Monday, December 15, 2008

Pub Fight with BookNet Canada

One of the more entertaining and interesting presentations at the Supply Chain Interests Meeting in Frankfurt this year was one by Michael Tamblyn of BookNet Canada. He spoke about several things but the one subject that was most interesting to me was a development project that his team devised that can help junior staff members learn how to make accurate sales and demand forecasts. The 'game' PubFight was initally tested with a small group of publishers but was so successful that it has been expanded to include students in publishing programs as well as some booksellers.

While the principles are core to under standing how planning works effectively the tool was constructed so that individuals or teams can 'play' each other and the team that maximizes sales while minimizing costs wins over a period of time. In the first version of this program the BookNet Canada group used real sales data for real titles over several months. In the version for publishing education programs, the tool can be loaded with historic data to compact the - say four months, of selling time into a week of elapsed time.

I hounded Michael until he finally answered my five questions on Pubfight.

1. You spoke at Frankfurt about two different initiatives but it was the PubFight which caught my interest. Can you explain the program?

BookNet Canada is an agency tasked with supply chain and technology innovation, so our primary focus is service delivery -- point-of-sales tracking, process improvement, EDI, standards and the like. During a series of meetings with heads of some of our larger publishing houses two years ago, I started to hear a recurring theme that went something like this: "Our tendency in publishing is to hire passionate, well-read arts graduates who can communicate well about books. But we need people who can combine their love of books with an understanding of the numbers behind those books, people who can forecast sales, assess opportunity or analyze sell-through and stock position in real time. How do we build those skills for the people we already have? How do we make sure that people joining the industry have those skills when they arrive?" That request -- to get people familiar with the numbers behind the industry -- was the genesis of PubFight. We whipped up a quick in-house version, tried it with staff and a few guest-stars from the publishing community last fall, and turned it loose this year.

PubFight is basically fantasy publishing. It's fun, competitive and only accidentally educational. At the beginning of the (real) selling season, leagues form up, usually co-workers or students in a publishing program, sometimes competitors and real-life rivals, about 8-12 per league. It all starts with the auction, "Fakefurt". Each person gets a fake budget to acquire the real titles that are going to hit the shelves that fall. Everyone has to fill a list of fiction, nonfiction and juvenile/YA titles. Once the list is built, each player has to forecast initial print runs and pay for them. Then, as the books hit the street, you accrue the (real) sales on your titles as long as you have enough (fake) stock to cover it. You can reprint if you need to, with real-life lead times and unit costs. The most profitable house at the end of the season wins.

On one hand, completely frivolous. But on the other, it is encouraging people to do some analysis, take risks, and make mistakes without putting their jobs or the firm's money at stake. Along the way, they build an understanding about how books sell that could take years if they were just learning by experience. Both publishers and some publishing educational programs are using PubFight.

2. How do they use it? What about retailers?

Every publisher is different. For some, it's pure team building. For one large house, they excluded their own titles from the auction in order to deepen their understanding of what the competition was doing. For some of our small press players, it's a chance to look at how the more commercial, mass market end of the industry behaves. In all cases, whether they intend it or not, it's helping them become more familiar with forecasting, sell-through analysis, competitive title analysis and the other techniques that publishers need throughout their organizations.

At the colleges and universities, it's often the students' first look at how books actually sell, which can be a real eye-opener. It takes them right from theory to practice: How much is a book worth? How can you tell? To what degree is the past a predictor of the future? And it puts them in some remarkably true-to-life scenarios, like when the book that you bought for nothing becomes a runaway bestseller that you can't keep on the shelves. With schools, we can also run lightning rounds, where we run a complete past season in a couple of weeks, with a new week of data dropping every couple of days. Much easier to fit into an existing curriculum.

We see retailers as the next stop for PubFight. One of the biggest challenges faced by store and chain managers is identifying new talent. Which of the people working on the floor has the aptitude to become a buyer? Who can look beyond their own interests to predict what book-buyers are going to be interested in? It's reasonably easy if you're a small independent, but much harder if you are spread out across multiple stores or in a chain. This might help junior booksellers start to get a sense of how the industry works at a larger scale, and pick up some tips about the demand for different kinds of books along the way. It would also help senior managers get a sense of who has a knack for picking winners.

3. How do you see this program expanding? Is there are more practical implementation of PubFight – can the tool be used in actual forecasting?

It's more about encouraging the practice of forecasting than becoming a tool for forecasting. At the same time, we're interested to see if the positions taken by publishers and retailers at auction and on print-runs can act as a lightweight, EasyBake prediction market for future sales. In a frivolous and non-serious way, of course.

4. Are you considering licensing this tool outside Canada? It would be great if you had a Flash version of your Frankfurt presentation to explain it in full to publishers and retailers. Is this under consideration?

It's a possibility, if we can find a licensing model that makes sense. It does require a direct connection to a continually updated source of point-of-sales information, which limits the pool of licensees somewhat. In the meantime, this might just be one of those things that makes it worth a trip north, right along with colourful money, free healthcare and baroque parliamentary politics. When we get some time, we'll try to get the Frankfurt presentation online and let you know when it's available.

5. Do you have any development plans for PubFight?

This is a sideline thing for us, an experiment that has escaped the lab. To the extent that we put more resources against it, the focus will probably be on things that help it scale on its own -- easier set up, self-administration -- and resources that help the educational/professional development focus: demos, sample analysis and training tools for students and junior staff. But the user community is quite vocal, so they are sure to have a few ideas of their own.

Michael - Many thanks. He can be reached at mtamblyn @

No comments: