Your subject-matter expert sent you a 90-slide deck, two policy documents and a note that says "everything in here is important." Three weeks later, after you built the first module, they rewrote half of it. Now they have stopped answering emails. If that sounds familiar, you are not doing anything wrong. It is a very familiar pattern in course building, and it is fixable. How do you work with subject-matter experts without endless rewrites?
Working with subject-matter experts goes best when you protect their time, ask for decisions instead of documents, and agree on a fixed review process before you build anything. Start with a focused interview about what people do on the job and where they go wrong, Turn what you learn into a short list of learning objectives and get the expert to approve that list first. Plan a set number of review rounds with clear jobs for each, and handle late changes as new requests with a visible trade-off. Most rewrite cycles start because those agreements were never made.
Working With Subject-Matter Experts: What Do They Actually Need From You?
A subject-matter expert (SME) needs you to make their expertise usable without taking over their week. Most SMEs have a full-time job that is not your course. They agreed to help because the training matters, but every hour with you is an hour away from their real work.

What they usually need from you is:
- A clear picture of what the course is for and who will take it.
- Short, specific requests with deadlines, not open-ended asks.
- Confidence that you will not distort or oversimplify their field.
- A sense of when their part ends.
That last point is underrated. Picture an engineer who agreed to "help with a course" and has no idea whether that means two meetings or two months. Of course they get cagey. Give them a simple plan with dates and hours at the start, and you will often find they are far more generous with their time.
What Should You Ask in the First SME Interview?
In the first interview, ask about the job and the mistakes, not the content. The content is in their documents already. What is not written down is how experts think, what they notice that beginners miss, and which errors cause real trouble. That is what your course needs.
Here are questions that tend to get to the good stuff:
- What should someone be able to do after this training that they cannot do now?
- What mistakes do new people make in their first month?
- Which of those mistakes are expensive, risky or embarrassing?
- What do experienced people notice that beginners miss?
- Can you walk me through a recent tricky case, without names or details that should stay private?
- What do people look up most often, even after they are trained?
- What in your documents is nice to know but rarely used?
Record the session if they agree, so you can listen instead of scribbling. Keep it to about an hour. Ending on time is one of the easiest ways to show respect for their schedule, and it makes the next meeting easier to book.
How Do You Turn a Content Dump Into Learning Objectives?
You turn a content dump into objectives by sorting it against the mistakes and tasks from your interview. Every piece of content either helps someone do a task correctly, or it does not. The material that does becomes the course. The rest can live in a reference document or a link.
Write each objective as something a learner will visibly do. The Iowa State University teaching center's guidance on writing measurable learning objectives explains the difference between broad goals and measurable objectives, with action verbs from the revised Bloom's Taxonomy. "Understand the returns policy" is a goal. "Decide whether a return qualifies for a refund" is an objective you can teach and test.
Then send the expert a one-page list of objectives and ask them to approve it before you write anything else. This is the single most useful review you will ever ask for. It is short, it is easy to say yes or no to, and it settles the scope of the course in plain sight. When a new topic appears later, you both know it was not on the list.
The same approach works when the content dump is a slide deck. If you need to turn a PowerPoint into an e-learning course, the speaker notes and the expert's stories are often worth more than the slides themselves.
How Many Review Rounds Should You Plan?
Plan a fixed number of review rounds, each with a clear job, and tell everyone the plan before work starts. Two or three rounds covers most projects. More than that usually means the earlier rounds were not focused enough.
A plan that works for many teams looks like this:
| Round | What reviewers check | What they should not change |
|---|---|---|
| Objectives | Are these the right things to teach? | Wording style, visuals |
| Storyboard | Is the content accurate and complete? | Objectives already agreed |
| Built course | Does it work, and were changes made correctly? | Content already approved |
It also helps to send a short review guide with each round, written specifically for that round rather than reused from the last one. A few sentences are enough: what has changed since the previous version, which screens deserve the closest attention, what kind of comment is most useful at this stage, and roughly how long the review should take.
Reviewers who know they are being asked for about ninety minutes of focused checking tend to book that time, while reviewers handed an open-ended request tend to put it off until the night before the deadline.
Ask for comments in one place, tied to a screen number, by a firm date. And if there are several reviewers, ask them to agree among themselves before sending feedback, or name one person who makes the final call. Conflicting comments from three people are a common source of rewrite loops.
How Do You Handle Late Changes Without a Fight?
You handle late changes by making the trade-off visible instead of saying no. Late changes are not always bad. Sometimes a policy really did change, or the expert spotted an error that matters. Your job is not to block every change, it is to help everyone see what each one costs.
When a late change arrives, try a calm, three-part reply. First, thank them and confirm what they are asking for. Second, explain the effect in plain terms: "This affects four screens and the quiz, so it adds about two days." Third, offer options, such as making the change and moving the date, making it in a later update, or swapping it for something else in scope.
Keep it friendly. The expert is not trying to sabotage your schedule. Often they are seeing the course for the first time in a form that feels real, and that is when good ideas appear. Treat those ideas with respect, and they will treat your schedule the same way.
What If the Expert Goes Quiet?
Silence is usually a time problem, not a sign that the expert has stopped caring about the course. Instead of sending a third reminder about the full storyboard, shrink the request until it fits into a busy day: "Could you check just screens 12 to 18 for accuracy by Thursday? It should take about twenty minutes." Small, specific asks are much easier to say yes to than a large document that sits unopened in an inbox.
If that still does not work, talk to the person who sponsored the project, and frame it as a scheduling question rather than a complaint about the expert. Their manager may be able to free up a couple of hours, or suggest a second expert who can share the load. Either way, it is far better to raise the issue early than to discover at launch that nobody ever approved the content.
Fewer Rewrites Start With Treating an Expert's Time as Scarce
Working with subject-matter experts is less about managing difficult people and more about designing a process that respects their time. Ask about tasks and mistakes, get objectives approved first, run focused review rounds, and make the cost of late changes visible without drama.
Do that, and the expert who once sent you 90 slides and vanished may become the person who asks to work with you on the next course. The process takes a little more planning up front. It saves you, and them, from the rewrite loop that wears everyone down.
