Differentiating based on pricing alone is tricky. I'd suggest convenience of setting up dedicated servers -- 0 to frag in 15 seconds or less.

In terms of pricing, charge hourly but have an easy discounted day/week/month pass option. (1 hour for $0.25, 1 day for $5, 1 week for $20, 1 month for $40 -- I have done no market research, these are just to show the discount curve.)

I strongly suggest selling credits rather than doing actual utility billing. Gamers are pathological customers in a lot of ways. Wait for credit cards to break, massive fraud issues, etc.

I do not like Cloud* branding for someone who is not selling to developers. That doesn't seem to communicate value to the customer for me, since Cloud is a meaningless buzzword outside our industry. MultiplayerOnDemand, your strap line, sounds better. (GameOnDemand?)

In terms of getting users for this, you have many options. Advertising rates for gamers are low... for a reason. You probably monetize better than T-shirts, so that is a definite plus. I would start trying to recruit folks for it right now, individually -- your gaming buddies, clans, what have you. If you can't give it away, you'll never be able to sell it.

Thanks for the feedback Patrick.

I had not thought about the day/* pass option, so far I have been looking at hourly rates. Users will deposit funds to their account, and when the server is started the cost will be deducted from their balance. I definitely do not want to get in to utility billing, it seems like a headache I don't want!

Wrt the name - I came up with it on the spur of the moment and did not really think about it much after that. Cloud is indeed irrelevant to a lot of the target market, so MultiplayerOnDemand might be a better option.

1) I think you could charge in 24 hour increments pretty easily. Maybe the longer the server is running the cheaper it gets, so you could incentivize people to leave them up for longer periods of time.

For example (not real prices) maybe $5 per 24 hours and then $16 per week then $22 per month or whatever.

That way it's easy to take no risk at $5, but if you end up wanting to keep it up you're not paying $5 x 30 days which would suck.

2) I don't think you need to have the server end automatically. Letting me close it myself would be fine as an option, especially with the payment scheme I outlined above.

3) Let use prepayment. They fill up their account with $100 and when it runs out there's no overbilling. This adds a huge sense of security to the buyer. Probably also let people pay on-demand if they want by leaving a credit card on file.

4) Latency is soooo important. Offer servers in west/central/east coasts. Do not use shitty oversubscribed bandwidth providers!

If you do Urban Terror or BC2 I'll try it out for sure. Good luck!

1) The charges will be in hourly increments to keep things simple. I might explore different options in future.

2) I'm not sure about this one, I think it might need to be an option the user can set. I would hate to be in the position where someone asks for a refund because they forgot to turn their server off. It could be ok if there is a big warning that ensures they realise they have to pay if it is left on.

3) Yes, definitely pre-payment. Credit card on file is a nice idea, I will look at that in future.

4) Couldn't agree more. There will be multiple locations so people can choose the closest one.

Thanks staunch, I appreciate the feedback.

This is a really cool idea.

I started a game server hosting company years ago, and my biggest challenge was renewing existing customers. Customers rarely lasted for more than 2-3 months or so. My business's biggest enemies were customers' parents (a lot of gamers are under 18, after all), and in-game clan/community politics (groups break up, people switch games/communities, etc.). I know these things seem petty, but it's hard to achieve growth without strong renewals.

I suspect that a lot of potential customers hesitated during the buying process for these same petty reasons: they were uncomfortable with the risks of prepaying for a month (or 3 months or 6 months) of services, knowing they might lose interest in the game, have their clan break up, or get grounded by their parents.

If this is as serious an impediment to finding new customers as I think it was 7 years ago, then on-demand game servers may be the solution.

Good luck!

Thanks for the feedback and good wishes snikolic.

I was in a similar situation to your customers - I would rent an L4D2 server for a few months, then let it expire if I was spending "too much" time on a new game. This was the initial inspiration for Cloudfrag.

It sounds like you might have some interesting ideas and experiences in this area - if you would like to share, my email is in my profile.

Thanks again!

Very cool - my friends and I are fairweather online gamers, we really prefer to play with each other via co-op or maybe once in a while we run TF2 or L4D2 on a whim. An on-demand service would be great.

Feature request (not sure if it's there): would love to be able to start a server, and send out email invites to friends with clickable join links that would connect them to the server.

Related: if you need help with designing, sending, and managing emails (especially for that last feature), I'd be happy to help. I'm the Product Manager for PostageApp (http://www.postageapp.com) and we're all about email. Let me know if I can help!

It sounds like you and your friends are the type of people I am building this for.

Email invites are on the to-do list, although I am not sure whether or not I will enable that for the beta yet. I haven't thought about the implementation yet, but I will keep you guys in mind.

Thanks for the feedback!

Then please let me into the beta! Haha, no problemo though, I always dream about booting up my server and hosting for a limited time, but it's always such a pain in the butt to manage it all.

If you can do it at a reasonable rate, would be a very happy customer!

I like the look of the landing page...it's clear, concise, and descriptive. And the design isn't overwhelming. From a UI standpoint, I approve.

One thing that might be helpful is labelling the games you do intend to support. That was the first question that I had. I see at the bottom you can view images of TF, Minecraft, Left 4 Dead, etc...but it would help if you were a little more explicit. It has those 3 listed under "Games, Games, Games", but I would love to know about the "many more". Overall it looks really nice!

Thanks for the feedback!

I'll work on making the games list more prominent. As for "many more", I'll be posting a poll to see what gamers would like to play first.

Great idea Mike. A few years back when i gamed online more I would've been all over a service like this.

Have you ever used webex, gotomeeting, etc? It'd be slick if you could make the startup experience a bit like those so you can start your server and get the connection details into your friends hands in one step.

Question: can multiple people chip in some money for a server? Would it make sense to allow this? (E.g. a group of friends may wish to split costs.)

Yes, this will be possible - I will go in to more detail in a blog post, but essentially you will be able to create a group of people who can pool resources, and then pay for servers with funds from that pool.

It might not be in the beta - if not, I will add it soon after. I have a few features aimed at groups/clans, as anecdotal evidence suggests they are the people most likely to pay for a server.

Would personally love this feature. It's too expensive to pay hosting alone, just to play games once a week or so.

Clickable: http://www.cloudfrag.com