Nick, this is extremely impressive and I applaud you for putting this together so concisely. Some comments:
1. Kahneman and Ariely both mention in their writings how their is a difference between free and even just 1 penny. Especially given how popular Freemium is I thought this might be a good topic for you to cover. (Ariely calls this "market norms" vs "social norms")
2. A common theme here on HN (mainly from people like Patio) is to "charge higher". I couldn't find a place that explicitly mentioned this. Many people perceive higher quality and value of products and services simply because they cost more (regardless of whether they actually do or not)
3. No mention of Kahneman?!?
This is awesome regardless...saving it and sharing it with some clients.
Feel free to contact me directly if you want me to elaborate on any of this.
1. I remember reading about that concept in Ariely's book. And I actually considered adding it to the list, but I wasn't sure where to incorporate it.
2. I think everything boils down to your ability to communicate the value of your offering. People will pay any amount, as long as they perceive to be getting more value from it.
3. And I can't believe I didn't mention Kahneman anywhere! His research was a large foundation behind a good portion of those techniques.
Tactic X: "Give away free product or usage of your service" (talk about Ariely's theory of social norms vs market norms)
Tactic Y: "Up-sell free users into a more compelling value proposition" (distinguish between the free product/service and use it as a framing exercise to get your users to pay)
1. Barrier to entry is very important. Games have an investment in learning the mechanic so getting people to make that investment is something that makes sense to split from the investment of money.
2. Check out the history of Chivas Regal. They were a low shelf blended scotch and at some point decided to simply double their price. No label change, nothing different. Their sales actually went up (and their profit was much higher per bottle obviously).
I know the arguments behind it, and I know the research shows it brings slightly higher conversions, but I always see charm pricing (e.g. 2.99 vs 3.00) as an insult of sorts to my intelligence.
It's like the company is saying they don't think their customers have the mental capacity to reason that 9.99 is effectively the same price as 10.
It's even doubly meaningless for some products in Australia, because with 5c being the lowest coin in circulation purchasing a single item worth $9.99 will be rounded up to $10 anyway.
Can't find it any more, but in some literature I was reading that actually it's not so clear as people think. In some situations -1ct works well in other situations other prices like $10.23 work much better. I don't remember the requirements, though.
In any case it's bad to do something that you haven't tested yourself for your market. So if you sell something try different prices and pricing strategies and see what works best.
What certainly works (but is illegal in many countries, I think) is price tagging something as $20 and offering 50% off.
Well, tfa says it pretty clearly: When asked to estimate the real value of TVs priced at 4998.10, 5000 and 5012.56 the middle one was estimated nearly $500 lower than the others.
Same goes for houses and other situations where you're likely to be conceived as having rounded up.
This says it clearly for TV prices in the interviewed customer segment and house prices in the interviewed customer segment. It doesn't mean that other situations don't do better with other prices.
Working in IT I have some friends who would pay $503.71 more for the same TV (compared to the price tag on the TV itself) if you present them a complex mathematical formula for why that TV should be valued higher, but who wouldn't pay $503.70 more because that's not what the formula says.
What I want to say by that example: Different people have different thinking and test your customers, don't assume you are right because a book says so.
Interesting. Wasn't something like this in the undercover used car salesman article a while back as well? You should always come up with weird numbers so the customer doesn't think you made it up.
Isn't it more like: "Turns out we can manipulate people"
Many of the arguments in the article are about subconscious behaviour. After reading all of these tricks for pricing tactics, I'd say it's rather like a magician fooling people with misdirection. No one likes to be fooled but you only notice it when it's too late.
>It's even doubly meaningless for some products in Australia, because with 5c being the lowest coin in circulation purchasing a single item worth $9.99 will be rounded up to $10 anyway.
While I agree these charm pricings are an insult to our intelligence, they do have a slightly positive effect on the final price. Firstly, the rounding to the nearest 5c only occurs on the sum of items bought (e.g. if you buy 3 items at $9.99 the rounded total is $29.95). Secondly, the rounding only applies to cash payments as electronic payments are not rounded.
"If you decide to downsize your product, you should reduce the size of all three dimensions — height, width, and length — by an equal amount. Consumers are less likely to notice a change in all three dimensions."
"People will perceive your price to be smaller if you display that price in a smaller font."
