Showing posts with label aggegator. Show all posts
Showing posts with label aggegator. Show all posts

Tuesday, November 13, 2018

"The Case Against Micropayments" versus "Subscription Hell" -- Finding Flexibility

This was initially published as"The Case Against Micropayments" -- From Fear and Surprise to The Comfy Chair
Both subscriptions and micropayments, as currently applied, are far too inflexible to satisfy more than a small fraction of potential paying customers. What is needed is a more flexible strategy that blends elements of both in a way that minimizes risk to the customer -- whether they access, and enjoy more or less than they expect in any given period. This more descriptive title reflects that core message. [2/1/19]

...and, see the update at end on how the beautiful theory of NFT micropayments is being murdered by a brutal gang of customers. [1/18/22]

Part 1: Micropayments  
(Followed by Part 2: Subscriptions and a unifying perspective)

Micropayment hope springs eternal. Clay Shirky and Andrew Odlyzko drove a stake in its heart way back in the dot-com era, but here it is again -- with new, more frictionless payment solutions and new content aggregators, some counting on the magic of blockchain and cryptocurrencies. Some micropayment content services have gained limited traction, primarily in Europe. But as Shirky said, "their weakness is systemic." Decades later, these systemic problems remain unsolved.

But that is true of micropayments as currently conceived: small payments at pre-defined rates. When the rates at which micropayments are charged become more reflective of the dynamics and behavioral economics of actual value -- as received and perceived by the customer -- that systemic problem can be solved. How can that be?

The problem as we now conceive it

Shirky summarizes the systemic problem:
The Short Answer for Why Micropayments Fail
Users hate them.
The Long Answer for Why Micropayments Fail
Why does it matter that users hate micropayments? Because users are the ones with the money, and micropayments do not take user preferences into account.
In particular, users want predictable and simple pricing. Micropayments, meanwhile, waste the users' mental effort in order to conserve cheap resources, by creating many tiny, unpredictable transactions. Micropayments thus create in the mind of the user both anxiety and confusion, characteristics that users have not heretofore been known to actively seek out
Odlyzko pinpoints the behavioral problem, drawing on the century old history of micropayments, and quoting Kara Swisher:
What was the biggest complaint of AOL users? ...Their overwhelming gripe: the ticking clock. Users didn’t want to pay by the hour anymore. ... Case had heard from one AOL member who insisted that she was being cheated by AOL’s hourly rate pricing. When he checked her average monthly usage, he found that she would be paying AOL more under the flat-rate price of $19.95. When Case informed the user of that fact, her reaction was immediate. ‘I don’t care, I am being cheated by you.’
Odlyzko's conclusion: "The lesson of behavioral economics is thus that small payments are to be avoided, since consumers are likely to pay more for flat-rate plans/"

Now it is not so much the "ticking clock," as "the ticking meter," but the problem remains. Much like Monty Python's Spanish Inquisition: "surprise and fear." Fear that we may be surprised to have run up a large bill without realizing it. It may be only a little regrettable, or it may be very seriously regrettable.  We will be stuck with that (or try to plead with the Inquisition's customer service department for forgiveness).

Even if you make the micropayment process totally frictionless, surprise and fear remain.

The pricing theory of relativity -- removing surprise and fear

We think of micropayments as immutable quanta of price. So many cents or micro-tokens for so many units of service. But why are we stuck with such Newtonian pricing, when Einstein showed us that clocks and meters can expand or contract relativistically?

We forget that prices need not be pre-set, but can be dynamic, and that they should adapt to whatever the customer and the business agree is fair. Prices should be relative to value, as I said above: reflective of the dynamics and behavioral economics of actual value -- as received and perceived by the customer.

