Analyze success through attribution
Attribution helps you understand whether the money and effort you invest in your strategy are actually driving results. As you explore Klaviyo’s reporting tools, keep this in mind and take a critical look at your data to identify areas for improvement. The most successful brands focus on key performance indicators (KPIs) and understand how revenue is being attributed.
Attributed revenue, also known as Klaviyo-attributed value (KAV), refers to revenue generated through your marketing channels, such as email, text (SMS and RCS), WhatsApp, and push notifications. For example, if a customer engages with your email and makes a purchase within your attribution window, that purchase is counted toward your attributed revenue. Understanding this relationship provides a more comprehensive view of performance and helps inform your next steps.
What is attribution and why is it important?
Attribution indicates which message, channel, or strategy is responsible for a customer conversion. With this data, you can gain a deeper understanding of your subscribers and tailor your marketing efforts to drive more revenue for your business.
There are 2 types of attribution tracked in Klaviyo:
- Event attribution
When Klaviyo attributes events to specific messages and customer actions (for example, a placed order can be attributed to a text (SMS and RCS) or WhatsApp campaign), event attribution data helps you understand how your marketing efforts influence your customers' behaviors and make informed decisions about your messaging strategy. - Message attribution
When Klaviyo attributes message data to when the message was first sent (for example, if an email campaign was sent in a recipient’s local time zone, the attributed message data will be dated at the earliest sent time of the campaign.
How does Klaviyo track attribution?
Klaviyo uses a cooperative, multi-channel model to track event attribution. This means that Klaviyo provides distinct attribution windows for each channel (i.e., email, text (SMS and RCS), WhatsApp, and push). Klaviyo then attributes certain customer events (e.g., placed an order) to touchpoints in a given attribution window.
In your Klaviyo account, you can view and customize your attribution settings to best fit the needs of your business.
Default attribution windows
Klaviyo has its own pre-defined attribution window that takes into account the optimal time when a recipient would engage with your message and take an action. Message attribution comes into play here when determining where a message falls within a distribution window.
Klaviyo’s default attribution windows are:
- Email clicks: 5 days
- Email opens: 5 days
- Text (SMS and RCS) clicks: 5 days
- Text (SMS and RCS) deliveries: 12 hours
- Push opens: 24 hours
- WhatsApp clicks: 5 days
- WhatsApp opens: 12 hours
In all cases, these windows start once the message is sent from Klaviyo. However, you can also customize your attribution window settings for all of your channels in Klaviyo.
To do so, select your account button in the bottom left-hand corner. Click Settings > Attribution, and scroll to the Attribution windows section.
Customizing these windows allows you to better match your unique customers’ typical behaviors when interacting with your message, providing a more accurate picture of which messages led to customer action.
Select which interactions you’d like to track by clicking on the checkboxes; then, customize your timeframes using the accompanying fields. Click Save when you’re done.
Filtering out transactional messages, bot clicks, and machine opens
In your Attribution settings in Klaviyo, you can exclude transactional messages, bot clicks, and machine opens from tracking or reporting. This helps you get a more accurate depiction of how your subscribers interact with your marketing messages, without being skewed by automated clicks and opens.
To do so, open your Klaviyo account and navigate to your account in the bottom left-hand corner. Click Settings > Attribution, and then scroll to the Tracking & reporting data section.
Select any of the following to exclude:
- Exclude transactional messages from attribution data.
- Exclude bot clicks from email clicks.
- Exclude bot clicks from text message clicks.
- Exclude Apple Privacy Protection opens from email opens.
Click Save when you’re done.
When you implement this setting in Klaviyo, your open and click rates are expected to decrease. Do not panic; this is expected!
This occurs because you're no longer counting superfluous clicks and opens in your data. All the click and open data you have is from the recipients you are trying to reach, providing more accurate insights into your subscriber base.
Note: Klaviyo’s attribution model will recalculate historical data after attribution settings in the account are updated, so past and future data remain consistent. Updates to attribution models may take up to 36 hours to be reflected within your account. Learn more about attribution settings.
Choosing an attribution model
By default, Klaviyo uses a last-touch attribution model, meaning every customer purchase is attributed to the last Klaviyo message that influenced the order. This helps you see which campaigns or flows are most effective at driving conversions in the final step of the customer journey.
If you're using Marketing Analytics or Advanced KDP, you also have the option to choose a linear multi-touch attribution model. With this model, credit is spread evenly across all Klaviyo messages a customer engaged with before making a purchase.
For example:
- Sam clicks a text message with an exclusive discount to browse your site.
- Later, he clicks on an older promotional email from your brand and makes a purchase.
With last-touch attribution, the purchase would be credited entirely to the email, since it was the last interaction before the purchase.
With linear multi-touch attribution, the purchase credit is split evenly: 50% to the text message and 50% to the email, because both messages influenced Sam within their respective attribution windows.
Regardless of the model you choose, purchases must occur within the attribution window for each channel to be counted.
Key considerations
When tracking attributed revenue, several key considerations should be taken into account.
- Consider your customer’s buying journey
Consider your customer's buying journey to better understand how they engage with your brand and how this might impact attribution. Some products and services may require more time and engagement to encourage your customers to make a purchase. Being aware of how long this typically takes and how much engagement is required to drive revenue will serve you well. - Consider the behaviors of each marketing channel
Think about how you interact with different types of messages you receive. When you receive a push notification, a text (SMS and RCS), or a WhatsApp message, you likely interact with these messages very shortly after you see them. It may take you a little while longer to see an email and take action on it. Keep these behaviors in mind as you set up your attribution windows to ensure you accurately reflect which messages are driving action. For example, it usually makes sense for brands to have shorter push open and text message click windows, and longer email open and click windows. - Consider how your attribution windows work together
When setting your attribution windows, you may track multiple interactions per channel (e.g., email opens and email clicks). For example, if you set a longer click window than the open window, you ensure that you capture those who have high intent to purchase and may take longer to progress through the purchase funnel. Coordinate these windows to align with your customers' behaviors. - Consider the seasonality of your business
Analyze how your customers engage with your brand on an annual basis and consider this when reviewing your attribution data for the year. For example, a custom sandal business might experience a significant increase in revenue in the lead-up to summer, as hotter weather drives sales. Other seasonal events, such as Black Friday/Cyber Monday, may yield wildly different results and require a separate marketing strategy.