INFORMS Revenue Management and Pricing conference - some feedback!

Dear Readers,

Earlier this summer, Julien went to the INFORMS Revenue Management and Pricing conference at Columbia University. Held in NYC, this event “is the premier forum for both academics and practitioners who are active in research in the fields of pricing analytics and revenue management”. The objective was to get up to speed with the latest research, practicesand burning topics in the industry.

He was amazed to see how Revenue Management and Pricing (RMP) techniques are developed and widely used in the United States. Most of the conference participants originate from the US and this discipline seems far more advanced there than in Europe. There are business cases in many unconventional areas (e.g. Google, with its Ad Exchange is using RMP techniques to grow the online advertising pie in an unconventional way – more about that in an incoming article), and the markets have really understood the potential of the price dimension in the 4 Ps marketing mix.

We would like to share a few findings with you…

1) The prerequisites for RM revisited 
This conference enabled us to put this discipline in perspective; it also got us to rethink several ideas we had about pricing:

• Market power: it is commonly accepted that individual consumers represent a fragmented demand, which doesn’t have much bargain power when it comes to pricing. This statement doesn't hold in a context of "social learning", where information is shared, and expectations tend to converge.

• Market segmentation. For a while, we were wondering which came first: the chicken (pricing) or the egg (revenue management)? Well, for sure, pricing is not a sub-set of RMP (as well as inventory management is)- it goes along with it and is on a higher level of analysis. It has a strategic aspect that would lead RMP in the organization. Actually, proper market segmentation (and calculation of pricing break-even points) is probably the real pre-requisite to any price related activity. 

• Price as a signal: It tends to be accepted that RMP cannot be implemented in an organization where the price a signaling characteristic regarding the product. Research has gone further and presents the price as a signaling device of product availability (demand forecast vs. inventory available). On this regard, we are looking forward to the publication of "Markdown Management: Pricing as a Signaling Device", by Gad Allon (Northwestern University), AchalBassamboo (Northwestern University), Ramandeep Randhawa (University of Southern California). Adapting pricing to product availability is an issue that most companies are facing : Who has never wondered what is communicated by the price a good is sold (eg. Used items on ebay, amazon marketplace, craigslist etc)?

Above all, the concept of perishability remains: a product/service is lost when not sold due to a limited life span. Of course, some pre-requisites vary, and must be stabilized on a case by case basis.

2) A taxonomy of RM
Acknowledging RMP by sector is, we believe, not the right level of analysis. After having attended this conference, and thus analyzed the latest research, we believe that RMP could be classified into three core areas.

RMP focuses on the “Maximization of Revenue” through:

• A proper segmentation and an optimized pricing policy: This is a key component when willing to market you product differently to different groups (bundling questions, break-even point for segmentation, demand and promo management). This must be sufficient when no real binding constraint can be found 

• The management of a limited inventory: This is the origin of RMP 

• The management of a network’s congestion where capacity has a very limited life span, and could not be stocked: This is under implementation by Telcos, Energy providers … Stochastic demand analysis is at the heart of it. 

3) The power of social learning in RMP
As mentioned previously, strong social learning reshapes a market's forces.

A very interesting presentation dealt with the issue of whether the customers manage to get a rationale and can understand basic RMP concepts. Therefore, they could act upon it and compare the quality of products (don’t mix the comparability of products and their lookalike)

“Our model postulates a fairly simple learning mechanism. Given the number of past agents that purchased and of them the number of agents that liked the product, each agent forms an estimate of the quality sensitivity parameters (QSP) of the marginal agent that purchased and liked the product. The agent then compares this marginal QSP to his own QSP.” In "Monopoly Pricing in the Presence of Social Learning", by B.Ifrach, C.Maglaras, and M.Scarsini.

This presentation offers a vision of customer behavior and its impact on market actors. More to come in a separate article

4) The impact of research on RM practices
The RM industry is interesting in its ability to absorb state-of-the art research, and to always go further into details. Lately, this has resulted in an ability to:

• Build disruptive and even more robust RMP : several business case emerged for various new industries. Demand shocks management is one of them. 
• Use an industry minded research, with a focus on reality: Cancel and rebooking behavior and how to act upon 
• Build incremental innovation: Bundling services and its efficiency 
• Gather a growing number people from various horizons and nationalities 

5) RMP and value creation revisited
Boarder line subjects are lightly emerging and we are thrilled to see that coming.

