Hacker News new | past | comments | ask | show | jobs | submit login
Don't Blindly Model Your SaaS Pricing on 37signals (ginzametrics.com)
149 points by rgrieselhuber on Nov 16, 2012 | hide | past | favorite | 62 comments

Entire books are written about pricing[1].

It's time for me to point out, again, that no other business parameter will affect cashflow and profitability more than your pricing.

You can adjust your costs all the time.

But prices are sticky. Customers remember that something used to be cheaper.

Really, again, let me emphasise this.

Pricing is the single most important business decision you can make.

Rails vs node.js / Ruby vs Haskell / bootstrap vs foundation / backbone vs sproutcore / trendoid vs hipster / cats vs dogs?

Only tangentially relevant to your success.

Pricing? Critical. Central. Essential.

Do not leave it to gut feelings.

[1] http://chester.id.au/2012/09/12/review-the-strategy-and-tact...

Pricing is secondary to product.

If you don't have a good product, nothing else matters.

Once you have a good product, then you focus on pricing.

Once you figure out the product and pricing, then you ...

If you want to sell services to large companies--even low cost ones--you need to offer invoice payment. At a big company, the corporate credit card is for meals and travel, not for paying vendors.

In my experience, if you send a proforma invoice, a W9, a federal tax id, and a soul sourcing statement, at the end of the signup form, many times a check will just show up in the mail. It is possible to do invoice payment without having a high touch sales process. You just have to preemptively provide all the stuff the AP department needs.

I really think 37 signals is leaving quite a bit of money on the table by not offering PO payments. I understand they don't want to do a whole bunch of customer hand holding, but it is possible to accept POs without ever talking to a customer.

soul sourcing statement

I think you meant sole, but I had to laugh at this meaning as well in the whole scheme of dealing with finance departments.

Just dreaming here, but is it possible that this could be outsourced, the way you can outsource credit-card payments with Stripe? Have another organization invoice them, collect the check, and then deposit the money into your account, without you ever having to make more than an API call?

Interesting idea. How much would that service be worth to SaaS companies?

Enough to give me a business plan ;)

Would you not have to customize this for every country that you do business with, though?

Hmm. Sounds like a complete and utter horribly painful schlep. A schlep that puts you right in the path of the money. Interesting.

You don't have to customize anything. If you don't want them to call, you can add a sentence that you accept (or don't accept) foreign currency checks at a market exchange rate. That is about the only question foreign businesses ask.

"Where there's muck, there's brass".

> You just have to preemptively provide all the stuff the AP department needs.

Do you have an example of a SaaS operation doing this well?

I don't know for sure, but I imagine Salesforce has this down.

This is a really good idea.

I intend to use invoice payments as a segmentation device. Want to pay by invoice? Expect to pay a lot more.

(I think I picked this idea up from Patrick McKenzie -- another one is offering SLAs to expensive clients).

This is true even for medium sized companies.

And for schools or any other organization.

Another great point. There's so much obsession with billing credit cards I've never seen this mentioned in any pricing blog.

"Once you get a lead, if it takes even a single phone call to close them, you’re no longer doing self-service and Customer Acquisition Costs vs. Self-Service pricing is going to kill you. If you’re not doing that phone call and you’re also not closing those deals, you need to figure out of if you can solve the problem through conversion optimization or if you really do need to be speaking with customers in order to close them. "

Wow, can you bold this? Touchless sale/conversion = win.

David Skok talks about it brilliantly and has an eagle eye on these metrics when evaluating his SaaS investments - http://www.forentrepreneurs.com/business-models/the-touchles...

No one really talks about backing into pricing based on LTV/CAC ratio (affected by for example measurable inbound marketing or cpc/cpm costs).

Pay back period (ideally CAC < a year) is also a factor. http://www.forentrepreneurs.com/saas-metrics/

I think the OP makes some very important points here. There's definitely a somewhat fixed mentality when it comes to SaaS pricing and it's easy to fall into the trap that you should be charging something akin to 37signals.

Pricing isn't actually something that's as easy to change as you'd believe either. It has huge implications on your actual product, your target market and sales strategy - as the OP points out.

