I could theoretically go out and buy a pallet load of twenty, 360W rated, 72 cell panels at something like $160 per piece, costing something like $3,200 + $400 LTL freight. But it's going to cost way more than that to put those 20 on my roof or ground mounts and make them useful.
https://www.solarserver.de/photovoltaik-preis-pv-modul-preis... has €0.43 per watt peak (STC watt) only for high-efficiency panels, the kind you buy when you're tight on space (or the cost of your installation is dominated by the cost of labor, as you say). "Mainstream" is €0.33/Wp and "low cost" is €0.22/Wp.
Also, these costs seem to be much higher as a result of the current supply-chain crisis. The low point so far was August 02020, with high-efficiency panels at €0.30/Wp and low-cost panels at €0.16/Wp. Probably at some point shipping will get back to normal and prices will go even lower than that.
This project (CO2 to CH4) would benefit from maximum raw production per panel, if you can ramp the plants up and down, so aim them south. I've been deploying, both for my solar and for some other residential projects I'm helping with, east-west facing panels, to get production up as early in the morning as possible and run it late. You get less kWh per panel than south facing, but you get a system that, on a good day, is producing from sunup to sundown, even when those set far north of east/west. Out here, peak days are almost 45 degrees north of E/W for sunrise/sunset.
Iron Ridge XR1000 and some locally welded frames work pretty darn well for this sort of thing, if you have the ground area for it.
Now, with panels sub-$0.50/W, you can accomplish the same thing (more power) for far less money by just putting more panels on. You get less production per nameplate panel watt, but the total system costs are usually lower. And you can push your DC:AC ratios pretty high if you want - cap your inverter out on sunny days, but still get more power on cloudy days. It just depends on what you're trying to optimize for.
I still see trackers on occasion - they're cool. But the only place they're useful is if you have some stiff limit on nameplate capacity. Out here, you're limited to a 25kW system for residential, and you can't exceed your local transformer capacity in panel area - not inverter output (why, I have no idea, I've gotten a range of BS reasons that mostly center around somehow changing the system, blowing up the transformer, and then the power company being on the hook for a new transformer). So if you want to run a large "residential" site (think a ranch or something like that with a bunch of outbuildings), you can hit the 25kW limit quickly, and then need to go to a tracker to increase kWh generated on your 25kW system.
But outside edge cases like that, just put more panels on. The systems I'm helping people with are mostly A-frames, with east-west facing panels for long solar days, and a fairly high DC/AC ratio. I've got 7kW of panel on 6kW of inverter, though I rarely see the inverter past 4500W, which is by design - I don't like pegging out power electronics for long periods of time, and prefer to run things at 80% of design load or lower for longevity reasons.
And the article seems to miss that wind is still beating solar from the LCOE charts I've seen. They keep making the towers taller and taller for better economies of scale.
Reverse metering is basically a shadow subsidy.
And I'm not arguing that the grid doesn't need to be adapted or needs work, but if we have a lot of home generation, then we don't need to worry as much about increasing overall grid capacity to handle BEV home charging.
plentiful wattage would be an interesting DIY project.
