See my Pricing articles at IterativePath blog

I have moved all my pricing articles to http://iterativepath.wordpress.com
Please search there for other unbundling articles.

Further Unbundling of Baggage Fees

United Airlines was the first US airlines to start charging separately for bags checked. Now they are further unbundling this service with a separate fee for bags checked in at airport.

Baggage Fee Unbundling

There is an additional fee to check in bags at the airport vs. checking them before flight. Once again this is very well executed pricing change with focus on customer reference price. Look at the right upper corner of the image, they are positioning online pre-checkin for bags as a convenience and time saver. The additional fee for airport checkin is not something that is forced on all customers  as most will add bags when they checkin from home/office/hotel.  The additional price is not about the cost since cost is irrelevant to pricing. If there is value to customers in delaying baggage checkin till they reach the airport then they United Airlines must be able to get a fair share of that value.

If there is one thing they could have done differently is to take a cue from Prospect Theory and position this as one price $25 (not $20) and for online checkin offer a $10 discount instead of listing the fee at $15.

Pay Per Article – Possible Business Model

The problem we identified before with newspaper pay per article is the transaction cost. One way to address that problem can be to have a third party pay for the article access. It is not much different from Ad supported except that the article (or certain number of online impressions of it) are sold by the newspaper to a sponsor. The latter then decides how to distribute it, whether to charge for it or not.

I was at a bookstore the other day and I remembered reading WSJ article on summer readings. I looked it up on my BlackBerry using my WSJ online subscription. An article about a list of summer books is of value not only to the readers but also to the bookstores.  So WSJ could sell that article to Borders (and anyone else) for a fee. Borders then make this article available free to all. Their monetization is by adding “Buy Now” links to the article for each listed book and generating book sales.

Such a pricing scheme is still unbundled, each article is priced differently and sold separately  not to every single customer who wants to read it but to  distributor/middle-men.

Unbundling Across Product and Usage Dimensions

Unbundling can be done either across product dimensions, time or both. That gives four possible pricing scenarios:

Monolithic Pricing: Customers pay once and get all product features whether or not they need all the features and for all the time.

Subscription Pricing: This is pay-as-you-go scheme. Customers pay periodically and get all the features.

Add-Ons: Customers pay once and pay only for the basic version but can pay additional price to buy Add-On features.

Usage Based Pricing: Customers pay only for the features they want and only for the time period they need the features.

Note that both product and time dimensions can be further subdivided into coarse or fine granular units. For product dimension  customers can either buy next higher version with a set of features or buy specific a add-on without having to upgrade to next higher version. For time dimension customers can either buy pre-set time units like monthly subscription or pay per use based on metering.

Here is a graphical representation of the two dimensions.

Started a Google Knol for Unbundled Pricing

I started putting together a Google Knol for unbundled pricing. This blog has been a collection of multiple ideas and examples and does not have an implied structure. The knol will attempt to make it easier to understand unbundled pricing with links to details.

Should att reduce its iPhone subscription price?

Business Week reported that at&t may be considering reducing its current $30 per month data plan to $20/month for new iPhone subscribers. at&t already subsidizes the price of iPhone up to $200 and currently require iPhone subscribers to pay a minimum of $30 per month for the data plan. The service is offered only as a bundle – getting the data plan is required to activate iPhone.  Should at&t reduce its price by $10? What is the impact on its current subscribers? How should they do this so the incremental profit from this price reduction is greater than their current profit? Let us do this by looking at why they are doing this, industry cost structure, competition and customer perceptions.

US Mobile phone penetration is more than 90% and a service provider can grow its subscriber base only by stealing market share. As the Business Week story reports most customers in the high income bracket with a higher willingness to pay for data service are already taken either by at&t or by Verizon. The current churn rate (% of subscriber base that switch service) is about 1.1% for at&t, this means an average at&t subscriber stays with them for 8 years.

