@klhicks912 Leading with the reader's situation works better than leading with the book. A question about a reading slump gets replies, and each reply puts the post in front of more people before the link ever shows up.
@Rakshaweb3 A simple tip: every few weeks, ask what they're stuck on right now and sort the replies by repeated words. The most repeated words/themes should be the next topics, and they tell you what a first paid product could be as well.
@davesaunders The calendar also stops channel hopping. Without it, a slow week on cold email turns into trying LinkedIn, then Reddit, with none of them getting enough reps to show whether they work.
@mxochai Turning a custom service into a fixed scope with a fixed price makes it something you can hand off, teach or sell as a product later.
Custom work is always tied to your calendar, trading hours for dollars instead of leveraging hours.
Give an AI answer something to quote. Broad guides that hedge every point give it nothing to pull.
We're starting to shift also to prompts/questions that force a recommendation. Citations are getting more and more hidden.
@SOOPLLC Don't discount the work of cutting. The first draft finds out what you think, and the second removes everything you wrote down trying to get there.
Viewers stay for a question they want answered, not for data they didn't ask for. Open with the question and strategically hold the answer until the point it pays off. Consider cutting any fact/info that doesn't move toward that answer, even if it's true or interesting.
@simonowens Publishing natively on each platform trades control for reach. If you can move that audience to somewhere you own, like a newsletter or membership, you build somewhat of a moat for the whims of a platform and/or algorithm.
@thegoldeenhand True. The first ten videos matter so much. Mixing topics out of the gate teaches the system you have no clear message or audience, and recovering from that takes longer than starting narrow. You can widen later.
After a program runs once, run the same version for the second group and change only the lesson where most people got stuck. If you rebuild the whole thing, you can't tell which change helped, and the second group gets an untested program.
Before you publish, send your first chapter to twenty strangers who match your reader and ask one question: would you buy the rest? Their answers tell you whether the opening works, and a cover poll among friends can't, because friends are judging you and not the book.
A member back after a week away often can't remember where they stopped, and hunting through the outline is enough to make them close the tab. The Lockers Classroom opens with a Continue card for their last unfinished lesson, with the time they stopped at for uploaded video.
Members answer faster when your question asks for a fact from their own week. "What do you think about pricing?" asks for an essay, and "What did you charge your last client?" takes ten seconds. A Lockers feed post can also carry a poll with 2 to 6 options.
A one-person coaching business earns more by selling more hours or by charging more per hour. Hiring adds hours but also payroll and your time managing people, so test a higher price on new clients first. Current clients keep their rate, so nothing breaks if it doesn't work.
A member who finished your course is the best proof to a new member that it can be finished. Ask finishers to post what they made. On Lockers, the course finish page has a Post in the community button that opens a draft the member can edit before posting.