The most systemic solution to the problem with micropayments is to apply post-pricing, in the form of post-bundling. We are talking about micropayments for digital "experience goods," which are unlike traditional "goods:"
  • They have little marginal cost.
  • Their value is not really known until after the experience. 
  • They are typically bundled such that the mix of items and amount to be metered is not known until the entire bundle is chosen by the customer, on demand, during the course of a billing period.
Why should such services be metered and priced at a pre-set rate? That is antiquated thinking:
  • The vendor can afford to take on most of the pricing risk (since the marginal cost of service is negligible).
  • The unit price should be discounted to provide quantity discounts (and to adjust for items sampled but not finished).
  • Price caps can be applied to ensure reduce surprise and fear. Such a cap might be higher than the corresponding rate for a simple flat-rate subscription, to compensate for the expectation that the customer will often pay less than the cap or even the usual flat-rate, but still low enough to eliminate the customer's fear.
I have written about simple forms of doing this:
The need for a relationship perspective

The core idea is that micropayments are most relevant to recurring business relationships, whether with a single content/service provider, or with an aggregator of such content/services. In either case we need to look beyond individual transactions to the aggregate value transfer over a period, in the context of the ongoing relationship. Subscription businesses already recognize that Customer Lifetime Value (CLV) is their primary success factor.

Advanced forms of FairPay take this farther, eliminating fear by allowing the customer to pay no more than they think fair for any given billing period -- as long as they do not abuse that privilege. That is just another level at which to leverage the "free" replication of digital content/services to eliminate the customer's pricing risk.

Whatever degree we take it to, when we shift to this relationship view, we realize that the vendor can absorb most of the short-term pricing risk, as long as the overall relationship is profitable over the lifetime of the customer. They can use the meter as just a guide, applying it to get a price that is adaptive to the nature of the relationship. The business can track each customer's fairness reputation over time, and use that to decide how much pricing power to grant and when. That enables prices to be set in a way that eliminates the customer's fear of nasty surprise. If we can remove that "anxiety and confusion," users' hatred of micropayments will turn to love. Good relationships build and thrive on comfort -- give customers the comfy chair!

Part 2: Subscriptions and a unifying perspective 

"Subscription hell" -- the case against flat-rate subscriptions

Interestingly enough, the critics of micropayments argue that the success of flat-rate subscriptions dooms micropayments, but now it is becoming apparent that the success of subscriptions is self-limiting. So many services are turning to flat-rate subscriptions that consumers are facing what Danny Crichton called "Subscription Hell."
Another week, another paywall. ...I’m an emphatic champion of subscription models, particularly in media. Subscriptions align incentives in a way that advertising can never do, while also avoiding the morass of privacy and ethics that plague ad targeting. ...Incentive alignment is one thing, and my wallet is another. All of these subscriptions are starting to add up. ...Worse, subscriptions aren’t getting any cheaper. ...I’m frustrated with this hell. ...And I’m frustrated that subscription pricing rarely seems to account for other subscriptions I have, even when content libraries are similar.
...For product marketers, the default mentality is to extract a lot of value from the 1% of readers or users that are going to convert to paid. Subscriptions are always positioned as all-or-nothing, with limited metering or tiering, to try to force the conversion. To my mind though, the question is not how to get 1% of readers to pay an exorbitant price, but how to get say 20% of your readers to pay you a cheaper price. It’s not about exclusion, but about participation.
...Subscription hell is real, but that doesn’t mean the business model is flawed. Rather, we need to completely transform our thinking around these models, including the marketing behind them and the features that they offer. We also need to consider consumers and their wallets more holistically, since no one buys a subscription in a vacuum. For too long, paywall playbooks have just been copied rather than innovated upon. It’s time for product leaders to step up and build a better future.
I have made similar points in a number of posts, most pointedly in Beyond the Deadweight Loss of "All You Can Eat" Subscriptions.

