Something I've been wanting to run by the HN community - would you pay anything for a backup home Internet connection? How low do you think the price would have

Something I've been wanting to run by the HN community - would you pay anything for a backup home Internet connection? How low do you think the price would have to be to get 50-80% of your neighbors to buy it? Would $20/mo be low enough? I've spent my career building and maintaining small, regional Internet Service providers. One of the realities of the market is that no matter how good your service is you can't get more than around 15-25% of people in an area to use your service. Most of the companies trying to do this on a large-ish scale (like Starry) are trying to break that barrier by selling wholesale contracts (like exclusive contracts to entire apartments buildings / communities) and/or by piling other services on top of the Internet service and reaching higher and higher speeds. I have a hypothesis that a company that could go the other direction and provide a really cheap, basic Internet connection (<20$/mo, 30mbps) could gain more market share in an area specifically because it would be cheap enough to be used as a secondary Internet connection. I'm thinking of three types of customer here: 1) Very price sensitive customers who just want the cheapest thing they can get that's still usable. Maybe they have Internet access at work and use satellite TV at home and are OK with very basic Internet service at home. 2) Customers who use the service as their primary connection and it saves them enough money to add tethering to their phone plan, so if the service is unreliable they can switch to tethering. 3) Customers who have cable TV and Internet and can switch over to this service when their cable Internet is slow or down. I'm thinking of two types of plans, one for $20/mo and one for 10$/mo + 1$/day that the service is used. These numbers mostly work out on paper as long as >60% of the homes in a community sign up to use the service and it's understood that support is going to be slow. Like, no phone support (email/text only), response times in 1-2 business days, and it might take a week to send someone out if your service is down (but we won't charge you for the time it was down.) Obviously this class of service only works if it's redundant with something else. I'm kind of thinking about it as the 'MagicJack' of Internet service - as home phone service started to become less essential (due to the redundancy of having a phone in your pocket) it became possible for a very low cost and not particularly reliable service to have a place in the market. They would not have had the strict geographic customer density constraint, though. The cool thing is that once you get to that 60-80% market penetration you can start offering really great plans at really great prices. Like 1gbps symmetrical for $50-60/mo, 10G symmetrical for ~$250/mo. If anyone is interested in chatting about this my email is in my profile. I'm not quite committed to pursuing it, but it's something I think about a lot and would love to get some outside perspective.