A service provider’s network costs are mostly sunk and even the ongoing maintenance costs are fixed its marginal cost to serve a customer is $0 – so it makes sense to add as many new subscribers as possible to amortize the fixed costs and deliver profit. So no surprise that at&t wants to grow its current subscriber base. The only relevant cost is the customer acquisition cost which is about $200 (including iPhone rebate) to customers.

Given these numbers, the lifetime value of a iPhone subscriber over  8 years (at 10% discount  rate) is $4400. So adding every subscriber and keeping them longer adds directly to their bottom line.

Despite reporting strong profit and increase in number of iPhone subscribers, at&t added fewer subscribers than Verizon wireless who beat out at&t with its portfolio of devices. at&t has not said explicitly how many unique iPhone subscribers it has. One source is to find from Apple’s earnings statement but ArsTechnica summarizes the number of  iPhone activations that it culled from at&t quarterly earnings report. The total iPhone activations since it was introduced is 8 million,  of which 2.1 million were pre-3G. If we assume 50% of the pre-3G iPhone subscribers upgraded to 3G the number of unique iPhone subscribers is  about 7 million.

If at&t simply reduced the pricing by $10 its current 7 million subscribers are not going to be happy. Like the iPhone price drop outcry that happened when Apple dropped the price of its iPhone by $100, there is bound to be customer demand to reduce their pricing. This means a total monthly loss of $70 million. So just to get back to its profit levels before the price drop at&t should acquire  1.16 million new iPhone subscribers.  Note that this 1.16 million is above and beyond its current subscriber growth rate  which is 1.6 million new subscribers per quarter. Since the competition is not going to stand still it is going to be a tall order.

So at&t will definitely introduce versioning and introduce the low priced plan with service restrictions/impairment. While it helps to avoid the need to reduce the fee across the board it does introduce two new problems. First, some of its current subscribers will switch over to the lower priced plan and second, majority percentage of its new subscribers will pick  the lower priced plan. If we assume a 50% downgrade and 100% pick $60 plan, then the  new break-even number of new subscribers is 580,000. This is still a high number.

So should at&t reduce its price? I think it should not to keep its profit at current levels.

An Alternative To Unbundled Pricing – Service Reduction

Are there cases where a business can simply eliminate a service component instead of unbundling it and charging separately for it?

There is a very nice Italian  restaurant called Trattoria La Siciliana in  Berkeley, CA. Their offering is premium – authentic sicilian cuisine, great wines and  great service. This is also a small place and is always packed. There is one thing – they accept only cash. No checks or credit cards. Checks are simply pain to process  but most people  expect to pay with credit cards especially when they add a bottle of wine to the dinner. Why does not the restaurant accept credit cards? I did not ask them but here is my theory on this.

Credit card processing may be free to you and I as users but the merchants pay a fee to the bank for each transaction. The fee, referred to as interchange, runs close to 3%. So for an average tab of $60 for two people, the net intake is reduced by  $2.8. Small businesses like restaurants are hurting with slowing economy and 3% fee is  directly affects their operating margin.

Paying with credit cards is a convenience. So why not restaurants unbundle this fee and pass it on  to the customers?   Why not ask the customers to pay for the convenience by simply adding the fee when a customer pays with credit card. Some businesses like gas stations do just that. But as we have sen before in this Unbundled Pricing blog, customers are bound to fee nickel and dimed.

In the case of premium priced restaurants,  like Trattoria La Siciliana, adding a separate fee does not fit the brand promise or its product offering. So instead of unbundling the service and risking brand erosion a business can simply not offer a particular service component. This is product/service  unbundling as well except that pricing is not unbundled – the unbundled service component is simply not offered.

Unbundling The Product For Micro-Targeting

PowerFrameworks is a website that offers Microsoft PowerPoint frameworks for visually showing concepts. Their pricing strategy stands out as an example of pricing done right. Their pricing is a very good example of product unbundling and pricing the unbundled components separately.

What they sell it as a subscription service with recurring revenue. Their pricing is unbundled along both usage and product features dimensions

1. Subscription model is unbundling the product along the usage dimension – instead of buying the product all at once the subscribers pay for usage.
2. What they offer with subscription are PowerPoint slides  with the frameworks, about 400 of them,  that a subscriber can download. For those customers not willing to pay the subscription fee ($149) they unbundle their product and sell individual frameworks at prices starting from $5.50.

