Nothing in life is free. If you hire Jr. Mahmud to take over 80% of your job, and your aggregate productivity drops by 1000%, you will probably lose your client and be forced out of business.
So don't do that.
A steady $200/hr consulting gig offers an awful lot of wiggle room to figure this stuff out. When we hire a new consultant, (a) we're very picky and (b) they tag-team with us on projects, often for months, without being billable. That costs money in the short term, but pays for itself in the long term.
Also, it's wrong to assume that a consulting gig has to be performed by one person. Time and again friends of mine have found that they could hire cheap for 20-40% of their project work (documentation, test harness coding, automation, etc) and then had that person grow quickly into a full-on replacement. And all due respect to 'mahmud's domain knowledge, but the examples I'm thinking of are pretty high end.
At the end of the day, this problem is a fundamental challenge of running a business, and I have to respectfully/tentatively suggest that if you can't figure out how to hire and train a replacement for your own work, you stand very little chance of running a successful software company, where you'll also have to hire QA, ops, marketing, and sales.
Think of this current challenge like a warmup.
