Game Theory, the Prisoner's Dilemma and... Winery Membership Organizations, Part 2: Measuring the Value of Membership

Last month, I wrote a blog piece about how game theory, and specifically the classic example of the prisoner's dilemma, relates to participation in winery membership organizations.  If you haven't read part 1, this post will make more sense if you do so now.

OK, welcome back.

In this piece I'm going to tackle the first of two questions part 1 left unanswered: how to measure the value created by being a member of a winery organization, and whether it's worth the (often considerable) expense.  I'll tackle the second question (how game theory suggests an organization or a community respond to wineries who opt out of the communal marketing) in part 3.

Membership in a winery organization is typically not cheap.  The membership page on the Paso Robles Wine Country Alliance (PRWCA, for short) Web site lists membership prices from $675 (for wineries producing fewer than 250 cases) to $17,500 (for wineries producing over 40,000 cases) with a range of gradations in between.  In many cases, this membership is the single largest annual marketing expense for a winery member.  Is it worth it?

There are many ways to look at the benefits of membership, some of which will be applicable to some brands and not others, and some of which are more easily measurable than others.  I'll focus on a few principal tangible benefits in this piece, but want briefly to address some of the benefits of membership that I think are valuable but hard to quantify.  They include:

• Participating in the branding and market awareness of your region.  This is the central goal of most marketing organizations, and yet it is both hard to quantify the value of success and impossible to limit this value-add to members.  If a campaign is successful, each bottle of wine you produce is a little easier to sell, as there is a larger pool of people out there potentially interested in purchasing your product.  But do they discriminate between wineries who are members of the organization and those who are not?  Unlikely.  Yet most wineries and growers would list this as the single most important goal of a membership organization.
• Having representation at the many intersections of the wine community and state and local government.  Whether it's building, tasting room, or event permits; water rights; labor supply; or regulation, one role of a membership organization is to represent its members in front of local government.  Elected representatives are much more likely to hear the aggregated voice of an organization that represents hundreds of businesses and thousands of voters than they are to listen to any one individual, and it takes an organization to have the resources to attend meetings and stay informed.  A winery may go years without having a regulatory issue threaten its business, but when one does, its impact can be enormous, and having a say in the outcome invaluable.
• Being a part of a community of ideas. There is significant value in rubbing shoulders with your peers as they navigate the same challenges that you face.  You share (and borrow) ideas both directly from other members and through the educational events organized by the membership organization, and it only takes a few good ideas to make a measurable difference in your bottom line.  The PRWCA has monthly meetings for tasting room managers as well as regular seminars for growers and winemakers.  Could you replicate these by paying for the ones that you want, or by spending lots of time with your neighbors?  To a significant extent, yes.  But that process is a lot easier when you're a part of a group with the mission to facilitate this exchange.

All of the above are valuable, but I don't actually think you need to get value out of the intangibles in order to justify the cost of membership.  In fact, let's just look at one benefit: the value of the additional people that membership brings into a tasting room. 

Wineries should know what an average customer who walks through their doors is worth to them. This will vary from winery to winery and depend on the winery’s average sale (AS), the percentage of walk-in customers who sign up for a wine club (CP), the average purchase per year of a wine club member (AP), the average duration in years of a wine club membership (LM) and the winery's profit margin (PM). The calculation goes:

Average Value = (AS + (CP * AP * LM)) * PM

For us, that figure is about $130. We’re one of the larger wineries in our area, so we pay a relatively high total in dues, but even at our size, we end up ahead if membership brings at least another 125 customers per year into our tasting room. That's 11 customers per month.

Once I went to the trouble of calculating the value of each additional customer visit, the calculus of whether or not membership paid for itself got a lot simpler.  Do we get 11 extra customer per month because of our membership?  I'd say yes, many times over.  Here's how: 

Bringing customers into the area. 
The ongoing marketing and advertising campaign, the organization-sponsored events, and the media outreach that the group does are all undertaken with the goal of bringing people into the area each year.  Once the people have booked their trip, they may or may not further consult the PRWCA for guidance on where to go.  So, the visits that these customers make in the area will be distributed among the various wineries, members and not.