Relationships and share of wallet -- a unifying perspective

With a broader relationship view, we see that the apparent dichotomy between flat-rate subscriptions and discrete "pay per view" micropayments is an artifact of our narrow, transaction-level thinking.
  • We think of micropayment transactions as isolated quanta that add up in ways that cause "anxiety and confusion" because we do not think about how the metered units map to actual value.
  • We think of flat-rate subscriptions from the isolated perspective of a single provider, because our vendors do not think about the whole customer, and what other subscriptions make competing demands for "share of wallet." 
But if we look past those blinders, we see that we exchange variable levels of value, and each should draw a fair share from the consumer's painfully finite wallet. To solve the systemic problems of payments that sustain the creation of digital services -- whether micropayments or subscriptions -- we must take a systemic view of value, share of wallet, and pricing risk. That is why my book has the subtitle "Adaptively Win-Win Customer Relationships."

Most of us can have nearly all of the content and services we want, at a fair and affordable price -- if businesses get smarter about sustainably exploiting the nature of digital services in a cooperative relationship context. Total removal of surprise and fear from pricing is inefficient and impractical, and benefits few. Even with flat rate, we have the converse fear that we will not get our money's worth in any given month. But businesses can leverage digital abundance (that costs them nothing) to put limits on the fear. They can seek to put each of their customers into a comfy chair that is cooperatively and adaptively designed to fit them just right. Failing to do that will be a tragic waste, for businesses and consumers alike.

[Update 2/16/19:] Many advanced examples of better flexibility are on this blog, but one of the simplest is this one: Post-Bundling – Packaging Better TV/Video Value Propositions with 20-20 Hindsight.

[Update 7/2/19:] FairMicroPay -- simple, relationship-value-based adjustments to micropayments. Simplified forms of FairPay might be applied to make micropayments more flexibly value-based. Consider how this might be done in a blockchain-based micropayment system. Relationship-value-based adjustments can be overlaid on micropayment models.  The idea is to add a FairPay layer that identifies the user, and that allows the user to modify a standard base price within limits permitted by a smart contract -- downward as a refund/discount for lack of desired value, or upward as a value-based bonus or sustaining contribution. 

[Update 1/18/22:] Crypto Enthusiasts Meet Their Match: Angry Gamers: "Game publishers are offering NFTs, but skeptical gamers smell a moneymaking scheme and are fighting back...Much of their resentment is rooted in the encroachment of micro transactions in video games." It seems the beautiful theory of NFT micropayments is being murdered by a brutal gang of customers.

------------------------
More about FairPay

For a full introduction to FairPay see the Overview and the sidebar on How FairPay Works (just to the right, if reading this at FairPayZone.com). There is also a guide to More Details (including links to a video). 

My article in the Journal of Revenue and Pricing Management, "A Novel Architecture to Monetize Digital Offerings" also provides an overview of FairPay (summarized more briefly in the ESADE Knowledge article "Three building blocks to monetize a digital business," and previously in Harvard Business Review, "When Selling Digital Content, Let the Customer Set the Price.").

Even better, read my highly praised book: FairPay: Adaptively Win-Win Customer Relationships.


(FairPay is an open architecture, in the public domain. My work on FairPay is pro-bono. I offer free consultation to those interested in applying FairPay, and welcome questions.)

Monday, December 28, 2015

FairPay-What-You-Want for Costly Products? Etsy? Everlane? Tiffany?


[UPDATES 12/27/17: Everlane PWYW sales on again.5/10/17 article broadly covers "What do people want to pay for their clothes?" and mentions FairPay (as "Fair PWYW") - reports 13% pay more than minimum.12/29/16:  Everlane has again done a PWYW sale. It is advertised as "Twice A Year," so it must be successful!
7.12/16 article reports 10% of customers voluntarily pay more than the minimum price.] 
---

Not just digital? Can FairPay work for real products that have significant replication cost? -- such as for a fashion retailer like Everlane or an artisan marketplace like Etsy? There are actually some very interesting opportunities.

FairPay (short for Fair-Pay-What-You-Want) is a new architecture for participative price-setting that adaptively seeks win-win value propositions in ongoing customer relationships. As discussed throughout this blog, the case for FairPay is most obvious for products that have negligible marginal cost to replicate, such as digital content -- since there is no out-of-pocket loss when the occasional customer does not pay fairly-- but a minor variation of the process promises to work well for costly products as well.