Their unbundled pricing for each framework is a clear recognition that the needs of the their customers are not all the same. Hence by unbundling the product they target each niche and allow these customers to purchase only what they need. This reduced customer resistance to purchase and exposes new revenue opportunities compared to a single price.

Another notable point is their unbundled pricing for each framework  ensures that  those who need more than a few frameworks will buy the subscription and not buy them one at a time.Even at the lowest price of $5 per framework and assuming an average customer needs only 50 frameworks, the sum of the prices of the unbundled components is same as that of the annual subscription.

Powerframeworks’ unbundled pricing offers lessons to newspapers trying to unbundle their pricing.

The Wall Street Journal Pay Per Article Coming

The Financial Times reports that the WSJ is considering unbundling its online publication WSJ.com to offer pay per article scheme and is working on a Micro-payment scheme. Currently WSJ offers subscription only pricing for its online version for $99/year and is available at a discount when bundled with its paper version. Previously I have written about unbundling the WSJ. Here is what I had to say then,

…in 2007 WSJ had 1 million paying subscribers and 20 million monthly readers who read the free articles. This 20 million readers are the potential market for unbundled pricing. When these readers click on a subscriber only article, the website will ask, “This is a subscriber only article but you can read it now for $0.25″. Suppose if 10% would take this offer and read just one such article at a price of $0.25, that is half million in additional profit per month, or $6 million in yearly profit.

This was just based on my assumptions on numbers. The WSJ has not announced their pricing for pay per article customers. The Editor-in-Chief of WSJ, Mr. Robert Thompson said (from FT.com story):

Pricing for individual articles and for premium subscriptions had yet to be decided, he said, but would be “rightfully high”.

This is of course pricing common sense. The price of the unbundled WSJ should be higher on a per article basis, otherwise readers can put together their own bundle at a cheaper price than the yearly subscription fee.

There are two problems that need to be addressed with unbundling a service and WSJ is solving both successfully:

1. Reference Price: If customers had had never paid for an individual article before then their reference price would be $0. But the reference price can be improved even with related offerings or decoy offerings. WSJ charges a high price of $4.95 per article for archived articles (older than 90 days). This serves as the reference price for the new pay-per-article plan even though the two services are different except for the fact that these are pay-per-article plans.
2. Transaction Costs of Unbundling: The WSJ cannot ask readers to pay every time they want to reade the article. This is cumbersome to the readers and there are transaction costs to WSJ. The FT.com story reports that they are working on their own proprietary Micro-Payments scheme to address this. It is conceivable that their Micro-Payment will take the form of prepaid mobile, asking you to buy $10, $25, $50  value credits and deduct from this every time you read an article. They will definitely have ways to send these credits as gifts as well.

One last point is cannibalization, loss from current subscribers switching over to pay-per-article scheme. If the price of unbundled WSJ is done correctly there will be minimal cannibalization. WSJ can further minimize this by communicating the value to subscribers by constantly reminding them how much it would have cost them if they had switched to unbundled pricing scheme.

A quick survey of articles written this morning on this topic shows that most bloggers and commenters in their blogs are not happy and do not think this plan will work. Techdirt blog calls the pay-per-article schema a sure failure. I disagree. The net is, I think this is a great plan. Getting $6 million or more incremental profit is well within reach with an unbundled WSJ.

Note: I quote the FT.com link here and not one from WSJ.com because I could not find it easily on the latter.

Going from unbundled to bundled

The reference price experiment my colleagues and I conducted recently, proved our hypothesis that people do not like unbundled pricing because their reference price for these was $0.00 but can be nudged towards unbundled pricing if their reference price can be improved. But what if the marketer who offered unbundled option wants to shift the customers to the new bundled option? How should the choice design be?