Once again, through this blog, we are trying to explore how organizations can implement Revenue Management and Pricing techniques.

The first hurdle we have to overcome is the financial evaluation of a RMP project. Several options are available: i) The traditional NPV, ii) an NPV model associated with a Monte Carlo simulation (used by Hilton Hotels’ Revenue Management), iii) through an option-like valuation (This method, and the business case for RMP will be available later on this blog). Let’s not neglect the other aspects: the strategic fit, Finance and cash flow considerations, the existing IT infrastructure, People and the corporate culture, and the Marketing and Communication.

Implementing RMP should also mean bringing superior value to the organization. That is the article that we have initiated earlier and that we will keep on updating.

In short, this conference helped us realize how fast the Revenue Management and Pricing field is moving. In order to get a grasp of how innovative and far reaching RMP is, we will publish a couple of industry focused articles within the next few weeks.

Thanks for following us!

Julien & Yoann

Introduction to Revenue Management as a sustainable value creator for the company

Value creation is driving companies at every stage of their activity: from products to shareholders, including employees, the community, the environment, etc. The definition of Value creation is wide, and there is no consolidated literature about it: IndustryWeek has a whole section dedicated to value creation, and companies have their own view on this issue (see what value creation is according to BHP Billiton). The finance oriented approach is central. Corporate Finance from Brealey Myers & Marcus explains extensively own a project "creates value" for its owner/investor. Marketing Management, from Kotler and Keller, drags the concept of value creation all along the book. Strategy mainstream authors, and especially Michael Porter  in his early papers (Competitive Advantage, 1985) prefer to define it using the concepts of competition and the company : Competitive advantage brings a sustainable added value vs. competition and the market. We believe this last definition symbolizes the most Revenue Management in its expected effects. Now, let's try to make our own generic and simple definition... there we go: Value creation is what brings a superior worth to a good or a service, on the perspective of a market. A good can be a company, a business unit, a product and it is to be viewed as positioned within an environment.

This is much discussed at every level of the corporate ladder. We believe this question is key for any RM implementation proposal. It is the role of the manager supporting the idea to bring some milestones along.

Evaluating the financial benefits of the implementation of an RM system in a company quickly proves to be complicated: We will dedicate an article to address this issue. Indeed, one can calculate an expected rate of return on an investment in software, machines or any tangible asset, where the material’s life span, cash flows and depreciation period are known a priori. 

Here, we are talking about a change that is probably one of the most difficult to assess in a company: The impact of a new, structural management methodology, which is all but certain. The greater the expected returns are, the higher the risk is (yes... we have heard of this too!).

In this context, we think the concept of sustainability makes sense. Implementing a new management methodology is a long term project. To create value, Revenue Management first has to be seen as at the crossroads of strategy, finance and organizational theory; and secondly, optimization, demand forecast and customer behavior issues can be addressed. Building a transverse management methodology is a key prerequisite to generate organizational alignment and thus survive external events.

That is why we believe that a structured approach must be drawn: This article proposes a series of steps that would enlighten the feasibility study to support such a project. We therefore list a couple of steps or milestones that one should address when you coming with the idea of implementing Revenue Management.

Step 1: Be aware that you may find yourself in a difficult position

Due to the influence of capital markets and private equity (and the related expected returns), business today (and probably yesterday as well) tends to focus on high yield and short term payback, The investor follows the risk paradigm and expects a proportional (or at least not contrary) return / yield. Therefore, everyone can understand that an investor, who is endowing in a firm in capital need to finance projects, is experiencing a risk, especially if the investment is about implementing a new, challenging management technique!

From our experience and readings (see Talluri and Van Ryzin, The Theory and Practice of Revenue Management), it takes quite some time to implement Revenue Management; even more if it is the first time: the needs have to be properly defined and studied, the organization’s rules, design and processes have to be reengineered to support RM. This means that, whatever the size of the business, a transition time is needed to get from the current to the targeted /optimal situation. An implementation can take several months up to several years. 

This puts the manager in charge of the implementation in a difficult position: Investors could expect a “return” before the end of it, and will ask for an assessment of the first results of the newly implemented Revenue Management system. There is an inherent contradiction between a latent short-termism, and RM’s value creation which can take a bit of time before ramping up. This underlines the importance of sustainability.

