Em ..
bad idea.
As You presented it.
There is approx. zero percent probability of You being able to make the products, and sell them to the same customers, that your current clients sell to.
In any meaningful volume, in any short timeframe less than a year.
While likely quickly losing, as You intimated, 60% of Your sales volume.
If You wanna start down that path, buy the furnace "for other business" and start looking at other similar products You could be making.
And sell them to totally other customers in other areas that don´t hardly compete at all, preferably internationally.
It will likely take 6-24 months for the new business to be running smooth, and be making a profit.
After 12 months of profitably and successfully selling the new stuff You make, then have a rethink.
It might well be that one or more of your customers, at that time, when you are already experienced and proven, might be willing to start using you to do it all.
If You could get that, that is what I would recommend.
You would get much less arpu, but zero sales/marketing/collections cost and a much larger volume.
Sales/marketing/collections (often support) tend to cost 10-30% of revenue.
Yes, they do.
Even if You don´t know it.
98% of people in small business don´t.
And if they don´t cost 30%, Your sales probably are not very good and not doing well (in professional terms).
Good sales/m is very expensive but/and brings tremendous profitable revenue onto the table, +/- immediately.
Bad sales/m/c is just a waste of money.
Most money spent on marketing s/m/c is just wasted. Everyone has an excuse for no results.
It´s a GREAT idea to be a better supplier for Your current clients.
Flextronics is the ultimate example, with 200.000+ workers.
They make iPhone metal shells.
65 operations, per phone, including multiple milling operations and laser drilling and media blasting.
They get about 9$ per phone for it, with 2$ or so going to consumables and materials.
I suspect about 5$ goes to amortisation and r&d.
(Almost) No-one here on PM would be able to do it that well.
Even If given 200m$ free money, to start with, approx zero-2 pm members could do that. Imho.
I could not, would not guarantee smooth running.
300M phone shells in a 4 months period, at 9$ each, 65 ops each, for 7$ each.
Absolutely no way.
For about 2-3$ in OPEX, leaving a gross margin of about 1$, or 0.5$ after secondary corporate expenses.
And Flextronics is one of the largest businesses in the world.
It´s a GREAT idea to be an integrated supplier to Your clients.
You just need to be good, or really good, at it.