In this reverse unbundling case there are two possible scenarios

1. The bundled product is the sum of its parts – no other value-add. In this case the price of the bundle should less than or equal to the unbundled option as previously discussed.
2. The bundled product has a new component that was not offered  before and is not made available in the unbundled option. In this case the price of the bundle can be more than the sum of the prices of unbundled components, if and only if the value proposition of the new components are clear to the customers and they have a non-zero reference price.

In the second case, if the value communication is not clear then customers will be highly unlikely to pay the premium for the bundled option. More importantly, if the reference price for the new value-added component is $0.00, then customers will be more than likely to prefer the unbundled option.

So in both unbundling and bundling cases, reference price is key. If marketers want to capture part of the value-added they should focus on setting a higher reference price in the minds of their customers.

Unbundled Pricing Cannot Be Lower Than The Monolith

Unbundled pricing should be about exposing revenue opportunities and capturing part of the value created to the customer. The pricing should be such that the sum of the price of unbundled components should be greater than or equal to that of the monolith. Otherwise customers are incented to go with unbundled option and the marketer will end up making less than they did before unbundling.

One exception is that there is value inherent in the monolith that is not available in the unbundled version even if consumers buy all the components. This usually means that the unbundling leaves out a component (physical or logical) that must be provided by the consumer. Since the consumers provide this component they must be compensated for it in the form of a total lower price than the monolith.

Let us walk through an example. A computer system can be sold as one bundle in which case the manufacturer provides the additional value in the form of assembling, system integration, verification and warranty. Or a consumer may buy the parts and assemble this themselves and hence may be able to purchase all the components for a total price that is less than the bundle but will have to do the assembly, integration and verification and must settle for warranty of components.

In true unbundling all components must be unbundled and available to for purchase in a la carte mode to consumers. In the computer example, the manufacturer must sell assembling/integration/verification/warranty as separate components available for purchase.

So for true unbundling, for a marketer to be indifferent to unbundling vs. selling the monolith the sum of the prices of the unbundled components must be greater than or equal to the monolith.

Unbundling At Checkout Counters – Paper Or Plastic?

Almost all supermarkets checkout clerks ask you this question, “paper or plastic?”. You are not charged extra for the bags but it does not mean it is free. The cost is covered by the margins earned on the goods sold. Most stores give you a bag credit of 5 cents for every bag you bring in. Is that the right approach? People respond to incentives but they respond more to disincentives (Loss Aversion).  It is not obvious to the customer at the checkout counter that they are forgoing savings. Even if it is , according  Kahneman and Tversky, the intensity of positive feeling from gain of five cents per bag is not as high as that of the negative feeling if they were to be charged 5 cents per bag.

When people see a 5 cent charge per bag they are more than likely to bring their own bags than if there were given a credit for each bag. Charging for bags is unbundling. It fits the pre-condition I set for true unbundling, the service is truly optional. Unlike other unbundling scenarios there is no need to train customers or improve their reference price. A reference price is already set through the per bag credit practice.

At least one retailer, IKEA, figured this out and charges you for every bag used for packing your stuff.

I propose businesses unbundle the price of shopping bags, for the sake of environment and for profitability?

Is Unbundled Pricing A Cure All?

Unbundled pricing is not a panacea and it is not for very product and service. In my previous posts I talked about the cases where unbundling is not the right solution.

1. Five reasons to not unbundle pricing
2. Do not use pricing to solve operational inefficiencies
3. Brand erosion from unbundling (Hulu and  Networks)

However unbundling is not to be shunned as this blogger says :

By avoiding the practice of unbundling, you might take a hit in the short term because of your competition, but it’s a long-term play that will win out in the end.

It is true that  unbundling made as a tactical reaction may turn off your customers. But customer perceptions can be managed. Unbundling should be part of your pricing strategy that flows from understanding your customer segments and what they value. The key rule of proper segmentation is “Offer them only what they want and none of what they do not want”. For example, in the Airline unbundling case,  baggage fees are not a concern to most business travellers who travel with just cabin bags or whose tickets are paid by their businesses. Why offer them a free baggage service if they do not need it or value this service? There is nothing more strategic than customer segmentation. Know the segments and the value proposition a marketer can successfully practice unbundled pricing without brand erosion.

Unbundling also exposes new revenue opportunities, by modularizing the service a marketer can see the value to customers from each module. Understanding the modules will enable the marketer to improve them, there by creating value to customers and capturing some of that value. For example, I talked about Amazon free shipping and how they can unbundle their free shipping option and provide a better service to  some segments.

From a customer perspective it is not always nickel and diming. Unbundling is  also about fairness, making sure a customer is not paying for services they do not use and hence not subsidizing others who use it. You can see this in airline baggage example. Another example is if BART (Bay Area Rapid Transit) were to charge a bundled fee for ticket and parking. Why should people who do not use parking at BART pay for this service? You can see the fairness argument apply in most cases.

The net is, Unbundling is an effective tool that should be part of every business’s strategy.

Using Conjoint For Unbundling

Here is a quick tutorial I put together on Conjoint analysis and Cluster analysis. This is a work in progress.  I need your feedback to improve this so it helps you and others.

Unbundling the Airline – Measuring Customer Utilities for freebies

The last experiment I conducted was on pricing freebies and how Airlines or any other marketer can improve their customers’ perception of unbundling pricing by focusing on the reference price. The logical next step in the unbundling research seemed to me to find out the relative value of the components of the monolith and what is the best mix of unbundled pricing that maximizes customer value (utility).

For example, for Airlines how much does the customer value the no-baggage fee option over price? Should an airline increase its ticket price or unbundle its baggage fee? Is charging $2 for a drink bound to affect customer utility.

Please note that this study is orthogonal to the previous study on improving reference price through multiple options. In that study we concluded that presence of expensive options  tend to reduce  customer aversion to unbundled pricing of freebies.

One thing that stood out in the last study is that despite the reduction in customer aversion the overall preference of the airline was unaffected by the presence of options. In the current study, a team of us here at Haas School of Business designed and conducted a conjoint analysis to study just that.

Conjoint analysis :

Infers the relative importance of product features and how objective values of products relate to their desired values by analyzing how consumers evaluate an experimental array of hypothetical product profiles.

The results are out, please stay tuned for the marketing implications.

Unbundling the dressing from the salad

Most health conscious people order a salad with dressing on the side or with no dressing. But the price quoted for a salad in any restaurant includes dressing. In other words the sald and dressing were seen as a monolith by consumers. Whether you like the dressing or not you pay for it in the bundled price. So isn’t that an ideal candidate for unbundling? That is exactly what Mama Stella’s Pizza Kitchen is doing in its San Diego Sea World outlet. The salad price does not include dressing which is sold at an additional cost of $0.75. Mama Stella has also unbundled the bread sticks from the sauce. Their bread sticks do not come with dipping sauce, you need to purchase it for a fee.

In an amusement park that serves tourists and seldom any regulars, unbundling is not a difficult task. Since people come to expect very high price and those who buy had no other option there is no pricing resistance. In fact instead of jacking up the price of Salad unbundling it helps to distribute the price increase across multiple components.

Unbundling in Mama Stella’s and other such amusement park restaurants is interesting for  another reason, the pricing for the parks. The parks practice two-part pricing, an entry fee for the park but no per ride fee.

Results from the Reference Price Experiment

Unbundled Pricing – A Reference Price Solution

View more presentations from ragsvasan.

The Downside of Unbundling

Unbundling is not all about identifying new revenue opportunities and converting cost centers into revenue opportunities. In one of my previous posts I wrote about five reasons to not unbundle. One of them is losing control of your main brand and the emergence of component brands.  For example, TV network shows like NBC’s  “30 Rock”. When viewers watch NBC on TV, they get not only all episodes of  “30 Rock” but every thing else from NBC. But NBC has unbundled its offering and is distributing individual episodes of its shows through Hulu.com.

