
AI can help an endurance coaching business with repetitive administration, draft communication, organizing information and exploring programming options. Whether it improves your business depends on the problem you give it, the time you spend checking its work and what you do with the time you recover.
If you need more qualified inquiries, generating training plans faster may leave your biggest problem untouched. If your roster is healthy but onboarding takes your whole Friday, a better intake process may be the useful place to start.
Michelle Lake's challenge is the right starting question
In episode 154 of The Endurance Coachpreneur, “AI Won't Fix Your Coaching Business”, published November 11, 2026, Michelle Lake questions whether AI plan-writing apps solve the problems coaches are currently facing. Her published show notes emphasize automating other business tasks first and learning to write a plan before outsourcing that work.
PR Project's approach starts with one rule: a tool should earn its place by solving a problem you actually have. Begin where mistakes are easy to catch, and add programming support when you can explain and evaluate the result yourself.
Start here: Name the task holding up your week before choosing a tool. Count checking and corrections when measuring time saved.
Find the constraint before choosing a tool
For one week, keep a simple record of where your working time goes. Separate athlete decisions from repetitive preparation. Then identify the task that is delaying something valuable.
| What is getting stuck? | A useful first experiment | What to check |
|---|---|---|
| Inquiries arrive but follow-up is inconsistent | Draft a follow-up checklist and message from your existing process | Every promise matches your service and availability |
| New athletes ask the same setup questions | Turn your approved answers into an onboarding guide | Instructions match the tools athletes actually use |
| Weekly reviews take too long to prepare | Organize check-in notes into a review outline | Every summary agrees with the original notes |
| You repeat workout explanations | Draft reusable explanations of sessions you already coach | Purpose, terminology and instructions are correct |
| Marketing feels generic | Outline a post from a lesson you wrote in your own words | The example is real and the conclusion is yours |
You may discover that a template, calendar reminder or ordinary automation solves the problem more reliably. Use that when it is enough. Use AI for language or messy information when the benefit holds up after review.
Four places AI can help, with the coach in control
1. Administration: document the process you already trust
Start with a task such as turning your onboarding notes into a checklist. Supply the steps you actually follow, the documents you send and the questions athletes commonly ask. Ask for missing or ambiguous steps to be flagged rather than invented.
Review the checklist against a recent onboarding. Did it forget the initial conversation? Did it promise a response time you cannot maintain? Did it add an assessment you do not offer? Correct those before using it.
Keep payment amounts, account access and contractual commitments under direct review. A polished draft can still contain a costly invented promise.
2. Programming support: compare options within your methodology
An experienced coach might ask AI to list scheduling conflicts in a draft week, rewrite a workout explanation or suggest alternatives from an approved workout library. These are opportunities to inspect options before making a decision.
For example, give it a fictional athlete with a Tuesday travel day and a Saturday long run. Ask it to flag conflicts in a schedule you wrote, keeping the week's intent unchanged. Check whether it quietly moved a hard session next to another demanding day or changed the total work.
If you cannot explain why a proposed session belongs in the week, do not send it. A new coach needs to develop that reasoning through education, practice and feedback. Generating more plans does not establish that skill.
For a related product example, read how PR Project's Auto-Push builds training. That guide describes coach-defined structure and selection logic. It should be evaluated separately from what a general-purpose chatbot produces.
3. Analysis: prepare questions, then check the evidence
AI can help organize observations into questions for your athlete review. It should not turn incomplete data into a confident explanation.
Suppose an athlete missed two sessions and reported unusually high effort. Ask for a list of observations, missing information and questions to discuss. Do not ask it to diagnose the cause. Compare the output with the actual log and the athlete's own comments before drawing conclusions.
Check calculations directly. Look for missing workouts, mismatched units and dates. A plausible summary does not establish that the underlying arithmetic or interpretation is correct.
4. Communication: draft from a decision you have already made
Write the substance first: what happened, what you decided, why and what you need the athlete to tell you. Then ask AI to make that message clearer or shorter without adding advice.
Read the result as the athlete would. Does it acknowledge their actual concern? Does it sound like you? Would you be comfortable explaining every sentence on a call?
Send personal responses yourself when an athlete is disappointed, worried or questioning the coaching relationship. Drafting assistance can help you find words, but the responsibility to understand and respond stays with you.
A small experiment you can run this week
Pick one repeated task with an output you can inspect before anyone receives it. An onboarding FAQ is a good candidate. Use fictional details or information cleared for the tool you are using. Avoid uploading identifiable athlete logs, private messages or health information into an unapproved service.
Try this prompt with your own approved notes:
Turn these onboarding notes into a checklist and a short FAQ. Use only the facts supplied. Flag missing information as a question for me. Do not invent services, prices, policies or response times. Keep the language direct and specific.
Measure the whole process across a few attempts:
- Record how long the task normally takes.
- Count prompt preparation, generation, checking and corrections in the AI version.
- Note anything omitted or invented.
- Decide whether the result is clear enough for an athlete to use.
- Keep the workflow only if the time or quality improvement survives that review.
If it takes longer, simplify the task or return to your existing template. If it saves time, give that time a job: a better athlete conversation, a follow-up you have been postponing or a protected finish to your working day.
What AI cannot settle for your business
AI cannot decide which athletes you want to serve, take responsibility for a training decision or build trust on your behalf. It can suggest language and options around those decisions. You still have to make them, act on them and learn from what happens.
That is where Lake's challenge matters most. Before buying another tool, finish this sentence: “The problem holding my coaching business back is ____.” Then choose a process that addresses that problem and check whether it works.
For coach-defined training workflows, see PR Project's vision for scalable endurance coaching.
If capacity is the problem, use our coaching business growth exercises to check your service promises against your calendar before adding another tool.