What can you achieve with better internal search functionality?
Behavioral data allows you to see how your users interact with your search second by second, with each interaction recorded as an ‘event’, i.e., a row of information with contexts (entities and properties) added.
An example might be:
Event = User search.
Entity = User (female, 34, USA, etc.)
Event Properties = search number, session number, user intent, item type…” and so on (Snowplow has 140 out of the box, and unlimited custom entities and properties).
This behavioral data needs to be accurate and granular:
- Accuracy allows you to stitch users across devices, channels, and platforms without significant errors which could skew your results.
- Granularity means you have enough metrics to ask ever-more interesting questions, such as “are users starting with one intent and buying a different product?” (here’s a technical read on user intent).
This better search experience has several outcomes:
- Delivery of more relevant search results, which leads to great user satisfaction
- Raising ‘clickthrough’ on top-ranked results, which drives traffic to high-value parts of your site/app
- Integrating conversion results into the search ranking
- Removing the need for error-prone and expensive human intervention, cutting development costs and permitting greater automation
Customization is key to optimizing internal search
Companies with better data which is well understood outperform the rest. Your search engine is not a commodity – it’s core IP for your business. Using the same standardized search analytics as your competition makes innovation extremely challenging.
Customization means shaping your data to your own context; this inherently distinguishes your offering from other products.