3 ms·

If you can not differentiate your product at all, it will be pretty hard to price non-linearly. You are right there: Customers will run away. It has happened o

by zwaps 9y ago

If you can not differentiate your product at all, it will be pretty hard to price non-linearly. You are right there: Customers will run away. It has happened on the internet (targeting), but people get very angry. Uber extracts from two sides of the platform, which is a more complicated matter. For the other examples, you are not exactly correct. Health insurance is clearly a two-part price, indeed it is a classical example. You pick your tariff and coverage. Whenever you pick, it's a nonlinear pricing scheme. Salaries are another thing. A base and a bonus are very, very common. That's two part. Also comissions and somesuch. Hotel rooms? Rate and extras? Minibar? Cars? Extras and bundling... All of that is non-linear pricing. So your point is basically correct. If the customer notices you get to extract all surplus, he might get annoyed. But a two part tariff most often seems beneficial to the consumer, because he gets to choose!

paulsutter 9y ago
Uber’s complexity and lack of transparency around paying drivers is a really bad idea for the company, and only an MBA could think it’s good. Someday Uber will discover that they were just a missing feature in Google Maps, and this is why all the drivers will switch. Your examples have convinced me that a 3 part tariff is generally bad and an indication of extractive desperation on the part of the vendor, and fundamentally based on deception. If you wonder why people hate their banks, pricing is the reason.
- zwaps 9y ago
  The pricing scheme is derived as an optimality solution for the company. Obviously, it always (mathematically speaking, weakly) benefits the firm. An all-knowing monopolist will extract 100-epsilon% consumer surplus. That's why economists are big about regulation and taxation. Is it deception? Not necessarily. A pricing scheme is a menu of deals from which you pick the most beneficial for you. You then optimize as well - just that the company gets to set up the menu. If a company were to decide to leave a minimum surplus to you, the optimization problem would be almost identical. We just assume that companies set this minimum value small. Perhaps, however, this may not be the best long term strategy? Not all companies may try to extract everything. But what is a good and fair amount?
  - paulsutter 9y ago
    Business is about more than extracting the most you can. Most Wells Fargo customers are unhappy, but they stay because the competitors are just as bad. Unhappy customers is bad for business. Startups are often about giving customers a better experience than the incumbents. Toms writing is targeted to entrepreneurs who need to consider all aspects of the business, unlike product managers at Faceless BigCo, who “win” solely by the shitty little optimizations that you so love.
    - zwaps 9y ago
      Ignoring your needlessly hostile tone, it is Tom who proposes nonlinear pricing schemes. If there were no interest in controlling the surplus left for consumers, neither him nor anyone else would have need to employ such schemes. Entrepreneurs probably employ such pricing methods more often than BigCo as well, simply because they sell less uniform products to less anonymous customers. Your point?
      - paulsutter 9y ago
        Yes my point is that the term "optimal" could be misleading. People reading the document are likely to interpret "optimal" to mean that it's the best business decision. But in many cases it's not the best decision. So my question (not point) is, when/for what is a 3 part tariff the best choice, when you consider all the other issues (like, what would annoy customers, etc). In the case of Salesforce, the day I signed up, if there were a significant base cost over-and-above the per-seat price, there's a good chance I would have avoided Salesforce, and Salesforce would never have gotten that (eventual) $300K a year from me. So it seems that their pricing is better than a 3 part tariff.
        zwaps 9y ago
        First, your point about optimality is "not even wrong". Like I said, the surplus value you want to give your customers is really up to you and the optimality starts from this point. Yes, people may misunderstand this in the article. But I think that's the reason why the article does not offer any concrete recommendations. Non-optimality just means you are leaving surplus on the table, out of your control. It certainly doesn't mean you will be better off in the long run. If you are comfortable with that decision, go ahead and choose a flat tariff. Since a business is also driven by cost, I would bet you will consider that decision sooner or later. Non-linear pricing is "the best" when you are doing a theoretical argument, since you just can't do worse with more instruments than less - if you employ them wisely. That's what is written in the article, and that's what I elaborated on. It may just mean it gives you most control in deciding about consumer surplus, whatever that strategic decision may be. Second, with regards to salesforce: A x-part price does allow you to set fixed components to zero, if that corresponds to the customer base. The "best" mentioned in the article is a theoretical construct in the sense that nonlinear pricing is simply more general. You are not the only customer, and in the end it depends on the whole demand base as to what is optimal. A product delivering high value per seat may just simply be best off with a low-to-zero fixed component. Again, it all depends on who you are targeting. You spend 300k a year? So what if I offer you a contract with 200k a year plus a fixed 50k? Will you not take it? What if you do, and I tell you we both win in this deal? My point is that there is a general thing to understand about pricing, and there is a reason why this article refers to something as best, while being sparse on actual implementation. The best business decision, as you put it, depends entirely on your assessment of what customers want. This much will never change. But this article wants to point out that you can deal with information asymmetries with a nonlinear pricing scheme, something everyone should study deeply before pricing one's product. As a consumer, you are faced with price menus in almost every purchasing decision you take outside a supermarket. That's for a reason. If you are an entrepreneur, and you do not understand why or when one may employ a nonlinear pricing scheme, then you are lacking knowledge.