The variation is very simple: set a minimum price floor that allows the buyer to set whatever price they want above that minimum. That can ensure that sales are not at a loss, and limit the FairPay adaptation process to apply only to the profit margin that the seller should receive above the cost. This builds on the simpler idea of Pay-What-You-Want (PWYW) with a price floor, which has been common, as described below.

FairPay goes far beyond PWYW to add in seller controls to nudge buyers to price fairly and to exclude those who do not. (It also shifts price setting until after use, when the value of the experience is known.) For a full introduction to FairPay see the Overview and the sidebar on How FairPay Works (just to the right, if reading this at FairPayZone.com). There is also a guide to More Details (including links to a video). [Added 12/30: Also see this background on studies of conventional PWYW.]

A nice example of conventional PWYW with a floor was just provided by fashion e-tailer Everlane.com. They have a 5-day Christmas sale that offers an array of items at any of three different prices [see 12/31 Update below], Using the example described in a news report at Racked a woman's coat said to normally sell for $250 can be had at any of three prices, and a mouseover frames the rationale for those options:

  • $110: "$0 to Everlane. This only covers our cost of production and shipping."
  • $132:  $22 extra to Everlane. This helps to cover production, shipping, and overhead for our 70-person team."
  • $225.  $115 extra for Everlane. This helps cover production shipping, our team, and allows us to invest in growth. Thanks!"
FairPay is a new concept that would enable such sales to become a regular option for selected customers (including those found to pay well on special sales like this Christmas sale).  For customers who develop very good reputations for pricing fairness, many items might be offered that way all the time.  For customers who gain moderately good fairness reputations, such offers could be more limited, but still for many items (often, if not all the time). Thus FairPay becomes an interesting next step for such a retailer, as explained further below. 

FairPay with a price floor -- for retailers and for marketplaces

Beyond the example of a single retailer like Everlane, FairPay can also apply to a platform for serving many sellers, like Etsy.com, a platform that provides a marketplace for many designers and artisans. 

Let's look at the details of how FairPay can apply to either:
  • The conventional offering is for products to be sold at prices pre-set by the seller. All of the issues with conventional pricing apply here -- notably no allowance for individually varying value perceptions, and no post-pricing that enables the price to be set after the value is known.
  • A 100% FairPay offering would allow buyers to set any price they think fair, after receiving and trying the product, even as low a zero. The seller takes the risk that buyers will not be happy or fair, and that they will set very low prices, possibly well below cost. Even with FairPay's reputation tracking and limitation of sales to those who do not maintain a reputation for fair pricing, sellers face the risk of not recovering their costs on some sales.
  • A solution is to add FairPay with a price floor -- similar to PWYW with a price floor (as used by Everlane), but with the added controls of FairPay that I propose.
  • This hybrid version of FairPay could provide for a minimum "floor" price that is pre-set by the seller, plus a profit margin bonus that is set by the buyer. This floor price set by the seller might be paid prior to shipment (as with conventional sales), to ensure coverage of costs. The FairPay portion would address the profit margin bonus price, which would be set by the buyer, after experiencing the product, as with pure FairPay. 
  • Note that just changing conventional PWYW to have sellers set prices after using the product can have significant benefits in getting better PWYW pricing -- buyers no longer need to discount their prices for fear being disappointed by an untried item. Thus sellers should seriously consider this idea of post-priced PWYW, even before moving to the more advanced FairPay process.
How FairPay works with a price floor

