Very interesting comments and if I may I would like to add that not only needs analysis needs to be carried out in a far more approachable and realistic way but I consider the second part of Nick's approach a major development in what we should be paying attention to. I do not teach ESP very often, I must admit. But as an EFL teacher, I adopt as a thumb rule to incorporate authentic texts that have a format that resemble not only what they do see in daily routine but also, it's actually funnier to compile things that are pieces of actual language.So, if you can grab actual can labels, underground panel photos, bus/theater tickets and other similar items, students will feel much more connected to the information contained in that particular vehicle. Because the probability they are going to use it in real life is very high. But that is my personal experience. And again, it is mostly EFL, not ESP.

Very interesting comments and if I may I would like to add that not only needs analysis needs to be carried out in a far more approachable and realistic way but I consider the second part of Nick's approach a major development in what we should be paying attention to. I do not teach ESP very often, I must admit. But as an EFL teacher, I adopt as a thumb rule to incorporate authentic texts that have a format that resemble not only what they do see in daily routine but also, it's actually funnier to compile things that are pieces of actual language.So, if you can grab actual can labels, underground panel photos, bus/theater tickets and other similar items, students will feel much more connected to the information contained in that particular vehicle. Because the probability they are going to use it in real life is very high. But that is my personal experience. And again, it is mostly EFL, not ESP.

Very insightful, indeed. First, start by looking at the tasks students need to perform in their professional life, and then do some discourse analysis. I couldn't agree more that it's the right approach. ESP should be task-based or performance-based. It's all very well if one teaches professionals. However, I always have a dilemma, teaching university students, whose future careers can vary - some will stay in academia, some will go and work in various industries related to their present studies, some might even go and work as teachers in schools... What then? It's a bit more tricky to decide what to teach - what they will need in the future.

Very insightful, indeed. First, start by looking at the tasks students need to perform in their professional life, and then do some discourse analysis. I couldn't agree more that it's the right approach. ESP should be task-based or performance-based. It's all very well if one teaches professionals. However, I always have a dilemma, teaching university students, whose future careers can vary - some will stay in academia, some will go and work in various industries related to their present studies, some might even go and work as teachers in schools... What then? It's a bit more tricky to decide what to teach - what they will need in the future.