Personal digital devices are making their way to schools. They will change significantly the way technology is used for teaching and learning. The time spent on learning activities, which take place in digital environments increases significantly. As a result, rich data about learner’s activities will be stored in different databases. This data can be aggregated from different sources, mined and analyzed in order to inform teachers, learners, parents, policy-makers and researchers about learning needs, gaps, success factors etc. As today’s practices in educational data mining and learning analytics are practically not connected to learning theories, it is difficult to make pedagogically meaningful inferences from large scale data analysis. A typical educational data mining study today strives to measure what can be measured easily (eg. counting page hits). In order to inform developers of next-generation virtual learning environments, educational researchers’ community could start from reaching consensus on domain ontology for pedagogy. We are going to discuss our first attempt in this direction: the domain ontology we extracted from Estonian teachers’ professional qualification standard. We are also going to demonstrate how this ontology was implemented in the design of two different software applications: organizational competency management tool OPT and virtual learning environment Dippler.