"People will perceive your price to be lower if it contains fewer syllables."
When I shop, I look at the quantity. If my chocolate has 90 grams instead of 100, I'll immediately put it back and look for another brand - because it's too hard for me to compute how much I would get swindled of.
I'm not sure whether it's a practice everywhere, but many shops in NYC also indicate the "per unit" price: per gallon/oz/pieces etc. I use it all the time, and I wish it was mandatory by law for all products.
I live in Spain and when you go to supermarkets, all price tags on the shelves and on supermarket-packaged products (pre-cut meat, for example) indicate the cost of the product per unit (big numbers), and per kilo (small numbers), so, if you want, you can quickly and easily compare a package of, say, 136g of meat with other packages, without weight interfering. You can see easily the price per kilo of each kind of meat.
a colleague sent me a link to "tactic 14." after reviewing the citations i'm skeptical of the rest. it appears the author misunderstood some research he cited.
the thesis for tactic 14 is:
> Anchoring not only works for prices, but it also works for any number, regardless whether that number is a price.
the SSN example demonstrates anchoring based on price (not just any high number). the results of this experiment would have likely been the same if people rolled dice.
and the only experiment [#4 of 5 from adaval and monroe (2002)] which tested "irrelevant dimensions (e.g. weights)" vis-à-vis priming with high/low prices seems to have been poorly designed and produced questionable results. (it should be noted the test methods were also sufficiently different than the UX illustrated in the article.)
so based on the cited research there's nothing that leads me to believe it makes sense to use "any high number" to make my price look better. and if i've understood adaval and monroe (2002) correctly this could actually create a contrast effect (and diminish value perception).
additionally, the suggestion in the article -- "join 2,387 happy customers" -- could also be an example of social proof, making it difficult to ascertain the cause of behavioral change. (was it any high number, or all those happy customers? or both?)
and lastly/separately, why base pricing strategy on the offline behaviors of a bunch of starving students looking for extra credit? it's 2015. where's the shopping cart data?
I skimmed through the article and saw some misunderstandings of research too.
Also another thing that made me raise an eyebrow - "You should avoid charging different prices based on past behavior, demographics, or any other factor besides natural supply and demand."
Clicked the "X" shortly after.
People are eating out of his hand it seems though, so he has that going on for him at least. Many ebooks will be sold.
i'm disappointed this made it to the top of the front page yesterday with such little criticism. it feels like something written by an intelligent yet naïve marketing undergrad, not something on which to base pricing strategy.
> "You should avoid charging different prices based on past behavior, demographics, or any other factor besides natural supply and demand."
wow. what an incredibly stupid recommendation. based on this, every major grocery chain should terminate their loyalty card program, many luxury products/services should never see the light of day, and perhaps the rest of us should just start businesses in perfectly competitive markets with little hope of long-term economic profit.
an aside: searching for "you should avoid" led me to another incredibly naïve recommendation: "you shouldn’t A/B test your prices."
there are so many ways to test pricing it seems irresponsible to dissuade people from doing this, especially in the context of a gigantic list of pricing strategies.
here are just a few types of tests that come to mind:
* practically any promotional/limited offer pricing scheme could (and probably should) be tested.
* prices of users' shopping cart items saved for later (e.g. amazon)
- can a/b test same items in different users' shopping carts and measure purchase patterns
* grocery coupons for people in different zip codes
- most people don't get coupons for other neighborhoods so they won't see two prices
Thanks for the feedback. I don't mind the criticisms - I think a healthy debate is good. :) However, I just want to clarify something.
> "You should avoid charging different prices based on past behavior, demographics, or any other factor besides natural supply and demand."
That quote is taken out of context. Obviously you should determine prices based on those factors. That quote was referring to dynamic pricing and how you shouldn't charge different customers different prices based on those factors (e.g., I'll charge $200 to Customer X, but I'll charge $150 to Customer Y because I don't think he'd buy at $200).
I also find this list fascinating, but I also agree with other commenters that energy is better spent educating people on these psychological pitfalls than to try to take advantage of them. If "do no evil" was ever a motto that a company believed in this would certainly qualify.