Such a hybrid two-level pricing process (conventional seller-set floor price prior to sale, plus FairPay bonus price set by buyer post-sale) could provide a very effective solution to adaptively seek win-win sales.
  • The process would be explained up front, so that buyers and sellers understand that the initial price is just a base price that only covers the cost of the product (and perhaps a very small profit margin), but that buyers who are happy with the product are expected to pay more than that, once they see the value of what they have gotten.
  • The seller could post a suggested bonus price (with profit margin), but buyers could decide to price higher or lower, based on their own judgment of fair value. 
  • Buyer risk is much less than at full price. Refund options could still be provided to deal with serious dissatisfaction over even the "at-cost" base price.  However, with the lower base price, fewer buyers would be so unhappy they would wish to bother with a return for refund. Many would be willing to keep a marginally satisfactory product at a "bargain" price, given that the value is now known, and there is no further effort to doing that.
  • Seller risk is low, because they will at least cover their costs (except for a smaller than usual number of returns for refund).
  • Both benefit by getting more customers to try the product.
  • The bigger benefit is in cooperatively seeking a fair profit margin. Sellers who are happy can decide just how happy they are, considering all relevant factors, now that the value is known -- and can set the bonus prices accordingly. Sometimes this process might lead to a price below a conventional price, sometimes above, but in any case it leads to repeat customers with loyalty.
  • The reason that is important is the "long tail of prices." Some buyers will happily pay more than a conventionally pre-set price, and that generates added revenue. Many buyers (the long tail) will be unwilling to pay a conventionally pre-set price, but would be willing to buy the product at the lower base price, and then consider adding a bonus. Any added bonus is added profit. Thus the seller sells more products and makes more buyers happy.
  • This can work especially well for quality producers who delight their customers and motivate them to pay generously by triggering the use of communal norms. If Everlane's experience is like that of other PWYW vendors, positioning as a dedicated provider of quality and service can elicit high levels of fairness under communal norms.
  • In the case of artisanal/craft products like on Etsy -- and building on the person-to-person nature of sales in such a marketplace -- communal norms of fairness should be especially applicable to motivate high levels of generosity.
The key to making this profitable and manageable is the fairness reputation tracking and feedback controls of FairPay. The seller (or platform) can track how individual buyers respond to individual offers (and sellers), to learn how fairly a buyer sets prices for what kinds of products (and from which sellers). This provides a database on value perceptions and fairness for each buyer that can be used to manage what is offered to specific buyers (by which sellers), so that sellers can control their risk and nudge individual buyers to maximize their fairness.
  • Offers can be restricted to only those buyers who have a reputation for pricing fairly for the class of product being offered, so that sellers have a reasonable expectation that they will set a fair bonus price.
  • Sellers can decide how much risk they want to take and how wide a market reach they want. Those who prefer a lower number of sales at higher prices can limit their offers to those known to price generously. Those more eager to expand the quantity they sell, at some greater risk to their profit margin, can expose offers to a broader segment of buyers who price fairly but less generously.
  • Some sellers will set liberal fairness thresholds for some products for unknown buyers, so that their behavior can be learned at manageable risk. They may do this with selected product lines (or for limted promotions like Everlane's Christmas sale) that they can use for testing. Tighter fairness threshholds can be applied for sellers or product categories for which they want only more generous buyers.
  • In the case of a multi-seller marketplace like Etsy, the personal reputation data of buyers that is collected by the marketplace need not be exposed to individual sellers (to protect privacy) -- the marketplace can simply avoid matching a buyer to offers from sellers who set a fairness rating threshold that is higher than the buyer's fairness rating. All the seller knows about the buyer is that any buyer who see their offer has at least the desired fairness rating.
  • This mechanism gives a buyer a strong incentive to price fairly and even generously, to maximize the number and quality of offers they see. Buyers will know that it is the most desirable offers (and the most desirable sellers in a multi-vendor marketplace) that set the tightest fairness thresholds -- so the less generous they are, the fewer top quality offers they can expect to see in the future.
Why would this work?

