Travel & Hospitality
Data & Governance
Optimize your funnel for conversion
When new users sign up for the product, register in the system, or add items to the cart, some will inevitably get stuck and not finish the conversion.
Maybe they’re confused by the UX, aren’t sure how to proceed, or don’t know the right pricing plan.
With accurate analytical data, you can optimize your funnel and increase the conversion rate by finding where the funnel leaks.
To analyze it, you need to extract data from multiple sources, transfer it to a data warehouse, and analyze it with a BI tool. In this instance, we will gather the necessary data from the Shopify connection (transaction data), Web, and iOS sources (web event data) and transfer all collections to the Redshift warehouse. After that, the visualization can be created by Mode Analytics.
1. Choose your conversion event and steps leading to it.
For most products and services, there is a sequence of steps you’d expect new users to navigate through to achieve the conversion.
The first step is to map out these events and start tracking those in Intempt.
Most common conversion events across different industries:
- eCommerce Order completed
- SaaS Trial/subscription started
- Travel & Hospitality Booking placed
Whatever yours is, make sure it ties to some meaningful part of your businesses’ revenue.
Once you have the one metric that matters, make sure you’re tracking the 3-5 steps which lead a user to the conversion event.
For example, for the eCommerce page, these steps could be:
Visited landing page → Visited product page → Added an item to cart → Opened checkout page
Create Account → SignUpPage1 → SignupPage2 → Registration Complete
2. Set up your iOS and Web Sources
Install js and iOS tracker on your website to start tracking your selected web events (e.g., page views, interaction events).
3. Set up your Shopify source
Connect a Shopify connector as a source and get all the transaction data from your CMS database.
4. Create pipelines
Raw data from different sources need to be transferred to the data warehouse (e.g., Redshift). After that, the BI tool (e.g., Mode Analytics) can query the data via the Redshift cluster to create funnel models.
You will need to create three different pipelines with 3 data sources you have already added.
How to create a pipeline:
a) Go to Pipelines
b) Select Shopify as a source
c) Select your warehouse as the pipeline destination
d) Add necessary transformations to your data.
Transformations must be done so the delivered data would match the schema in the data warehouse. Example transformation:
Property rename. Input schema from Shopify has a property name “order_id.” However, the schema in the data warehouse has the same property with a different name, “Purchase_Id.” Transformation is required so the output schema would match the schema present in the data warehouse. To do this, apply property rename transformation to the source schema - change the string “order_id” to “Purchase_Id.”
Data enrichment. Data is enriched by adding new columns based on the lookup tables. For example, after using the lookup tables store_number is converted to store_name, which makes the output data far more accessible.
e) Add additional configurations to the pipeline.
Select which events and attributes should be sent to the destination.
Apply the same principle to create Web and iOS source pipelines.
Once the pipelines are configured, you can load the data to the warehouse.
5. Analyze data
After the data is delivered, you can use a BI tool like Mode Analytics to query the warehouse’s data and visualize the five funnel events.
Each bar in the chart will show the number of users who have made it to the next stage and gives us a sense of the fall-off.
Mode analytics will also give us a view of the individual users who fell off at each funnel stage.
6. A/B test and measure
Once you have identified the biggest drop-outs in our funnel, you can A/B test different tactics to get the users back and avoid the fall off.
Here are some ideas:
- Limit the options of what new users can do. 3-4 paths are already too many. Limit the first steps to just a few to ensure they continue going down the funnel.
- Onboarding. Because it’s their first interaction with your brand, add tooltips or interactive walkthroughs so the new users are fully onboarded.
- Emails. Send well-timed and personalized reminder emails to get the users back on track to the funnel.