Viewers on Hulu now can watch just “30 Rock” and to them it does not matter that it is a show on NBC despite the fact that NBC logo appears at the bottom of the screen. This means the “30 Rock” brand becomes more prominent than the NBC brand as it becomes irrelevant to viewer decision making. Moreover NBC may end up losing control over what advertisements are shown during the show.

When the show is broadcast on NBC it can use the popularity of the show to push its other shows. But on Hulu, it loses that control.  Specifically one episode of “30 Rock” has an  advertisement for ABC’s new show Castle. ABC is able to do this only because of the unbundled offering of “30 Rock”.  Unbundling will lead to a shift from a strong brand platform to competing on strength of component brands.

If  the parent company controls the component brands then it should not matter much but if the component brands are owned and controlled by others the parent company risks becoming irrelevant. Before unbundling the product any marketer should consider this long term loss from shift in value chain despite the short term gain from unbundling.

Ryan Air Unbundling the In Flight Toilet – “Loodicrous?”

Ryan Air, the pioneer in unbundled airline travel and the first one to charge baggage fees, has come up with the plan for next level of airline unbundling:  unbundle the facilities and charge passengers a pound each time they want to go to, as the British would say loo.

Times Online quotes Ryan Air CEO from his BBC interview,

Michael O’Leary said that the carrier had been investigating fitting coin slots to the doors of aircraft toilets, similar to those installed at train stations.

“One thing we have looked at in the past and are looking at again is the possibility of maybe putting a coin slot on the toilet door so that people might actually have to spend a pound to spend a penny in future,” he told BBC Breakfast this morning.

Can the airplane toilet be unbundled? There are problems with this unlike the baggage fee and in flight drinks. The necessary condition I defined for a unbundling to be possible is, the component should be truly optional and there should be other options for the consumers.

For example

1. with baggage fee travelers can pack light,
2. with charge for in flight drinks travelers can bring a drink they bought outside .

While one could argue that travelers may choose to use the toilet or other means (it Depend), this is a service component that has no alternatives and is not optional. So people expect this to be a required component that is essentially not  unbundle-able.

A statement issued by RyanAir, in response to customer outcry, said:

“Not everyone uses the toilet on board one of our flights but those that do could help to reduce airfares for all passengers. Then again, maybe O’Leary was just taking the p*** this morning.”

Next there is the fairness issue. Unlike baggage service where those who do not check in bags feel it is fair that the airline charges those who do check in bags, everyone knows they will need the toilet and will use it. So the move will face backlash from even those who do not use plane toilets on a regular basis.

Could RyanAir have handled this better and managed consumer behavior with a Reference Price option?  I am not sure. One way could have been have two kinds of toilet, a free version and a paid version. Either the paid version could be marginally better or the free version could be significantly worse. This preserves the precondition I stated above and alleviates the fairness issue.   But in the end if an airline is really looking at revenue opportunities from people using the facilities, it is certainly loo-dicrous times we live in.

Training The Readers To Pay For Articles

The Pay-for-content argument should not be an argument at all. When material goods are not available for free, why should information goods be? But the reason for the current arguments  from those who want the online news to be free and from those who say this is bound to fail is rooted in the reference price in the minds of customers. Until now most information has been given away for free before. The problem with moving from free to fee model is exactly this, consumers have been trained to expect the newspaper to be free despite the value they receive. Hence in their minds the reference price, t the price they are willing to pay for the service, is $0.00.

Some of the objections to the fee model and unbundled newspaper are around the technology limitations of micropayments. Those are easy to address, for example a Korean gaming company Nexon offers free to pay model but charges for game paraphernalia like weapons and shields. They solved the micropayment issues by selling high value gift cards  in Target and other retail outlets.

So the question is not technology and will it work, but how can the newspapers train their free readers to value free? The answer is how to improve the reference price.

We saw this in the experiment conducted on MBA students on valuing free soft drinks in airplanes. This is not much different from the current argument on charging for online content. As our experiment showed, when we improved customer reference price the resistance to pay for “free” reduced considerably.

Before moving to fee model the product mix and marketing communications should be about improving the reference price. I will be happy to help any news media planning to move from fee to free model.