Online service analytics provides information to support service development

In her earlier Job Market Finland blog post “Building better working life services by means of co-development”, Henrietta Sarpakunnas described how the planning and implementation of Job Marketplace is based on the operating model of co-development. In such a model, users and the service provider join forces to develop a service in interaction with each other. Interactive work often takes place in small groups the participants of which have been selected to achieve as comprehensive representation of the most important groups using the service as possible. Examples of the qualitative research methods used include interviews and usability tests on the early prototypes of the service. These methods provide invaluable information about such matters as user diversity and motivations at the individual level. 

The digitisation of services has enabled us to complement the understanding we have gathered on the users in interactive workshops with data material. This material is created when people use the service. The data material provides important indicators of actual usage situations during real use: How many visitors have viewed each page of the service, how the number of visits changes over time, which functions of the service have been used, etc. 

A data analyst is like a chef who chooses and prepares a menu 

The data collected is rarely usable as such. The raw data must be refined and analysed before utilising the results in the form of reports. The work of a data analyst could be compared to the profession of a chef. A good chef starts by thinking about the serving situation: What is the nature of the event, who are coming and what kind of expectations do they have regarding food (cf. the report). Secondly, he needs to check which raw materials (raw data as a “text file pulp”) are available in the first place. Sometimes, it may require a laborious journey to acquire the raw materials. Once the raw materials have been brought into the kitchen, they must “be cleaned and prepped” for actual cooking. At this stage, the chef still has various alternatives by which to combine foodstuffs with each other in an appropriate manner. Once the courses and their ingredients have been decided, the chef finishes his work with a carefully considered presentation (data visualisation). The careful planning and implementation of the chef (and data analyst) culminates in a finished product that is appealing to the eye and easy to digest. 

Reports generated from the data material help us see whether the design of the service has been successful or not. The results collected may confirm that the user experience aimed at during planning and development was successfully implemented, or they may prove quite the opposite outcome. 

Combining analytics and small group work brings best results 

At its best, usage data analytics provides fact-based information for further development of the service not only to service designers but also to the entire production team. Reports based on analytics help the team prioritise development ideas for the service and identify new service needs. The reports help us identify which functionalities have been used the least and make decisions on the product accordingly: Should we remove the functionalities from the service altogether or shall we decide to invest in improving these functionalities. 

However, even large amounts of data collected from the customer experience cannot replace the quality that can be obtained by combining usage statistics with the results of face-to-face workshops in small groups. Data collected from tens of thousands of service use situations must serve the designer as he or she anticipates the expectations and customer experiences of individual users. 

Harri Klemetti 
Senior UX Designer, Gofore Oy/Job Market Finland