Agreed, but I wonder how many companies would educate people on these tricks? Would it give any company any advantage or benefit? And what if a large charity uses these tricks in their fund raising schemes? In the UK for example we have a number of street "chuggers" (charity muggers) who use a range of bad tactics on the street and doorstep. Legally they are allowed to get away with their bad behaviour because so many people support the charity in normal ways.
Nice write-up, but it hurts to see a generalized warning against dynamic pricing since that's what we provide with our product Baabedo[0] (we don't adjust prices on a per-customer level though).
One point where we observed the exact opposite effect is the splitting of shipping cost. Products with free shipping sell better than with split pricing, but that might be a marketplace specific effect of Amazon.
I'd say the dynamic pricing should be done transparently, that is, the user should know that the products have been re-priced.
It also depends very much on how you do it.
From the article:
>Is dynamic pricing always bad? Not necessarily. Dynamic pricing can be effective when adjustments are based on supply and demand (e.g., stadiums trying to fill remaining seats).
>Dynamic pricing becomes harmful when adjustments are based on a customer’s willingness to pay. You should avoid charging different prices based on past behavior, demographics, or any other factor besides natural supply and demand.
Does badbedo do re-pricing based on the positive or negative factors? From the blog it appears to be choosing optimal prices based on the market, not on the user.
> Does badbedo do re-pricing based on the positive or negative factors? From the blog it appears to be choosing optimal prices based on the market, not on the user.
Because we are currently only on Amazon we can't do optimization based on the customer because we only get information about him when a sale is made. So at the moment we optimize based on market and past sales history.
I am generally open to the idea of customer based pricing and think that it can be very powerful if done right and be benifical to both the merchant and the customer.
That's an interesting read. But it just feels wrong. And the worst is that it works. Could we all just stop buying things based on some annoying pricing tactics and start buying products of companies that we want to support for their good work?
Are you are suggesting consuming based on how the company makes you feel? If they make you feel good, or convince you that they are doing good work then should they get more business? Is that rational? What if they are lying or exaggerating? Doesn't a lot of marketing and advertising seek to increase the positiveness of the particular brand?
How about morality? How about if a charity uses these tactics in their fund raising? Is a charity raising funds to protect children allowed to use somewhat manipulative psychological pricing strategies to do good?
My point is that it does feel wrong and it does work, but it works because of feelings and psychology. It's fascinating morally. I agree, it's very hard to stop this and start consuming based on rationality.
> If they make you feel good, or convince you that they are doing good work then should they get more business? Is that rational? What if they are lying or exaggerating?
It is certainly more rational to believe a company should get more business if they can convince you that they're doing good work or better work than others. In order to protect customers from lying or exaggerations, there are (ideally) laws in place that should give you the right to return whatever it is that you bought. The whole discussion loses its purpose if we include fraud or otherwise illegal tactics.
> How about morality? How about if a charity uses these tactics in their fund raising?
That's a tricky question. I think you could reframe it into: Is it morally justifiable to do bad things in order to ultimately achieve a greater good? Morally, the answer is no since the morality of a society are a set of rules that determine what can be done and can't be done (instances of these rules can be found in the law). So, if something can't be done morally (e.g. child labour), you simply can't do it, whatever the context might be.
On the other hand, pricing tactics are not illegal and there's probably not enough consensus to derive morality from it. However, ethically, from my personal viewpoint, I would argue that it amounts to manipulation and therefore deceits people. Thus, I can not endorse charities that use manipulation, no matter the context. After all, if you want to do something good for people, why would you try to manipulate the ones trying to help you?
I like the idea about the charity. In the UK we have what's been called "chuggers" - charity muggers. Essentially doorstep and high street salesmen, these professionals work on behalf of many well loved charities but they harass and aggressively try to make sales.
Councils and people have tried making laws to enforce against them but they do not succeed because the charities are well liked by the majority of the population. So on the whole the charities do lots of good and people cannot effectively complain against their fund raising tactics. The charities do not suffer as a result of this, and those who support the charities do not suffer.
Is any kind of physical cold calling, being stopped in the street, or at your doorstep manipulative? Does it seek to control people into donating to the charity? Yes. Is it objectively wrong? I don't know, but personally I think it is subjectively. Do some people dislike the behaviour? Yes. Do some people like the charity nonetheless? yes.
Then why are you asking for my email? I hate it when they ask for your email for something can be provided without. And the worst part is that it doesn't tell you if it's going to send more emails.
