- You see that platform data does not match actual orders
- You want to know which channels are getting too little credit in your dashboard
- You are considering setting up a post-purchase survey and want to know how
- You want to know how to interpret the responses without overreacting to them
Platform dashboards only count what they can see themselves, and each platform often claims the same sale as its own. If you add the numbers together, they can exceed 100% of your actual sales. Furthermore, a click doesn't tell you where the customer first heard about you.
A post-purchase survey solves this by asking the customer directly. It is a short questionnaire on the order confirmation page, right after the purchase, while the experience is still fresh. You get the customer's own answer alongside what your tracking shows.
It helps in two ways: it benefits the technical side of your tracking, and it provides insight into who your target audience is and what motivated them to buy. Read more about why platform numbers don't add up in Attribution on Shopify.
- Place it on the order confirmation page, right at the top, so the customer sees it immediately after the purchase. Check whether it should be on the standard Shopify page or a custom page, and ensure no third-party apps are in conflict.
- Start with the attribution question. This should always be the first one, as it is what corrects your attribution data. Ensure it is linked to your attribution tool.
- Add a few fixed questions with a purpose. See the list below. Each question should provide an answer you can actually use.
- Keep it short. If it needs to be shortened, remove store-related questions first, then brand-specific ones, and keep the four fixed ones.
- Get the questions approved, before the survey goes live.
- Check after one week, that responses are coming in. If the response rate is low, adjust the placement or flow.
Our experience is that approximately 20 to 40 percent respond without an incentive, and approximately 40 to 60 percent with one.
The most important thing is that there is a thought behind what you are asking:
- Where did you first hear about us? The attribution question.
- How long had you known us before you purchased? Shows how much time passes from the first encounter to the purchase.
- What will you use the product for? Tailor the question to your industry. A jewelry brand might ask about materials, and a sports bag brand about the type of sport.
- What made you buy from us today? What triggered the purchase.
- Which content convinced you? Build the answer options based on the formats the brand actually uses, and always include "An offer or a promotion" and "I didn't see any content / other".
- For returning customers: What made you shop with us again?
If you want to ask more in-depth questions, they can be sent in a follow-up email some time after the purchase.
- Read the survey alongside your tracking, not instead of it.
- Look for the discrepancy. If customers mention word-of-mouth, Google, and social media more often than click attribution shows, it is a sign that those channels are doing more than the dashboard gives them credit for.
- Adjust your management accordingly, and keep an eye on whether the discrepancy persists.
- Wait for enough responses. We only use the responses to correct channel data once there are approximately 500.
- Use the other questions. We use the responses regarding what triggered the purchase and which content was convincing to evaluate creatives. The answer regarding how long the customer has known you helps with planning retention.
A survey alone is only indicative. It is based on what the customer remembers and chooses to answer, and only those who respond are included. Therefore, use it in conjunction with tracking and platform data, and do not let it stand alone when you shift budgets.
DVISIONMEDIA's approach
We build the survey at the start of a collaboration: the attribution question first, a few fixed questions with a purpose, and brand approval before it goes live. We check after a week that responses are coming in, and we read them alongside tracking and platform data so we can see which channels are getting too little credit. It is part of the approach we build into our work with attribution and through E-COM OS. Read alsoWhat is a good ROASandMER vs. ROAS.
1. What is a post-purchase survey?
A short questionnaire that the customer sees right after their purchase on the order confirmation page. It is used to find out where the customer heard about you and what motivated them to buy.
2. Which question should come first?
The attribution question, for example, "Where did you first hear about us?". This is what corrects your attribution data.
3. How many questions should there be?
Keep it short. We use four fixed questions and one about the content. If it needs to be shortened, remove the store-specific questions first and keep the four fixed ones.
4. Can I rely on the answers alone?
No. A survey alone is only indicative. Read it alongside tracking and platform data.
5. How many people respond to a survey?
Our experience is approximately 20 to 40 percent without an incentive and approximately 40 to 60 percent with one. If it is lower after a week, adjust the placement or flow.
.png)
