Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: [Pricing] Charging a one-time fee for a SaaS app?
30 points by sandeepshetty on Sept 5, 2012 | hide | past | web | favorite | 28 comments
Do you know anyone that charges a one-time fee for a SaaS app like pinboard.in does? When do you think/know it makes sense to let someone use your hosted app indefinitely for a one-time fee?

Yeah -- I've always sold Bingo Card Creator on this model. It is modestly less crazy than you think it is: if you're getting customers on an ongoing basis, attrition plus the natural near-zero marginal costs of servicing customers means the new customer keep the lights on and older customers get to freeload indefinitely.

Suggestion: make something that will let you charge on a recurring basis. (I once thought BCC could never sustain that. It probably could, if I had a mind to implement it, but the pain involved isn't worth it to me.)

Starting from $0 revenue on the 1st day of every month sucks. Starting from "I'll always have at least 90% of the revenue I did last month" is, on the other hand, a wonderful thing for the business in every possible way.

I've been thinking about this from the perspective of Customer Lifetime Value (CLV). Charging a one-time fee based on historic CLV data, IMO, has three advantages: 1. Customers perceive the indefinite nature of recurring payments more expensive than a one-time fixed fee, 2. I'll have more cash upfront, and 3. I'll make more from customers that might have unsigned up long before the average CLV point.

1. Customers perceive the indefinite nature of recurring payments more expensive than a one-time fixed fee

This is, for better or worse, exactly the opposite of how customers actually think.

I'm seeing this behavior especially for addons. When customers are paying a recurring fee for the main app and the costs of addons add up, they seem to prefer a fixed one-time fee (not necessarily equal to the CLV).

Good data, thanks.

That said, there was a recent blog post about how one HNer dramatically increased his revenue by changing his pricing model. Rather than charge for features, he adopted a few tiers of service, each one targeted at one type of customer's use case.[1] I don't know your exact situation, but that might be worth trying as opposed to configurable add-ons. In general, speaking in the language of use cases resonates with the buyer more, and is more likely to result in a purchase.

[1]: http://www.extendslogic.com/business/what-i-learned-from-inc...

You're doing pricing wrong then. If product A creates $300/month in value and you charge $50/month for it, and the user needs addon a for $15 and creates $8 in value and addon b for $35 a month but creates $20 in a value a month you're better off charging $150 a month for the package that has those addon features.

Maybe I'm unusual, but I just shelled out $350 for a smartphone in order to save myself 50% off of my phone bill by using a prepaid service.

When I decide whether or not to purchase or subscribe to something, I do the math to figure out which is cheaper in the long run. That being said, if the price difference is negligible and all else is equal, I would probably opt for a recurring payment.

You're unusual (no offense, so are most/all HN readers). The vast majority of smartphone buyers with a choice choose the subsidized (more expensive LTV) phone.

Wow. That's interesting. I'm sure you probably have data validating this but how does anyone explain it?

Is it because a recurring payment implies they only have to pay as long as they are using the product?

Take SEOmoz for example. I subscribed to them for a while.

I used it for the website: http://http://metalranchandfarmsigns.com/, which on average before Google killed me made $200 - $300.

In any case, my thinking goes like this: If I had to buy SEOmoz in one lump sum for rest of my life the value of the product to me is probably close to $5000. That's what I'd pay for a lifetime subscription just for MRFS.

I can either pay $5,000 (hypothetically) and have it forever, or slowly enter in at $99 a month. If I cancel within 4 years I win, if I cancel after 4 years SEOmoz wins.

I cancelled in a few months.

From my experiments 1 is definitely false. Customers often are willing to pay $50/month for several years when they wouldn't pay a $500 one-time fee.

This goes back to the concept of reducing risk. Customers are very risk-averse, so things like charging a lower monthly fee or offering a money-back guarantee will often give you a significant increase in sales.

See my reply above to patio11.

but the pain involved isn't worth it to me

Do you mind if I ask why not? It seems like something that should make a whole load more money, and if it puts people off, changing the code to say is_subscription_valid() {return TRUE;} and switching back should be trivial.

Implementing in the first place shouldn't be too painful if you can steal some code from Appointment Reminder.

