I agree with Mr. Graham about the importance of having a definition of "success".
For training evaluation, the Kirk Patrick model is frequently used/quoted, you might have known. This link offers more detail: http://www.ct.gov/ctdn/lib/ctdn/ttt_14_m5_handouts2.pdf
Assessing the success of any lecture , meeting or training program is dependent largely on what the objectives of the session was. So the success of that program would mean the objectives were met.
For instance, objectives could be knowledge, skill or behavioural based. Measuring the outcome of your meeting lecture or training would follow suit. So if you aim at imparting new knowledge you want to ask the participants questions before and after the session to determine the change your lecture has made. A written test or quiz could achieve this.
Pre and post pactical assessment could do the same for skills. Behavioural or attitudinal outcomes can also be appraised similarly depending on what they are.
I guess the key thing is to identify the domain where your expected outcome(s) is or are and plan to assess them from the onset. A pre intervention assessment forms your baseline against which you can march the post and draw inferences.
Definition of success is different for different task.
Success also depends on the objective and goal of any session and all. If the objective is fulfilled and goal is achieved then we can say that the task is successful.