It depends… Problems vary in size and complexity. I rely on a "toolkit" of workshop, design and research techniques to understand and then solve problems.
Identify the problem we are trying to solve by gathering insights from stake-holders or users directly. Gather any research that has already been done.
Gain an understanding of current user behaviour and expectations from analytics, user interviews, surveys and competitor analysis. Outline any metrics we wish to improve. Create some hypothesis of how to solve the problems and pains of the user.
If improving on a current solution, conduct user testing to understand users current behaviour and jobs they are trying to do.
If trying to establish a new proposition, run a value proposition workshop to align team around the what solutions we believe will solve our users problems best. Using personas, research and analytics to guide these decisions.
Map out user journeys, write stories, personas and service blueprints to gain understanding of the users holistic experience. Review these outputs and outline any further opportunities.
Wireframe concepts on paper in a sketching co-design session with people from different disciplines around the organisation. Unpacking and referring to research, insights and discovery outputs to inform the session.
Outline stories and features for an MVP and what will be out of scope. So we can build, learn and potentiallly fail as quickly as possible.
Create annotated wireframes outlining layouts, journeys and interactions, collaborating with development teams to ensure feasability. If needed, high- or mid-fidelity design is then done using Axure or Sketch.
Test prototypes either face-to-face internally, externally or remotely using usertesting.com. Group, sort and analyse feedback. Present key findings and any relevent clips to team/organisation.
Refer to objectives and metrics defined during kick-off to outline what success will look like. Whether to increase conversion or improve a specific metric. Set up A/B or multivariate test, and what should be tracked.
Analyse results of test and determine whether the solution tested was successful. If not, use data and feedback to iterate on designs. Alternatively, pivot and return to build phase focusing on a different solution, or return to kick-off phase.