I know I'm underestimating the total amount of work here, but is it really going to take more than a couple of days? Unless there's crazy invoice integration work required, in which case I understand - that's a painful world!

Edit: although I appreciate that consulting rates way outstrip revenues from SaaS a lot of the time. Still, it's nice to have some income whilst you're on holiday ;)

It's largely a question of messaging this change to customers, dealing with approximately 7,000 existing customers and answering "No, no, don't worry, YOU still get it for life!", changing marketing copy site-wide, yadda yadda yadda... and then, pot-of-gold-at-end-of-rainbow-style, getting a few thousand bucks extra in 2013.

Call it, oh, two weeks of work. Two weeks at my consulting rate is... umm... a more reliable method of getting a fairly sizable chunk of money up-front, if I am interested in getting more money right away. Two weeks of work on AR, by comparison, gets me fairly little money right now but gets me much closer to where I eventually want to be than the same marginal work on BCC. (e.g. I did two weeks of work on AR in the last two months, and increased paying customers by 50%. Not like rate of acquisition, like total. Feel free to guesstimate how much that is worth to either a) my income in 2013 or b) the implied value of owning 100% of the shares in AR. It's a heck of a lot more than either a) how much money BCC could reasonably make in 2013 or b) the delta in implicit sale values of BCC. There are other projects I could spend two weeks on, with a variety of professional and personal goals attached, which all sound more fun to me.

As of September 2012, BCC is a fun laboratory I get to play in (and draw stories from) more than it is a core line of business for me.

Interesting, thanks - I'd completely ignored the messaging the side of things ;)

If you were doing Bingo Card Creator today, would you have went with a subscription model instead of a one time fee?

I basically cloned the Bingo Card Creator idea in that I sell access to a PDF generating machine. I though long and hard about a recurring fee model, but decided that my customers are price sensitive, and to be fair, I'd have to charge $5/month rather than a one time fee of $30. Sure some people would sign up and continue to pay, but most would cancel after one month and I'd only get $5. I know this because 90% of my customers never log in again after one week.

Additionally, I think a recurring fee would scare customers off in this market, and I go out of my way to tell them it's a one time fee and that they get access forever. If you're wondering, "Do some of them feel a sense of entitlement for essentially... forever?" Oh boy yes, they do!!! But that's OK, I treat them real nicely anyhow.

Have you tried making the $30 an annual fee?

Have you tried making the $30 an annual fee?

One thing to consider is that annual fees have the potential for pain, purely given that lots of people forget about them. The only complaints I ever received about TweetingMachine were the people who were rebilled a year later, the tool having left their memories six months prior.

Sounds like a good idea for an A/B test.


Why not add it on as an upgrade? Sign up for all the cards you want only $59/year.

How much would you charge and over what period?

This is a bad idea. Any customer who really likes & uses your app as-is will eventually become unprofitable for you. Your "best" customers will use resources you have to pay for every month, but you only collected N months of revenue up front. You'll eventually have to decide to pull the plug and screw your users (like Joyent/Textdrive's forever hosting). Even before that, your financial incentives will be at opposites to the people who like your app the most, which is a bad place to be.

Ideally, you want your customers' usage to align with more money to you over time, not less.

If you need money now, offer your customers a discount for quarterly/annual prepayments. You're more able to predict your costs out 12 months, so you're less likely to make a fatal mistake here.

Charge yearly a X%(go figure it out what works best) of what you would charge one-time.

Charging a one-time fee is evil for the customer with common sense. If your service stops growing, what about the customers that already paid for it?

You don't pay for updates in a SaaS app as it's in the cloud, so... I can only truly see a recurring model working. UNLESS you know very well your userbase and you know very deeply it's the only model that would work.

I've thought of doing a slight variation on this for SaaS apps that are project based - charge once for each project for something like six months or one year access. I think it would work in situations where the customer only needs use of the product for each project for a defined amount of time.

If you've got ongoing expenses (e.g., bandwidth) you'll need to make sure to charge enough that income from the one-time fee is enough to cover the monthly bills + profit, in perpetuity.

Annuity tables (or the annuity formula) may be helpful here.

Don't do it.

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact