Q: Should you constantly be referring to risks and next steps throughout the case? If so, how to do this in practice?
After you have analyzed one area of the case (eg a graph), it is a good idea to propose the next steps related to that area. This is expected in particular in a candidate-led case.
The simples way to move to the next steps is to state your hypothesis and ask a question to the interviewer related to that.
From element XYZ of the graph, we can infer that the issue is ABC; from what shown, my hypothesis is that we could decrease ABC by renegotiating with the supplier. Do we have any information whether this is possible and if so, which would be the new rate we could get from the supplier, in order to calculate the possible savings?
The same could apply to risks, although a risk analysis is normally needed when you have all the elements to make a decision. As a consequence, it is more common in the conclusion.
Hope this helps,