You’re Pricing It Wrong: Software Pricing Demystified | Hacker News Reader

You’re Pricing It Wrong: Software Pricing Demystified

smashingmagazine.com

160 pointswillfarrell27 comments

One point that I rarely see made in this type of valuable articles, is how pricing affects your ability to do paid advertisement.

Let's say that for your particular niche, you pay on average $1 per click. These are the expected conversion rates to break even, depending on the price of your product.

$1 - 100%

$2 - 50%

$4 - 25%

$8 - 12.5%

$16 - 6.25%

$32 - 3.125%

$64 - 1.563%

$128 - 0.781%

In general, it's far easier to convert 1.5% of your visitors, with a product that costs $64 than it is getting 50% of your visitors to pay $2, 25% to pay $4, or even 12.5% to pay $8.

So in my experience, charging a premium has practical implications when advertising, that go beyond pricing as a quality indicator.

The lifetime value of a new customer must justify the cost of paid acquisition channels, and leave room for profit.

This is true with assumption that both products are offered in same segment of the market where cost per click is same. Often, products with $2 pricing are offered to much wider demographics than products with $64 pricing. For lower priced product there can be bulk targeting strategy as well where word of mouth and viral sign ups can also be seen. Also as pricing increases you go into B2B sales where complete different marketing/sales strategy needs to be adopted.

Let me try challenge that. For every iPhone app purchaser that bickers over a $3 price, a fairly pricey iPhone has been sold. Those are the same human beings but acting in different ways depending on the combination of perceived value and price. They're the same demographic, the same B2C target market.

Apple sold a seriously expensive (especially initially) device to people who would later complain about 2 bucks. They owned word-of-mouth and viral. Any Nokia user seeing their friend using an iPhone would immediately calculate the months to the end of their phone contract. There's been so much free marketing that it can't possibly be calculated. But the product is not cheap.

Our challenge is to find out how to harness that for our own software/services. First step is to avoid boxing ourselves in, e.g. by price.

  > Objective Value:
  > (Hourly rate × Development time in hours) − Price = Value

Um, no. This assumes the value of the product is based on the inputs, and is incorrectly modeled from the seller's perspective. Any product's value is determined by the buyer, who is the one making the purchasing decisions. The value is not in how much time the developer put into it, but what it does.

Yes, the article is about subjective value (and finding it), not objective value, but taking the developer's (biased) perception of value as a starting point is a bad idea.

Pricing should be determined early in the development process to inform go/no-go, and the amount of development effort to apply. Starting from scratch on pricing after the product is ready is backward.

Agreed. Value, price, and cost are three independent variables. When they line up advantageously, you have a viable product. But one does not determine the others.

Good point. There are actually four variables - one of them is usually money though, so we don't think about it that much. The four are

   buyer price (= contracted price + opportunity cost to buyer)
   buyer value (what the product is worth to her)
   seller cost (cost to create product + opportunity cost to seller)
   seller value (value of what he will be receiving in return, usually face value of money)

One of my favorite anecdotes about that comes from an article found here (can't remember the exact article): The author (a developer) went along with their lead salesperson on a call to MIT. After demoing the product, which normally sold for $150k, the MIT rep asked about the price. The salesperson immediately said "$1 Million" - the MIT rep chuckled and said, "Oh come on, that's ridiculous - you know we don't pay more than $650k for software." The salesperson responded, "Ok, we'll let you have it for $650k - now let's talk extras."

So yes, price is a function of demand first - don't tell the customer what it's worth, let them tell you.

I think the author was Steve Blank. I have an MP3 where he mentions this happening, but I can't for the life of me remember the name. You can probably find it on IT Conversations.

I was expecting this to end with a couple of kids at MIT built the thing for free and open sourced it.

I think you're referring to this story: http://venturehacks.com/articles/pricing

You misread that - the hourly rate and dev time in hours refer to the buyer (a developer) having to develop a functionally similar product (instead of purchasing it).

That's how I read it as well. It doesn't matter though, it's still the wrong way to look at pricing. Value, price and cost are not at all related.

"This rectangle represents the calculation of sales x price, and the biggest rectangle represents the biggest profit."

Err, no. The biggest rectangle represents the biggest revenue. Profit != revenue, unless you have zero costs.

Yes. I've seen a lot of folks that confuse profit and revenue, and most of them are living in a bubble of some form or another.

To the author's credit, my guess is that he's living in the bubble of sales&marketing. Their assignments are usually something like "Go out and sell as much as you can!" with little regard to product cost, production limits, or even cost-of-sales.

I'm guessing that is why he makes such an elementary blunder.

Isn't the marginal cost of software effectively zero?

Not if you operate a free help desk, a support web site, etc.

Not to forget bandwidth and other variable paid expenses.

I'm a bit surprised that the article mentions support, but at no point makes the connection that more customers == more support requests.

I think I'd rather have 5 customers paying $20 than 10 customers paying $10 - is this a thoroughly bone-headed point of view? Or is supporting more customers essentially a marketing cost?

This all depends on the type of customer. If your customers are high maintenance customers that require a lot of support, then having more customers paying less money may not be worth it. However, more customers often means more diversity so you're more protected against a few taking their business elsewhere. And then more customers might mean you have to hire more people which can affect the culture of your company. Seemingly simple decisions can run pretty deep when you think about the true cost.

So it really all depends, but it's definitely not bone-headed.

If you're interested, here's an article I wrote for Inc. that discusses some of these ideas: http://www.inc.com/magazine/20101101/go-ahead-raise-your-bus...

A great point. Some companies price in their very excellent support (Nordstrom comes to mind). At the other end of the spectrum, I worked with a guy who ran an wifi internet service on the side. His prices were so cheap that he refused to help clients troubleshoot and offered only two support options: cope or get a refund.

More customers can also mean faster growth, as you are getting more word of mouth referrals.

There can also be an element of customers paying more will expect more from you (ie. support).

It's discussed in the article, but I would recommend reading "Don't Roll the Dice" which is a free e-book on software pricing by Neil Davidson: http://www.neildavidson.com/dontjustrollthedice.html

(Neil is CEO of Red Gate Software and co-founded the Business of Software conference.)

       "Apple charges a premium because of the perceived value of its products"

Is this necessarily true these days? Air vs Ultrabook pricing for example?

I take the general point. In the UK college education sector there are two main players for providing virtual learning environments. One is open source, the other has a lease contract price in the tens of thousands per year (depending on the number of seats). Both are widely used!

Testing prices sounds great, but how do you test prices for SaaS subscriptions? I can't just lower or raise prices without making at least a moderately big-to-do about it.

A great question. The only thing I can think of is starting at a high price and then lowering it slowly. You definitely want to do some split testing to see which plan pages get you the highest conversions, too.

Start low and raise them gradually. Grandfather existing customers at the price they had when they signed up. If you need to cut prices, cut them for existing customers and new customers. It's really not that big a deal unless you're out of the startup phase.

Agree 100%. I did this with a consumable online retail product. It works well. Not only do you find the right price-point, but it gives you a solid base revenue from loyal customers to finance overhead and test marketing.

Reply on news.ycombinator.com