by zwaps 9y ago

paulsutter 9y ago

Uber’s complexity and lack of transparency around paying drivers is a really bad idea for the company, and only an MBA could think it’s good. Someday Uber will discover that they were just a missing feature in Google Maps, and this is why all the drivers will switch. Your examples have convinced me that a 3 part tariff is generally bad and an indication of extractive desperation on the part of the vendor, and fundamentally based on deception. If you wonder why people hate their banks, pricing is the reason.

zwaps 9y ago
The pricing scheme is derived as an optimality solution for the company. Obviously, it always (mathematically speaking, weakly) benefits the firm. An all-knowing monopolist will extract 100-epsilon% consumer surplus. That's why economists are big about regulation and taxation. Is it deception? Not necessarily. A pricing scheme is a menu of deals from which you pick the most beneficial for you. You then optimize as well - just that the company gets to set up the menu. If a company were to decide to leave a minimum surplus to you, the optimization problem would be almost identical. We just assume that companies set this minimum value small. Perhaps, however, this may not be the best long term strategy? Not all companies may try to extract everything. But what is a good and fair amount?
- paulsutter 9y ago
  Business is about more than extracting the most you can. Most Wells Fargo customers are unhappy, but they stay because the competitors are just as bad. Unhappy customers is bad for business. Startups are often about giving customers a better experience than the incumbents. Toms writing is targeted to entrepreneurs who need to consider all aspects of the business, unlike product managers at Faceless BigCo, who “win” solely by the shitty little optimizations that you so love.
  - zwaps 9y ago
    Ignoring your needlessly hostile tone, it is Tom who proposes nonlinear pricing schemes. If there were no interest in controlling the surplus left for consumers, neither him nor anyone else would have need to employ such schemes. Entrepreneurs probably employ such pricing methods more often than BigCo as well, simply because they sell less uniform products to less anonymous customers. Your point?
    - paulsutter 9y ago
      Yes my point is that the term "optimal" could be misleading. People reading the document are likely to interpret "optimal" to mean that it's the best business decision. But in many cases it's not the best decision. So my question (not point) is, when/for what is a 3 part tariff the best choice, when you consider all the other issues (like, what would annoy customers, etc). In the case of Salesforce, the day I signed up, if there were a significant base cost over-and-above the per-seat price, there's a good chance I would have avoided Salesforce, and Salesforce would never have gotten that (eventual) $300K a year from me. So it seems that their pricing is better than a 3 part tariff.
      - zwaps 9y ago
        First, your point about optimality is "not even wrong". Like I said, the surplus value you want to give your customers is really up to you and the optimality starts from this point. Yes, people may misunderstand this in the article. But I think that's the reason why the article does not offer any concrete recommendations. Non-optimality just means you are leaving surplus on the table, out of your control. It certainly doesn't mean you will be better off in the long run. If you are comfortable with that decision, go ahead and choose a flat tariff. Since a business is also driven by cost, I would bet you will consider that decision sooner or later. Non-linear pricing is "the best" when you are doing a theoretical argument, since you just can't do worse with more instruments than less - if you employ them wisely. That's what is written in the article, and that's what I elaborated on. It may just mean it gives you most control in deciding about consumer surplus, whatever that strategic decision may be. Second, with regards to salesforce: A x-part price does allow you to set fixed components to zero, if that corresponds to the customer base. The "best" mentioned in the article is a theoretical construct in the sense that nonlinear pricing is simply more general. You are not the only customer, and in the end it depends on the whole demand base as to what is optimal. A product delivering high value per seat may just simply be best off with a low-to-zero fixed component. Again, it all depends on who you are targeting. You spend 300k a year? So what if I offer you a contract with 200k a year plus a fixed 50k? Will you not take it? What if you do, and I tell you we both win in this deal? My point is that there is a general thing to understand about pricing, and there is a reason why this article refers to something as best, while being sparse on actual implementation. The best business decision, as you put it, depends entirely on your assessment of what customers want. This much will never change. But this article wants to point out that you can deal with information asymmetries with a nonlinear pricing scheme, something everyone should study deeply before pricing one's product. As a consumer, you are faced with price menus in almost every purchasing decision you take outside a supermarket. That's for a reason. If you are an entrepreneur, and you do not understand why or when one may employ a nonlinear pricing scheme, then you are lacking knowledge.