This is very high value, well presented, and I went to read some of your other work.
I appreciate your source cites. Well done.
Gold Star for never discounting without a reason. The amount of lost value due to this is insane. It is a constant issue I face. Most of the time, I find improving value perception to be the better, more sustainable answer to sluggish sales.
That's a good question. I think it depends on the pricing technique.
For example, research shows that anchoring effects occur even when people are explicitly warned about anchoring before they make their estimate.
On the other hand, if people are aware of other pricing techniques, then those techniques could backfire in a negative direction. For example, researchers have found a "reverse priming effect." In other words, if people detect some persuasive intent behind the marketer, then those people are more likely to resist that persuasion attempt. So it could make things worse.
So it might have negative effect on some people (who feel it as insult on their intelligence), but if twice as many people don't notice it and are positively affected, it's still a net gain.
I think, things like this work even if you know the underlying mechanism in the pricing strategy.
As an alternate example, my realtor regularly keeps in touch with me by email, using new letters, calling me on my birthday, etc and even though I know that he is being nice mostly because he wants referrals from me (definitely not because of my winning personality :-)), I still feel obligated to give him referrals.
Likewise, knowledge of the pricing strategy doesn't necessarily make you immune to it's effects unless you consciously stop and think and able to compare with other similar services/products and are prepared to take your time.
Our software could do this for us automatically, e.g., a browser plug-in that detects prices in a web page, and automatically rounds them back up, then adds taxes and shipping.
Some of these tactics are contradictory. Tactic 2 says to add cents for a "rational" purchase, i.e. 19.86 verus 19. This contradicts tactic 3 which says to use fewer syllables, i.e. 19 versus 19.86. How do you prioritize these tactics?
I think you need to look at it holistically and place yourself in the shoes of your audience. I imagine an example of a "rational" purchase to be something like searching for a good (eg. headphones) on Amazon and comparing prices between brands. A buyer will try to research and justify their purchase, and the cents may help push them over the edge and convince them that they've "done their homework". Meanwhile, an example of the syllable tactic is a restaurant menu. When I order from a menu, I'm time constrained and forced into a decision. I ask myself internally: "Would I get 19 dollar value from this meal?", and if the answer is "Yes", I'll order it.
Great list. Minor quibble-- I wish Nick had put the last part first. Pricing effectively is all about successfully communicating value. Your price is limited by your perceived differential value. Once you get that part, the rest of the techniques might be helpful to optimize, but if you start with the optimizations without really understanding how your value compares to your customers' alternatives, you may get a local maximum, but you won't get the best results.
I am biased (or the audience of this post, who knows) because neither I am nor a salesman or startups funder (not even american), but I really feels bad about the vendor when noticed that psychological tricks are used to convince me to buy something. Yes, I know that's how world works, but I am more keen to buy when I am sure and here is a room for valuable marketing.
It's a really good list, can work as starting point. I advice you to be careful before applying in your business.
Pricing is too important for we stop just with this list. The list just show topics, results and conclusions of experiments, not the psychology experiments made and the reason behind each strategy, so, very easy to be forgotten, or even applied wrongly.
If you want to lean in a way you will never forget, and apply it correctly, you need to read some books about pricing.
Excellent writeup. Just one point that I disagree on is the Shipping price split up. I tend to look for other sellers when I see one with paid shipping, and tend to favour free shipping sellers more.
- If you are combining a discount with the product, and mention a shipping charge, the discount registers as a false discount ('He is just shifting costs through shipping split')
- Maybe it registers as more of mental processing (X$ + Y$ needs more time to process than Z$)
- Sometimes, it is human nature to be greedy, and look for best value possible. Seeing a shipping cost feels like seller is not 100% working for the customer, or we just become plain greedy, that why should someone else get a part of the price.
Didn't read the article yet, but I tried this: 2 euro vs. 1.99 euro and it made 0 difference in conversion rates. I guess it depends on the product. Maybe I should repeat the test, too.
That's impressive. Unfortunately the source* of theses number is not very clear on how many items has been sold and if people buy by unit or by pack. I would love to see more tests result like that.
This list is great. I think it's funny how many of these folks criticize the list. They are probably the same ones who chose to OS their products for them to go off and die on Git rather than actually sell it.
