User needs and data engineering

When it comes to discovering user needs, the gold standard in tools is to talk to your users up front, sit with them as they use your applications and observe their interactions. This works. You get immediate feedback from your users, you can quickly see their frustrations, ask for their ideas and quickly form new hypothesis.

User needs are always an core focus in application development, particularly with government services. While sitting with users is clearly the most effective, they can also be a difficult way of extracting insight. They take time, people and expense. Something mature products often can't access.

Web Analytics are only part of the puzzle

Web analytic tools (i.e. Google Analytics) are useful in that they show us time spent on pages, how often users return and which pages are most popular when visited. Web clicks and dwell times are not a complete picture.

• Some applications run internally, ruling out most web analytics

• Often users have multiple touch points that are poorly measured via web analytics i.e. a customer orders a driving licence and calls to complain that the photo is incorrect. How do we join up this disjointed journey.

• Data in web analytics is segregated. If you wanted to display your promotional spend on a marketing campaign vs web traffic patterns, there are multiple exports and steps to merge data required. Web Analytics data is siloed unless you take a more defined approach to data sharing.

Business Intelligence is not user intelligence

BI often serves some of the most critical functions in a business, but rarely are user needs seen as critical to operational data. The audiences are different for user needs. Data warehouses are great at capturing long-term data points but less so at guaging user experience. How did our change impact user behaviour? You won't find it unless you create the right views.

Multi-touch attribution for user needs

To get the right views, we need to collect, transform and create the necessary views to recreate the user experience. Multi touch attribution is only possible when you combine web analytics, service metrics and operational data together to make sense. Easier said than done of course.

Technology leaders should be thinking to themselves, what piece of the puzzle is missing from the oranisations insight. How do I get access to it and how can I make it useful.

Your organisation may need a customer data platform (CDP)

The recent trends towards customer data platforms highlight this. More organisationas are combining their data to build on their strengths and improve their tooling. Connecting innovative services to enable end-to-end visibility of their processes. Think insight, think combined, think forward.