7 Ways Ope from Cowrywise Helps You Increase Sales Via Attribution Modelling
Google Attribution Modelling
Share this article

Every single product that you sign up for, takes you through a conversion path. Think about your last sign up. Think. 

What was it and how many of the company’s ads did you interact with before making up your mind? 

If you run any website, online business or work for one, this post will help you know what channels bring in the money, and which to focus on. You may just want to continue reading because this post is funny as hell.

User Generated Content Are Compelling, But thats Not the Issue Here

As for me, the last product I signed up for, at least before writing the first word of this article, is Cowrywise. My first encounter with any of Cowrywise’s touchpoints was at work when we were designing a campaign for a competitor. It was through a referral website, techpoint.

I was assigned to do a competitor analysis and I Googled, “top digital savings platform in Nigeria.” the rest they say, is history. However, the refferal article did not make me sign up. Ope did. and…

Here is how.

Ope from Cowrywise is seen as as a bully, a friendly one though for Cowrywise users on Twitter. Who sends emails to users with the subject, “Here is Why You are Not Wealthy,” if not a psychopath like Ope?

Why must the email pop up when people are about to TGIF or pop champagne. Well, the emails likely worked because some Twitter-Cowrywise users confessed to cancelling money-blowing plans after reading Ope’s email. 

Because of the seemingly organic User Generated Content (UGC) buzzing about Ope from Cowrywise, I signed up. Peer Pressure!

What Sayest Though About my Conversion Story? Straightforward? Confusing?

Which of the touchpoints or channels will take most credit in my conversion story? Ope’s email? The UGC on Twitter about Ope from Cowrywise? Or the techpoint article? 

Attribution modelling shows a marketer, a CEO or anybody which of the channels are bringing in the money/sales/conversion. There are several attribution models and a quick summary of them is found here, from Google. 

Image result for attribution models

You may want to see how to determine the right model for your business? Read below.

  1. First Click Model

The First interaction model gives all credit to the first click. In this version, my encounter with the techpoint article takes the entire credit and will be designated under referral. 

Cowrywise’s Referral for July

Remember that the techpoint article made me only aware of Cowrywise. In this model, there is no reference to Ope, who did much of the work in converting me. There is no mention of the users who talked about Ope and there is no mention of my direct search of Cowrywise. Just the first click. 

The issue, however, is that the marketer going by this model, channels more money to referrals, thinking that it brings all the revenue.

2. Last Click Model

Here, all credits go to the last touchpoint before conversion. My last touchpoint was direct search. I googled Cowrywise and simply signed up on the landing page. Using this method, more revenues will be assigned to optimizing direct search, so that the landing page is super friendly, and the bounce rate is down. 


Could direct search have done it alone, if not for Ope? 

We move!

3. Last Non-Direct Click

Using this model, the UGC on Twitter buzzing about Ope for putting them in check would take the credit. Unfortunately, Ope’s email may be left with little credit. However, the UGC is as a result of Ope’s emails right?

4. Linear Attribution

The next is Linear attribution, which gives all the touchpoints equal credit. Though I did not relate directly with Ope’s email, the email channel takes the same credit as the UGC, the article placement and even the direct search. 

Image result for attribution models

This is as good as not having done any modelling because it just tells you to reinvest in all the touchpoints a customer went through, both the effective and ineffective ones. You may want to use this when you are unsure, but this does not optimize for increased ROI. 

5. Time Decay

Nothing actually decays here. It is just a figure of speech about time flying by. Time decay modelling gives credit from zero to hero, as you move towards conversion, with the highest-ranking touchpoint being the last before conversion. Here, Ope is also rewarded via the UGC which is the second to the last touchpoint. 

https://twitter.com/TheNonye/status/1163711192222916610?s=19

However, this does not do much justice, as less money is earnmakred for refferals and more, for direct search and social media.

6. Position-Based Attribution 

The position-based attribution gives 40% each to the first and last touchpoint while sharing 20% credit to the touchpoints in between. Hmmm. This does not still do much justice to Ope and his/her UGC crew that did the work, so we pass.

7. Custom Attribution

The custom attribution model assigns a custom credit which is usually in percentage. Here, if the marketers know how much work Ope put in, he/she would take all the glory. However, the big question is can you do this for hundreds of customers simultaneously without losing your mind? Exactly, thats why I
love Ope from Cowrywise

https://twitter.com/Abdulakeeb/status/1162073990258671616?s=19

So What Next?

The choice of the best attribution model lies in the objective of the research. Do you want to know which channel helps create the most awareness? Do you want to know the best conversion channel or the most important channels leading up to conversion?

For me, given the Ope case study, I would choose the last non-direct click and my reason is that it comes very close to giving credit to Ope, the main reason why I am a Cowrywise customer.

Generally, this model rewards channels who usually do the most work, but are never at the forefront of recognition. I call them the backstagers.

Come to think of it, if the last touchpoint is direct search, then a rule of thumb is that the second to last touchpoint did all the work. Maybe someone referred you to a product, or you saw the product’ ad on TV, billboard, Twitter etc, and you decided to search for it.

Finally, no model can fully capture human behavior. The ones explained here, all have limitations. It depends on what works for your business.

Image result for attribution models

Which is your best Google analytics attribution model? Talk about it on the comment section.

Check in with me for the next phase of this article, “How to Conduct an Attribution Modeling that Works.”

Leave your vote

1 point
Upvote Downvote

Total votes: 1

Upvotes: 1

Upvotes percentage: 100.000000%

Downvotes: 0

Downvotes percentage: 0.000000%

Comments

0 comments

1 COMMENT

LEAVE A REPLY

Please enter your comment!
Please enter your name here