Imagine you were suddenly told you had to charge 10x more for your product. You might initially think it would kill you but think about what that would change. Would it necessarily be a bad thing?

If you're anything like me, it's easy to assume everyone lives in a world where $10 for Spotify is a pretty big deal. My friends moan about that price for unlimited music. $200 a month seems huge to me for anything I'd buy. It's only recently I've fully started to appreciate just how much money enterprise clients willing drop on "simple" products that meet their needs.

I really think there is a lack of good advice on pricing and how alternative pricing strategies could be used to open up new markets for SaaS type businesses. Sure, you may need a slightly "less scalable" sales strategy but with increasing competition in the web-app marketplace, moving to a different customer base with more cash may well be worth a few sales phone calls.

"If you're anything like me, it's easy to assume everyone lives in a world where $10 for Spotify is a pretty big deal. My friends moan about that price for unlimited music. $200 a month seems huge to me for anything I'd buy. It's only recently I've fully started to appreciate just how much money enterprise clients willing drop on "simple" products that meet their needs."

For businesses those SaaS services are a core need. Leisure services like spotify are the wrong thing to compare prices to. Think about how much of your budget you spend on housing, food, transportation. $200 / month is peanuts for those kinds of things. That gives a better idea about what businesses are willing to spend on software they need for their core activities.

That's a really good way to put it.

"It's only recently I've fully started to appreciate just how much money enterprise clients willing drop on "simple" products that meet their needs."

Not just enterprise clients - almost any business.

I run a tiny, cash-strapped film production company and do various bits of Internet Marketing stuff on the side. We're about the smallest entity you could imagine that can still be called a "company".

I'll cheerfully drop $100 or more per month on a service that will save me a day a month of work or increase profits on a project by 10% - because in both cases, I come out ahead.

Consumers are buying on "value", on gut feelings and emotion, and all sorts of other things. Businesses are buying on "will this make me more money than it costs?"

Hopefully this is useful. I've seen very little discussion on pricing models for SaaS in the startup community.

I would have liked it better if you had actually talked a bit about business. A lot of people who hang out here are engineers / developers who don't get a lot of exposure to that side of things.

Start by explaining a business 'model' which goes like this:

    Revenue - Cost-to-deliver = Gross Margin
    Gross-Margin - All-other-costs = Net Margin
Its a sad bean countery topic, but really its not that hard to see that if your revenue is less than your burn rate you die. And your revenue has a 'unit', whether its widgets or users, each 'unit' brings you $X in revenue. And somethings are dependent on 'units' (how many servers you have, how many customer support people you need, how many twitter tokens you buy, ...) those go into your Gross Revenue calculation, because growing revenue means growing those costs and Gross Margin remains 'constant'. Then there are things like salaries for engineers, rent on the offices, replacing laptops or servers every 3 years, maintenance, janitorial, lunches. Those stay the same regardless of your revenue so your Net Margin starts low and then gets bigger as you get more revenue.

A business "model" takes the Gross Revenue, and then fractionally allocates it to those other expenses. Maybe 20% of the GM is engineering, 10% is is marketing, 10% is sales commissions, 5% is facilities, 3% is swag/giveaways, maybe 10% goes into a bonus pool that you divvy up among your best performers. The remainder is 'net revenue' or 'free cash flow'.

If you're a C-level executive and you don't know what every single number on your tax return means, you're running the risk of having a problem staring you in the face that you never "see". If you're an engineer and you see a bunch of patched together spaghetti code that is poorly documented you "see" that there is a huge amount of technical debt that is going to have to be paid before you can move on to V2.0 or what not. If you're looking at your taxes and you see that you're depreciation costs are of the same magnitude as your revenue you need to be able to "see" that a cliff is coming when you're going to need new gear and you'll have no money to buy it.

The biggest challenge of "pricing" is that sometimes you realize that you can't sell your product for what it is going to cost you to produce it. You have a choice, either innovate around the costs to make it, or put on a really good show and try to sell it to someone before the truth is out. Hard place to be but knowing your costs will inform you on whether or not your pricing makes sense. The 'best' situation is when you can sell it and your costs scale fractionally with users. That is a very good business to be in.

