I did some freelance work once, as a sub-contractor to an agency, and joked that what I was hired to do was produce deliverables by the pound.
I never met the clients, had only the most rudimentary information about the larger context of the project, but had specific documents (content inventories and taxonomies, mostly) to create. The agency was happy with my work, presumably the documents I created were acceptable to the clients, and I got paid, but the experience was less than satisfying professionally.
Early on in my content strategy career, I created a set of questions when asked to create any project deliverables:
Who is the audience?
What is the context into which this fits? and
What decisions are going to be made based on this?
Essentially, these address the users and context pieces of the content/context/users triangle. If I couldn't answer all three of those questions, I didn't start work until I knew.
Too often, we embark on a project with our set list of deliverables (particularly true in consulting environments), often put into a project plan by a project manager who may or may not know what purpose they serve, we create them, hand them off, and hope that they are useful to someone.
I am a firm believer in the value of the kinds of deliverables we create as content strategists—content inventories, audits, content matrices—but not every project requires every one of them and I know I'd rather spend my time adding extra value to the ones that will truly inform the recipients and help answer the critical questions involved in the creation of a great user experience. Definitely beats creating deliverables by the pound.