Skip to content
  • There are no suggestions because the search field is empty.

Best practices for setting up your experiences

Some tips from our experts

Start simple. There is no need to begin with the most complex targeting or product logic on day one. A broader setup can help you launch faster, gather learnings, and identify where more advanced targeting will have the most impact.

Focus first on the placements that matter most. Higher-visibility placements often drive more engagement, so it is usually better to get those right before investing in lower-priority areas. If one placement consistently drives the most interaction, prioritize getting that experience right before expanding to secondary slots.

Be thoughtful about complexity. More granular targeting and custom product mapping can be powerful, but they also take more time to build and maintain. Before adding complexity, consider whether it is likely to change outcomes in a meaningful way.

A good approach is to launch with a version that is practical and easy to manage, then iterate based on what you learn. Once you have a baseline, you can test more advanced segmentation, targeting, or merchandising logic with more confidence.

AI can be a strong option for certain components or straightforward updates, especially when prompts are detailed and specific. For more complex experiences, custom interactions, or advanced logic, it may make sense to involve our team earlier in the process so that we can guide you to success. Feel free to reach out and we'll be happy to guide you!

To summarize:

  • Start with a clear goal Define what the experience is meant to achieve before building it.
  • Align setup to engagement level The more prominent the placement, the more worth it it is to invest in strategy and optimization.
  • Plan for iteration Treat the first version as a baseline, not the final state.
  • Add measurement early If engagement data will help you evaluate performance, set up tracking at the start.
  • Match the build method to the use case Most updates will be a good fit for AI, while more advanced experiences with logic e.g. price and shipping tests may need support or dev help.