I would add that doing bundles after payment is huge. Once a customer has paid, adding 15% to their bill is a very easy sale. Whether a soft good or in the form of version insurance/download protection. Conceivably, it could be done on Saas too, perhaps to lock in the rate for life with a one-time charge.
This set off each and every one of my slimy scumbag and no-good charlatan alarms. Please convince me that this protection is a good that's actually not extortionate and/or doesn't rely on the ignorance of one's customers.
A short heads-up: The page seems to freeze IE11 on Win 8.1 - tried 2 times (once with lots of tabs, one in only one tab/instance). IE has to be restarted both times.
This kind of stuff is a strong argument for communism. An immense amount of mental energy was squandered on creating this list (not Nick Kolenda, the people who studied these things). That same energy could have been spent on curing any number of psychological ailments, but instead it was spent on getting people to pay more for less. It's interesting, but as far as I'm concerned the energy was wasted.
If you want to organize resources and work in a centralized fashion, you still have to understand people, how they think and how they're motivated. You still have to decide who gets what, how they get it, who works on what, how to motivate them... You just do it directly instead of detouring through the abstraction of money and prices.
Getting rid of the intermediate pricing abstraction doesn't magically solve the underlying problems. Perhaps it lets you address those problems more directly, but it's not clear that that's a better approach—as programmers and computer sciences we should appreciate the power of abstraction and indirection to make solving problems easier.
The same energy would just have been spent in committees. Except, historically, the process would have ended up far more political than scientific.
I'll preface what I'm about to say with this: I don't think any economic system works at the scale on which modern economies operate.
> If you want to organize resources and work in a centralized fashion, you still have to understand people, how they think and how they're motivated. You still have to decide who gets what, how they get it, who works on what, how to motivate them... You just do it directly instead of detouring through the abstraction of money and prices.
Understanding people, how they think, and how they're motivated, at the scale of modern economies, is necessarily an approximation. We can group people by demographic, and that gives us better coverage, but ultimately some percentage of people will never be adequately accounted for in our models.
At a smaller scale, say, 50 people, it becomes possible for each of the members of the economy to understand each others needs and motivations. It becomes possible for each person to state their needs and motivations and for others to judge whether these needs are adequately being met or if the person is being too greedy.
> Getting rid of the intermediate pricing abstraction doesn't magically solve the underlying problems.
At small scale, the pricing abstraction offers very little.
> Perhaps it lets you address those problems more directly, but it's not clear that that's a better approach—as programmers and computer sciences we should appreciate the power of abstraction and indirection to make solving problems easier.
I think this is a mistaken characterization of abstraction in computer science. Abstraction doesn't take you further from the problem you're trying to solve, it takes you closer to it. Indirection and abstraction are there to pull you away from the problems you aren't trying to solve: memory management, bit twiddling, etc.
When you talk about addressing problems more directly, that to me means you have created a better abstraction, not that you've removed an abstraction. If a layer of abstraction means you're dealing less directly with the problem, then it was the wrong abstraction and should be replaced or removed. YAGNI.
> The same energy would just have been spent in committees. Except, historically, the process would have ended up far more political than scientific.
I think this is also a function of the size of modern economies.
Size is a difficult problem to solve in itself, but situationally it has been solved with great success: see kibbutzes and communes.
Translated: "That same energy could have been spent on something I care about and not what the author cares about."
Which kind of explains why communism always leads to tyranny: it is based on the essentially tyrannical assumption that communists alone have the knowledge to decide how everyone else should be spending thier time, and the moral authority to impose that judgement on others by force.
The problem is that no one has ever found a way of granting such power of central decision making in such a way that either:
a) central planners have access to the information and computing power distributed in the market such that they can make well-judged decisions even by the theoretically pure criteria they claim to be using (that is, communism is inefficient)
b) central planners have no incentive not to make decisions that serve their own interests to the detriment of all others (that is, communism is corrupt).
Given that theses results are incredibly robust and widely known, it is weird that anyone would suggest communism is an interesting alternative to anything. Reform capitalism all you want, but why bring up the political equivalent of a perpetual motion machine as the solution?
Just curious, but does communism by necessity or definition require central planning? Is a distributed power model not possible, aside from the fact that it hasn't been done?
I'm genuinely curious, as I've never studied political systems in any depth.
