Create 1-on-1 personalized storefronts that
increase conversions and AOV.
About the Author: Valentin is the Founder and Head of Growth of Up Collective. An e-commerce growth agency focused on performance and creative for Meta, TikTok and Google Ads.
Org: Up Collective
It’s nothing new that you can’t trust channel attribution data in 2023. Since the roll out of iOS14.5 it’s been increasingly difficult to make sense of your performance data.
For marketers this is a common situation: Google is claiming tons of conversions via Performance Max but you have a suspicion it’s overcounting. Meta went from drastically under-reporting conversions in 2021-2022 to possibly overclaiming in 2023. TikTok on the other hand is still under-reporting conversions in its ads manager.
So how can you make sense of all this messy data? The answer is simple: Ask your customers.
Post-purchase surveys are a great way to sense-check your media spend decisions. But how do you do this I hear you ask!
We’ll walk you through the exact 5-step process we just went through with a client. *Spoiler-alert* their ROAS improved 54% MoM in January.
Make sure you have your post-purchase survey tool aka Nudgem installed. This is a simple one-click install in the Shopify app store.
Set-up the survey with the RIGHT questions:
It’s important that you ask the questions in a certain way to make sure you get good data back.
The first question is the all important: “How did you first hear about <insert brand name>?”
The “first” is important because it prompts them to think back to the start of their journey.
Then you want to list all the channels where people could have heard about you. Make sure to include “Other” with a free form field.
- Facebook & Instagram sponsored post
- TikTok sponsored post
- Google search
- Bing search
- A friend, family or a colleague
- Instagram influencer post
- Other (free form)
It’s important that you phrase them in a way the customer thinks about the channel. For instance Facebook Ad would not be a good response as only marketers think of sponsored posts as Facebook Ads. Instead phrase it as “Facebook & Instagram sponsored post”.
Bonus: A good follow-up question to ask is “What made you choose our product over an alternative?” or “What almost prevented you from completing your purchase?”
It’s time to gather data. Just leave the survey running for long enough to get at least 50 responses but better yet 100+ so you have a representative data set. Depending on the completion rate and how many conversions per day you are getting this could take anywhere between 1 day to 1 month.
It’s analysis time. Take your data set of responses for the last 30 days and pull out responses and revenue per channel and the % each channel makes up of responses. This might look something like this:
Important: The post purchase revenue is not your actual revenue figure. Because only a certain % of customers are answering your survey. But it’s a perfect proxy to know how much of the total pie a channel is driving.
Now we can take this data and add your total new customer revenue for the same period.
If you’re on a basic Shopify plan you might have to use a quick workaround to get this figure. Filter for the desired timeframe. Then check your total revenue figure for the period. This is all customers new and returning. Now you can use the % of returning customers in the dashboard to work out what % new customers are driving. Finally you multiply your % of new customers with your total revenue to get your total new customer revenue.
Almost there! You now need to multiply your total new customer revenue figure by the % of responses for each channel.
This is how you’ll get your true revenue by channel. See the example below:
Now onto the final step of the process. Calculating your True ROAS:
Add your new customer marketing spend per channel for the same period. And then divide your true new customer revenue by the marketing spend per channel. This will give you your true ROAS and efficiency figures per channel.
In this example we can see that Influencers are driving an amazing ROAS of 26.5x at a low spend of $4750. This is an opportunity to double down and invest. The same goes for Bing Search 11.6x ROAS and $12.500 spent.
At the same time you can see Google is delivering a poor ROAS of only 1.7x against an average of 5.87x. But $56.700 budget is being spent on this, so it's time to take money out of this channel and reallocate.
The goal is that the % of responses roughly equals your share of marketing spend.
If you're spending between $10.000-50.000 per month on ads, running through this process monthly is probably fine. If you’re spending more, then it could be worth turning it into a reporting dashboard that you can automatically pull and review weekly.
If you have any questions about this process please reach out to firstname.lastname@example.org and we'll be happy to answer any questions.