Step 2: Build something called strategy, or at least, steer your organization. 

First of all, Revenue Management is not a secret formula, enabling managers to create growth and value, forever (neither is value creation in general by the way). We think that such a sustainable value creation is the coordinate action of an organization with aligned people, methodologies, and stakeholders, willing to make the move and bring the company to the next level. This implies the alignment on long run business objectives and changes together with a short run operational management. Making this change be part of a strategic positioning would have more success and meet expectations (Michael Porter,  What is strategy?, HBR Nov-Dec 1996) This whole is called “strategy”, and 2 companies out 3 lacks one – according to Henri Bouquin, Head of EMBA at Université Paris Dauphine, France in his book, Les fondamentaux du contrôle de gestion.)

Step 3: Try to assess a time period for Revenue Management. 

Revenue Management has currently no standard usage time (even if depreciation is a calculation made by accounting…but everyone knows how accounting can sometimes not represent reality!). This is a key as depreciation is based on utilization rate or time of the asset Obsolescence, from to our point of view, can only come from superior optimization and forecasting algorithms within the competition, preventing your company to reach all your usual revenue streams. To our opinion, the prerequisite to the kick-off of a project, before convincing any investor and being misled thereafter, is assessing timing and usage time. For example, usage time could follow the life cycle of the product or the foreseeable market macro-changes. This way, you can adapt your management to your activity. Basic idea, but robust in a globalized world!

Step 4: Assess revenue increase and value creation.

This will be the full content of an article to be published later in May

Step 5: Keep in mind your (high) fixed costs and the levers to keep Revenue in a good shape

Rule of thumb: Revenue Management will create positive results as long as the activity of your company is not affected by uncontrollable exterior factors (we’re pretty smart on this blog J). If the activity slows down, revenues may lower, and it will be quite critical to determine whether or not Revenue Management is generating superior value for the company. In this context, value is strongly needed by the organization which has, more than ever, to sell to high yield customers. This is when capacity adjustments have to come into play. Indeed so far, one of the hypotheses was a fixed capacity. . However companies can to take advantage of that and must have rescue plan that assess the opportunity cost of shutting down part of the capacity.

Even if Revenue Management works well in capacity constrained industries, it doesn’t mean that it should not be changed if needed. This is easy for some businesses like hotels and cruise lines for example, where you can scale down more or less rapidly (selling part of your assets) and try to diminish slightly your costs. Unfortunately, other businesses (parking lots, restaurants for example) have more difficulties to reduce their capacities due to limited levers. Indeed it is not easy to fire employees in restaurants in Europe (ok… in France!) to wait fewer tables for example. Anyhow, shutting half of a car park or of a restaurant off would not have any effect on reducing the costs, as opportunity costs remain for the whole occupied space.

Step 6: Have a look around.

Of course, your market is not revolving around you and key players could look at each other and find ways to set up alliances, leaving you behind, alone. This has happened recently in the airline industry: the three main alliances (Star Alliance, Skyteam, Oneworld) created JV to form oligopoly in some transatlantic markets. This leaves remote players like SAS, Finnair with low market shares. Their only alternative to keep the statu quo is low prices. This is for sure a more of a strategic subject, but RM is a topic for alliances (Can we align our pricing policy? Who will manage the inventory?).

Well, as an entrepreneur or a small business owner, you may not feel concerned with what is said above (even if we think that for every company size, there are markets and predators which will, as it is the goal of most corporations, grow!). We still believe distribution is a core lever for the successful implementation of revenue management. Indeed History shows that the first Global Distribution System (GDS) brought the first RM. Today, with the new techs, multi-channel distribution becomes easier than ever: Less complicated selling platforms have to be built (unlike GDS in the business travel industry. We are actively researching how could SMEs build their own distribution systems (based on a pool of companies, or one major) in order to secure the value chain and therefore increase overall chance to reach value creation expectations.

You might have noticed that in this article, value is not only addressed from a financial standpoint but also from an organizational and strategic point of view, as sustainability involves the whole company: This is our vision!

Despite the need for an overall view of business, numbers convince easily: We will go further in details for step 4 later this month.

See you soon!

Yoann and Julien