I see where you're coming from. Hopefully I can defend some some of the researchers in that field :)
Although most applications of that research are often applied to business and marketing, the underlying research has broader implications. For example, some of those studies progressed our understanding of cognition in general (the researchers merely used numbers in their study - which was then adapted for pricing applications). But we now understand more about our brain's processes than we did before.
And ,with a better understanding of cognition, we can help pave the way toward other, more benevolent applications (e.g., curing psychological ailments).
So even though that research may not produce benevolent results directly, I would argue that it helps indirectly (among many other benefits too).
And, with a better understanding of cognition, we can help pave the way toward other, more benevolent applications (e.g., curing psychological ailments).
An example of progress would be equally intensive research on how consumers could tip the balance back in our favor. I'm skeptical that it will happen for a couple of reasons:
1. Research supporting business and marketing probably has infinitely more generous funding. We are hopelessly out-gunned.
2. Research into how people respond to abuse (manipulation) doesn't necessarily translate into benevolent applications.
I agree with you, but the benefits of this kind of research have to be balanced with the harms as well. Rampant consumerism has driven environmental and economic damage, and while it's hard to weigh these kinds of benefits against these kinds of harms, my intuition is that if even a small percentage of the consumerism was caused by this research, the upsides far outweigh the downsides. Of course that's entirely subjective.
While I wouldn't go so far as agree that it's a strong argument for communism (though I do support guaranteed, unconditional basic income myself), I do think it's a strong argument for broad public education initiatives...imagine if there was a required class called "How To Tell When a Politician/Company is Bullshitting You." Seriously, no sarcasm, I think that's an extraordinarily useful skill that is often glossed over, and if someone made a curriculum for understanding psychological biases, that could prove insanely useful at mitigating the effectiveness of these ploys.
I would pay £19.99 for that course on "How To Tell When a Politician/Company is Bullshitting You.".
Seriously though - it's a great idea, and ironically it could also be a potential profitable start up idea. A start up that takes the side of consumers and decloaks: a) tricky pricing methods b) political bullshit c) herd mentality. Via a range of plugins and services - a bit like an on demand Snopes.
The strongest argument against communism is "How could a man plan economy if he can't plan his todo list app without producing bugs?"
People tend to fail in simple predictions but communists and hardcore socialists still believe they can manually control systems of enormous complexity.
Number of failed communist regimes confirms they can't. (But these regimes are quite good at producing death and starvation.)
Communism is a philosophy in the first place. It leads to a certain type of governing and managing economy. On practice it also lead into a totalitarian state every single time.
Communes and kibbutzes are almost universally democratically governed, so no it does not lead to a totalitarian state every time. That's a ridiculous claim.
Why do kibbutzim belong to communism? Communism is based on works of Karl Marx and Friedrich Engels. You can't just attach a "communist" tag on every commune that has shared property.
You seem to be confusing communism, a form of economy, with totalitarianism, a form of governance. There are many non-totalitarian examples of communism.
There is no reason at all why this information has to result in more for less.
Couple better value perception (and that's a great topic for Nick to do sometime in the future) with effective pricing and now you have them paying more for more and happy about it.
Even better, adding more value means adding more to the price too. Built in race to the bottom avoidance, while actually delivering great value, not diluted value at higher prices.
How this plays out depends on who is pricing and what their overall strategy is.
Just because I said there exists a strong argument for communism doesn't mean I am not aware of other arguments for other alternatives which might be stronger.
I was hoping to see a list based on how the price of a product better communicates a product's value. It's sad to me that there is so much energy focused on the manipulation of the buyer as the goal.
Frankly, I consider the aggressive use of these tactics to be a sign that the company is manipulative and untrustworthy.
1. Kahneman and Ariely both mention in their writings how their is a difference between free and even just 1 penny. Especially given how popular Freemium is I thought this might be a good topic for you to cover. (Ariely calls this "market norms" vs "social norms")
2. A common theme here on HN (mainly from people like Patio) is to "charge higher". I couldn't find a place that explicitly mentioned this. Many people perceive higher quality and value of products and services simply because they cost more (regardless of whether they actually do or not)
3. No mention of Kahneman?!?
This is awesome regardless...saving it and sharing it with some clients.
Feel free to contact me directly if you want me to elaborate on any of this.