Agree with most of your point Chuck. However, I think most people on HN are in the startup phase. They are doing customer discovery/development to find something that is actually valuable. This post is really optimization in the execution phase. It is normal for most of the engs/devs to not understand the business side.

Agreed, but if you're the founder, even if you're just one guy and an idea, you need to be able to evaluate your options.

Example, founder says "We launched and got 100K users in the first weekend!" That is fabulous, shows that they have really connected with their target market, but do they know how much money each user has to generate to pay for their expected burn rate? They should. You can say "We're build X for Y, and we're looking to monetize the landing pages with advertising." So what "CPM" do you need? (clicks per thousand) What RPM (revenue per thousand) do you need to make that work? Are there other businesses that have similar CPM/RPM numbers?

We're fortunate that a lot of folks can and do share their numbers with this community. I'm just encouraging founders to keep the whole equation in their thoughts between costs + engagement + revenue so they can think about ways to test against their models.

The three questions every founder has to wonder; Can we build it? Will people use it? Can we convince them to pay enough for it to sustain it?

I did make some assumptions about people understanding the core parts of any viable business model but the way you explained it here adds some great points.

As you mention, it's hard to know what your costs are going to be when you're just starting out so it's really important to be flexible enough to change your pricing / value prop to your customers as you grow in order to cover your costs and provide room for further growth.

What is transactional? And how is it different from enterprise?

That's explained really well in this ebook:


But, briefly, the transactional model is one in which you require some customer communication in order to close the deal and is typically closed via inside sales reps. Enterprise deals are larger, cost more and typically require on the ground reps and a lot more customer facetime in order to close deals.

That's interesting. In the freight, and oil services industries, transactional is often used to describe spot market arrangements versus contractual ones. I think that makes sense, but I need to think about it a bit more as it relates to rivalrous goods.

"Don't take pricing advice from Hacker News"

Huh, I've seen more recommendations for higher prices than lower on here lately.

Thanks for the post, it was the push I needed to get off my butt and run a pricing test, if only a weak one. I've been told I'm charging too little, and I've been told I'm charging too much.

Just to see if I'm leaving money on the table (and hurting my ability to grow by undercharging), I made the one-line CSS tweak to hide the cheapest pricing plan on http://www.improvely.com/pricing

If that doesn't hurt signup rates, I'll have this post to thank for it. If it does, it's easily undone.

Glad to hear it was helpful. Improvely looks fantastic, btw. I love the UI.


Really great article and very refreshing to read. We recently adjusted the pricing on one of our services http://www.advantly.com to include a lower price monthly package and 2 other value added packages that require annual subscriptions. It has proven to be successful for us and the subscriptions to our annual packages have increased.


One thing I found really confusing : On the gold package what does select features mean ? What can I select ? ( I later realized that it's the select package features but it took me a few minutes :) )

Thanks for the feedback on that. Maybe we need to rethink the name for that package. Now that you point that out, I can see how that can be confusing.

Pricing is complex in general, but for SaaS companies even more so. Enterprise SaaS companies spend a lot of resources to acquire customers. Annual billing to a large extent helps offset these costs. However a question remains: How does a new unfunded SaaS company acquire customers, when annual billing can be a big obstacle for sign ups?

You remove as many barriers to the purchase as possible. Allow monthly credit card billing, but incentivize longer term payments. For example, PivotalTracker.com charges monthly. However, I personally paid them for an entire year to get the 2 month discount! 12 months for the price of 10 months is enough encouragement for me to pay for the entire year.

The discussion around annual pricing seems misguided. You should charge customers in whatever method they prefer to pay. For some customers, putting $400/month on a card is much easier to deal with than a $4800 check. The oposite is true in other organizations. Requiring pre-payment for software and using customers to fund working capital to pay sales associates seems myopic. Get the working capital from somewhere else and do whatever is best for revenue.

Very few companies are going to turn down revenue, regardless of the shape it takes, so if customers are unwilling to pre-pay for a year, it's easy to figure out a way to make it work.

That being said, the startups we've spoken with who have switched to annual pricing are able to get paid in advance 60-70% of the time. This helps a lot.

