Ask HN: Is unlimited pricing a bad idea?
I have a SaaS that lets people extract data from video files. It is resource intensive and the more amount of work they bring the more it will cost me in terms of time, working with them, and etc. The hardware cost is manageable for the time being but I imagine if all of my users suddenly maxed out capacity it would cost me.
I offer an unlimited video processing plan for my small, medium, large plans. Is this a horrible idea? There is no reason for someone to move off the small plan into a medium plan.
Should I limit each plan by features instead and keep the unlimited processing? ex. small plan missing cool features but medium and large will have them. existing customer doesn't seem to care.
Or should all plans have the full set of features but different quota of the number of videos they can process every month? Something like a credit plan.
I've had some user say they will stop using my service if I remove the unlimited processing. This was their big reasons why they are using my service according to them. But some customers demand is just extreme for what they are paying (very little). Should I just tell them to go away?
In some ways I ended up getting customers directly from my competitors because of the unlimited processing. Is it too early to say this is working or this isn't working?
Cheers.
An open Ask HN community consultation from 2014 where @curiously convened hundreds of developers and bootstrappers to share real experiences, hard numbers, and tactics.
Hundreds of active indie hackers contributed case studies, pricing heuristics, and pitfalls to avoid when moonlighting on side projects.
Just general advice:
You want the cost to explore and adopt to be low.
You want the revenue of a retained customer to be high.
Unlimited pricing is a good sales tool when it helps minimize brain damage for your customer. They will over-value it if trying to figure out every possible use is scary, or difficult. In other words, it gives them room to experiment and uncapture hidden value (for them).
What you need to keep in mind is whether or not paying for this trial and error usage makes any sense. In other words, are they exploring your service? If so, they are likely finding value that once discovered they will be willing to pay a reasonable rate for. On the other hand, if they are exploiting you, they are just farming out their costs to you and this will never end (it always makes sense).
So, try to figure out why they want to stop. If they threaten to stop becuase of the brain-damage/experimentation issue,,,try to find a way to work around that. If they simply feel your service is simply un-economical priced any other way, you may want to expore that more carefully. In this case, you may have a bigger issue on your hands.
This gives you some frame work to think about features. Features which are necessary to explore vs features which allow simply "power usage". May want to treat those features and those stages of using your product distinct.
Some misquotes from the articles of patio11 (I hope there were a button to surmount him and get the real advice. You can try reading: http://www.kalzumeus.com/greatest-hits/ ↗ and subscribe to his "mailing list" aka "training course by email".)
* Double your price. (You are probably charging too little. Grandfather existing clients.) If the plan is too cheap, you get too cheap clients. (In one of the last article he explains that he had to discontinue the cheapest plan because it attracted bad clients.)
* Change the names of the plans from Small, Medium, Large to Hobbyist, Professional and Enterprise (or something like that), so the clients autosegmentate. (They know that they are an enterprise, but they are not sure if they are medium or large.)
* Put usage quotas but don't enforce them. Use that information to upsell the customers.
* Try to find the enterprisely features and add them to the Enterprise plan (I'm guessing here: allow multiple accounts to see a half processed, read only accounts, save usual processing filters.)
IMO if you do an unlimited plan it is generally wise to caveat it with a fair use agreement. You can word those in different ways, but if you really look at almost every successful SaaS out there you will noticed they all do this in some way. This stops a few clients from abusing you and allowing you to increase their pricing based on some factor.
2 other points:
An unlimited plan that is losing money is never good. Just because startups backed by VC money can loose money for a few years while building scale does not mean it is ok for the bootstrapped business to do the same. Depending on your situation and opinions it will vary.
The other point is that while A/B testing can be overused and useless if done wrong, this is one of those places where it really does make a positive difference. It will complicate (increase) your number of subscription plans, but it will let you test and hone in on the right mix of pricing without alienating existing users.
So unlimited processing is your core strategic advantage? Is it just the psychological effect of "unlimited" that's important? If so you could have a soft-limit (think bandwidth-throttling). Or describe the actual limit in language that makes it seem bigger.
The business model of "promising" more capacity than you actually have is pretty common, especially with cloud storage provides. To some extent every SaaS does it. As long as you do the math and structure pricing accordingly you're fine.