Consider the lessons of conventional pay what you want (PWYW) offers.
  • PWYW has proven reasonably effective for both virtual and real products/services. People can be motivated to willingly pay fairly even when they do not have to.
  • Many sellers of digital products like music and games have done PWYW offers with a minimum price set to at least cover download and credit card transaction costs, with good results. (Additional evidence may come from sellers like Everlane.)
  • Research studies suggest that price floors can be effective, but there is a downside to consider -- setting a minimum can signal a lack of trust in the buyer, or leave the impression that a fair price is not much above the minimum.
  • In the case of real goods with substantial costs, it seems likely that the risk-mitigation of a price floor is more important than the signalling concerns.* Care in framing the floor price as not really fair -- in that it provides no profit and is thus not sustainable -- can help push generosity upward -- as can care in how the suggested profit margin is framed.
So it seems there is good reason to think this could work well for many real goods. Everlane seems a promising example, as a retailer seeking to establish an image for value, fairness, and transparency. Similar advantages may be applicable for design/craft/artisan products -- the seller can emphasize the human value of the artisan. Such use of FairPay could benefit a multi-seller marketplace like Etsy, especially where buyers are unsure what to expect from a seller they do not know (and vice versa). This could be good for the buyer, good for the seller, and good for the marketplace.

Could it work for very high-end products? -- such as for Tiffany? Perhaps not as well as conventional pricing, since at the very high end, high set prices are a signal of exclusivity -- a vendor with cachet like Tiffany can command prices that less prestigious brands cannot. I would guess Tiffany will be among the last places to try FairPay, for that reason. But who knows what variations might become workable once FairPay becomes widely used and understood?

----------

[Update 12/31:]  Wide press coverage of this sale in Business Insider (twice), NY Magazine, Inc, HuffPost, Daily Mail, and others shows the promotional value of PWYW offers, Hopefully they will also report on Everlane's results.

Some of this coverage raised concerns about how well PWYW works, notably in NYMag. Here is an expansion of comments I posted on that article:

This raises many interesting and important questions about how to apply new participative pricing methods like pay what you want (PWYW) that try to find a win-win with the customer -- but we are at very early stages of understanding how to do them most effectively. I believe Everlane is on the right track, and that with proper framing of the offer, and what is expected of them, PWYW -- and more advanced variations on it, like FairPay -- will change how we buy things.

The cited research by Gneezy (which I included in my Resource Guide to Pricing) and others is very interesting, and offers many insights, but does not tell us what results can be obtained with better framing (and after people gain familiarity with such new approaches). A more established example of PWYW is tipping in restaurants. True, it makes some people a bit uncomfortable, and some want to eliminate it, but most of us manage to do it as second nature (apart from any arithmetical challenges that an online system would eliminate). We simply look back on the experience and consider whether service was better or worse than average (intuitively considering many factors, including how we feel about the server, and our plans to frequent the restaurant in the future), to come up with a tip that seems about right. With some experimentation, much more effective variations on how to present PWYW offers can be explored and refined.

(BTW, Panera has been doing PWYW continuously since 2009 in their Panera Cares locations, and are now serving about one million people per year.)

Simple improvements to PWYW:  Here are two very simple things Everlane and others could test to improve their results (even without the sophisticated feedback control process that FairPay adds):

  1. Provide a slider that allows any price within an allowable range. For the coat example, the quantum jumps in the allowed prices are quite large: from $110, to $132, to $225. Maybe I am willing to pay $150 or $175 because I want to support them, but $225 just seems a bit steep for an overstock sale -- but I only pay $132.  The same framing levels could be presented, but with the ability to pick an intermediate price that seems fair to me.
  2. Let people pay the base price up front, and then follow up to ask them to decide on the bonus price after they receive the product and know how they like it. This would be only a bit more complex to do, would still assure costs are covered, but would gain all the benefits of post-experience pricing. Instead of wondering if I will really like the coat, and pricing low, because I am afraid to end up disappointed, I would know how much I liked it, and not have to discount from what I would later agree would have been fair.
----

*Of course a price floor can be used in any FairPay context, including digital goods with low marginal cost (just as is done for PWYW offers). Market testing (as by Everlane) is needed to understand under what conditions that is desirable, and how to set and frame such a minimum price.

(Acknowledgement:  My thanks to Florian Gypser, an architect and designer based in Austria and Thailand who contacted me to inquire about using FairPay for consumer fashion products. That led me to begin this post, to highlight these opportunities. ...And to Everlane for provding a nice case study -- which I hope they will add to with a report of their results.)