On a quick and dirty conservative estimate of costs vs risk vs reward analysis:
The 'average wage' at PGI is roughly $200+ish per day. (public company, you can find the pay grades with a quick google)
To create Quads you'd need to develop a slightly different control set (relatively easy, strafe instead of turn on legs, turn on torso to turn mech), a new skeleton template (trickier), they're bringing back kinematics anyway, and a different crit / mounting system (this could lead to some things get leg mounted weapons though), layout, and a few other bits.
Based on the size of the studio, I reckon 10-14 would likely need dedicating to it for around 6 (to 12) months.
So that would be around 1,300 to 1,820 days of work (if 12 months, up to 3,640), give or take a few.
This would work out to a development cost of around $260,000 to $364,000 (up to $728,000 if 12 months).
Throw in extras, sundries, and other unforeseen overheads, that becomes an easy $300,000 to $400,000 (even $750,000) in development costs.
To make this properly profitable, by usual business standards, you'd be looking at a 100% return, so a $600,000 to $800,000 minimum take ($1,500,000 if it gets bumped to 12 months), which would mean a need to sell around 30,000 to 40,000 IS packages (or package levels), or 20,000 to 26,667 Clan packages (or package levels).
Lets say it'd be 2/3 IS, 1/3 Clan, that would mean needing to sell at least 20,000 IS packs (or levels of) and at least 6,667 Clan packs (or levels of) to be minimally viable.
Which means, of the core 7 (lets say they fudge it and expand to 10), they'd need to sell around at least 2,000 of each Mech to justify it. (If not 4,000-6,000 of each on average if it took the 12 months).
And these are all pretty conservative numbers, running off rough averages and reasonable industry estimates - you could probably double all those numbers, certainly increase them with a proper in depth analysis.
If we then consider, in that same period (6-12 months development cycle + 3-10 months sale cycle), they need to make enough money to keep the lights on (this has been reported at over $500,000 a month to pay overheads such as wages, rent, tax, payments, etc), and still need to pump out some saleable content each month, while the huge development cost of 'quad creation' was ongoing on top of the regular costs.
So this would all be on top of the required 'average estimate sales' of around 20,000 items a month (assuming an average item sale value of $25, between IS/Clan packs, MC sales, etc), meaning to 'justify it' they would need an increase in sales of around 20-30%
If instead, they carry on as-is, they can develop other things, bug fixes, etc, and pump out mechs using existing systems, only needing to dedicated artist time and some coding - meaning you'd have a mere fraction of the cost (likely only 3-4 people per month) for the same purchase value, meaning they only need to meet the usual targets, and everything past that is straight profit.