Some companies prefer this as well as it can reduce their own AP and record-keeping burden.

Annual pricing can be a great thing -- simplification and money up front: what's not to like? But anyone being paid long in advance of when a service will be delivered should keep in mind that, from an accounting perspective, some portion of that pool of advance payments may be considered a liability, and can complicate revenue recognition. As soon as you receive a payment, you now owe the customer a year of service instead of, perhaps, them owing you the cost of a month of service.

Rev rec definitely gets more complicated, this is a good point. Everything is a tradeoff.


Cashflow, cashflow, cashflow.

Getting paid up front is better for cashflow than getting paid monthly.

Not to mention there ain't no chance the customer's skipping to a different service in 6 months.

His math is completely skewed. 22 sales a month of a service that charges $200/month will generate $343,200 a year (unless he's assuming 100% churn rate).

At the end of the year, yeah. But at the end of the first month, in order to be able to pay your salesman, he must have done 22 sales. By charging yearly, only 2 sales are needed. That's cash flow.

So yeah, if he does his 22 sales / month, at the end of the first month, you're even, and starting from month #2, you're making money.

Combine this with "It's easier to sell one time a $1000 product than 1000 times a $1 one" and you get the main idea.

Good point! I was talking about startup costs for the first few months of hire as the rep ramps but didn't make that clear at all. I'll update soon.

Edit: Updated now.

Your math is still way off unless you assume 100% churn rate. Assuming a reasonable churn rate in the 2nd month you will have some fraction 22 existing customers paying $4400/mo (not really sure how $4400 breaks even with $4800).

The chart indicates that one can avoid the startup graveyard by raising prices. Is that okay?

Pricing low has nothing to do with being rude. That statement is a major disconnect from reality.

I also think that there are many service out there that are buying into this 'high pricing' doctrine and are giving prospective customers sticker shock. These starving young companies can't afford to do that so early in the game.

The other day I saw a site on HN selling business ebooks to startups: "how to drive more traffic to your website" $99, a customer segmentation ebook for $39, a copywriting ebook for $49. All this stuff is way overpriced. Dozens of quality articles can be googled up in a few minutes for the same information. I know because I did just that.

The people that tell these young startups to jack their prices are not being helpful or even honest. Maybe its a way of trying to justify their own high pricing.

I suggest young startups get users by any means possible, even if it means discounting heavily. They need habitual users, and that requires making it affordable enough that the enduser uses the tool / service for a long enough time that it becomes a habit and the tool becomes too entrenched in their workflow.

If anything is rude, it's telling young startups how to price their businesses when it looks like they don't have a clue on pricing even their own.

How many copywriting ebooks have you purchased in the last quarter? What is your budget for enterprise software?

If the answers are "zero" and "none" then is any price I could possibly quote to you giving a potential customer sticker shock?

It's probably over $100 this year that I spent on copywriting books, and almost that the year before.

I will pay for information if it will help me, but the examples I've given is very basic information that's been written about for free so many hundreds of times, I'm surprised anyone would buy those ebooks.

They are ripping off the very startup community they say they are helping.

I think there's a pricing bubble going on with startups. Many monthly subscription rates are too high, especially when there are so many choices available to the consumer, and at a time when the budgets and attention span of buyers is so limited.

There seems to be no imagination in pricing, only to charge more and more. It's bad advice and I'll bet its mortally wounding a lot of young startups and impressionable kids that are trying to make a go of it.

Again, it's a con game and it's trying to gouge and profit off the start up community they say they are helping. It's shameful, but its only time before the herd snaps out of it and realizes what is happening.

There's a difference between free information and compiled free information with a bit of extra. You pay for the premium of having it all put together and not having to spend the time to go find it all. Why buy a programming book? There's lots of tutorials on the internet. Still, many people buy books.

If I were completely new to a topic, like copywriting, I would probably look for a book on the topic, rather than blogs. That's because I'd spend more time trying to find free material that's good (there's a lot of free crap out there), than to buy a pre-compiled book from a reputable source.

