know.

Cheers,

Shai Frank, VP of Product

Why “Response Rate” should be your lighthouse

Determining how well a campaign is performing or the winner of an A/B test shouldn’t be done by looking at how many customers clicked a link on an email, but rather by how many customers performed the activity you wanted to incentivize.

Created a campaign to increase revenue? The difference between the revenue from customers in the test group to the ones in the control group should tell you if you succeeded. That is why Response Rate, the rate of customers that took the desired action during the measurement period, should be your lighthouse when navigating the marketing performance seas.

See how Optimove uses Response Rate when calculating the winner of an A/B test or a Stream by clicking here.

Leaving outliers behind

Data is messy by nature; we want to help make it less so. We all know averages can be misguiding, so in Optimove, you can also look at medians. Similarly, response outliers can skew your campaign analysis. All that is needed is a few ultra-eager customers (especially those in the control group).

When analyzing campaigns, if you’d like to review your results without outliers, remove them with the new “Exclude Outliers” functionality. Note that not all campaigns have outliers and that in some instances you will want to keep them as part of your results to view the complete picture.

Going beyond a single campaign

Of course, you don’t live in a world where there is only one campaign running at a time, so knowing how to attribute results to a series of occurrences of the same campaign, a sequence of campaigns, or a group of different campaigns is paramount.

For a recurring campaign just go into the campaign analysis page and toggle “Series Analysis.” You’ll be able to see cumulative results and, of course, remove outliers if they exist.
For different campaigns, use each one’s “increase in” metric to learn the unique marginal contribution each one had and your marketing plan’s “increase in” metric for the uplift of all your efforts. Read more about uplift and attribution here.
For sequences of campaigns useStreams, letting you test an entire sequence against a super-control group or run sequence vs. sequence experiments that will let you know which one has the highest impact on your desired KPIs. For example, test two alternative welcome series to find the one that yields better New-to-Active migration rates

Simplifying the creation of Streams

That last point is a crucial one. As a marketer, many times you will think in terms of a customer journey. Streams is the only tool out there that allows you to accurately measure the impact specific journeys have on customer migrations (for that 63%) or uplift on KPIs (for that 58% out there).

Now, creating Streams is easier than ever. If you already have target groups and recurring campaigns that represent a sequence and you want to test it, all you have to do is drag & drop those target groups into a Streams folder and they will immediately be added to the relevant Stream. Learn more about repurposing your target groups for Streams here.

> From Forbes

Article: Knowing the real impact of campaigns

Read what Optimove’s CEO had to say on how marketers can learn what is working in this recent article.