I agree with you that the materials have to justify their price. I've purchased a $99 information product before that was total crap, and the exact same content on the same person's website, just reformatted. In that case there's no justification, no added value. I don't know what books you're talking about to determine if they're actually any good or not, but to charge for what's normally "free" information there has to be added value.

Maybe there is a rash of startups with inflated pricing without any added value. However, as soon as the first competitor comes in that truly adds value for the same price (or less), then everyone will jump to it.

I'm not going to link to these ebooks because it doesn't deserve more attention, but it was front page HN only a day or so ago.

I will say this though: certain opinion leaders on HN are taking advantage of the readers and the prevalent herd mentality that prevails here.

RE: "I will pay for information if it will help me, but the examples I've given is basic information that's been written about for free so many hundreds of times"

I suspect your google-fu is stronger than most. Many people don't have the skills or time to do this. Should I spend 2 or 3 days searching and building a marketing strategy that works in 2012 by filtering through all the stuff written over the last 4 years and half of it is out of date or should I pay $99 dollars to get it in my inbox in PDF form. For many the $99 makes sense.

They are certainly not ripping off the startup community! If you even value your time as a startup founder at $100, it's cheaper just to by the dumb book.

Most startups doing subscriptions are doing b2b or b2h (business 2 hacker examples: github, seomoz, etc) business. Businesses and to a much lesser extent hacker don't care about the cost, but look at the value provided.

RE: "There seems to be no imagination in pricing, only to charge more and more. It's bad advice and I'll bet its mortally wounding a lot of young startups and impressionable kids..."

This is absolutely absurd! This is like saying "There seems to be no imagination in cars today, only to get better gas milage more and more". Of course, that's the point of a business is to charge so much your customers complain, but still pay you none the less. You're capturing the most value possible, while they still enjoy a net gain.

I'd argue even if this is bad advice for a majority of people, it's still good to get it out there that it is possible. Once we know it's possible, much like the 4 minute mile, we can achieve it. When it's locked away and only a choosen few know it is possible that is the real danger to impressionable kids.

I don't have special google-fu abilities. I just notice there's a glut of ebooks and courses out there that are rehashes of older content - and they are priced way too high. They are priced so high that those who need the information most cant afford it.

Stop drinking the Jonestown kool-aid that's been going around my friend. Your time isn't so valuable that you can't spend 15-20 minutes doing a diligent search online.

The ebooks I mentioned in the previous post were college textbook stuff. As basic as the stuff gets. You would need to be already under the influence of the kool-aid getting passed around here to think that's an honest price. It's not.

My point is this: the people trying to sell you stuff here are not your friend.

By being on this site you most definitely do have special google-fu abilities. The vast majority of people do not know that the top search result in yellow is an add and that you can do "exact match" searches.

If you can find 40-pages of well documented information and condense it down into clear, actionable information that you can then form a plan from in 15-20 minutes you are a much better man than I am.

For those of us that lack that particular skill set, these are screamingly good deal even at $99. For even more of us who don't personally pay for this, but put it on the company credit card, our company should fire us for wasting company time by trying to do it ourselves and not spend the $39.

Clearly, we disagree, but all I'm saying is that there are a lot of people for which these are a screamingly good deal.

You are substituting what you think are acceptable prices for what the market will bear.

There's a little more to it than that. If your not priced to keep users long enough to get addicted to your offering, the cost of churn and marketing spend trying to get fresh customers will kill you.

There is a lot of choice out there. Everybody is executing nowadays, and it's the same sort of apps that are flooding the market. The advice in this post is not helpful.

Most of the money in the economy is tied up in the capital structure -- ie, it's not actually in the hands of consumers, it's in the hand of producers.

Producers can and do pay a lot more for goods and services if it helps them to make a profit.

The market for todo apps and other consumer software is, you're right, absolutely a mad scrabble for volume.

The market for intelligently targeted vertical segments is juicy and prime for picking.

I know where I'm aiming my current venture.

Why is $49 for a copywriting ebook overpriced? Speaking personally, if I can get one useful idea from a book on copywriting, the chances are it's going to pay for that $49 somewhere between 10 and 100 times over, perhaps more.

And many copywriting books I've bought had 2 or even more useful ideas in them.

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