How many extra people does the work of a group like the PRWCA bring into the area?  It's a tricky question, as customers typically have multiple reasons that they make any particular buying or travel decision.  But I would suspect that it's at a minimum in the tens of thousands per year (the festivals alone attract over 10,000) and likely in six figures.  (For why I think it's that high, see the "But does it work?" section below.)

If we assume that each wine tourist would visit an average of 5 tasting rooms per Paso Robles trip, even the minimum figure indicates 50,000 tasting room visits created by the PRWCA's outreach.  Divide this up among 200 local tasting rooms, and you get a figure of 250 visits per year per winery.  That's already double the number that we would need to justify the membership outlay.  If, as I suspect, the visit total is significantly higher than 10,000, the extra is just gravy.

Now, does a winery need to be a member to reap these benefits?  No.  But it's not just a community-minded thing to do.  As as the prisoner's dilemma example in part 1 showed, the individual wineries in the region benefit more from pooling their resources in a combined marketing campaign than they would from spending their same money individually.  

Directing the customers that they bring (and others) into the tasting rooms of members.
There is also a direct result on traffic due to a member's inclusion on the map (below, or available in PDF here) prepared by the PRWCA, printed by the hundreds of thousands each year, and also distributed electronically to the 50,000 monthly visitors to the pasowine.com site.  That's a lot of potential customers who see you if you're a member, and don't see you if you aren't. How many of our customers have to find us through one of the PRWCA touring tools to make it worth the investment? Less than 1%.

Remember that we need 11 extra customers per month to pay for our membership.  We average 2500 visitors per month, so 11 is four tenths of one percent of our traffic. It seems impossible that being on the map and on the Web site where hundreds of thousands of customers are starting their research isn't going to bring us that tiny marginal increase in our customer base; one or two groups per week is enough.  For a smaller winery -- say one at the 3,000 case level who's paying about $3,000 in annual dues -- 24 additional customers per year does the trick.  That's one couple a month.

But does it work?
I can almost hear the reader question hovering in the background: "but how do you measure if the outreach really does drive traffic?"  I'm happy you asked.

Last year, the PRWCA started an online campaign, which ran from August through October and included targeted advertising on such high-traffic sites as Pandora, Centro, Snooth, Eater, Facebook, Twitter, Youtube and Google.  These various ad buys (coordinated by an expert digital advertising agency) produced nearly 40,000,000 impressions, and big increases in the traffic on pasowine.com and the PRWCA's various social media sites.  I'd been aware that our tasting room traffic had also spiked last fall, but hadn't particularly tied the two things together.  As a baseline for comparison, between January 2013 and July 2013 our tasting room traffic was up 2.7%.  During the August-October PRWCA ad campaign, our traffic was up 13.2%.  After the campaign ended, it was down 0.6%.  Now, this is not conclusive.  There could have been a number of other reasons that our traffic was up then (including that we had a run of good press, and that the tasting room was doing, I thought, a particularly good job).  Or it could have been a statistical blip.  But, it encouraged me to pay attention to the difference in 2014 between our traffic when the PRWCA was advertising, and when it wasn't.  So far this year, the results have been similarly suggestive. 

In 2014, the PRWCA has made two advertising pushes (around March's Vintage Paso/Zinfandel event and May's Paso Robles Wine Festival) that were separated by so little time that it seems likely that their impact would have run together.  Looking at the period before that advertising started (1/1/14-2/17/14) our traffic was down 2%.  Between the beginning of the first campaign and the end of the second (2/18/14-5/26/14) our traffic was up 8.2%.  And since the campaign ended (granted, it's only been five weeks) our traffic has been down 3%.  That's starting to be a useful number of data points.¹ 

The roughly 10% increase in our traffic that we've seen in the periods where the PRWCA has been advertising amounts to an extra 200 customers per month. That's 18 times the number of extra monthly customers we need to break even on membership.

Does a winery who is not a member get many of these benefits in traffic?  Of course.  That's the free-rider challenge.  But if they lose even a tenth of that extra traffic because they're not on the map, or not on the Web site where the customers are doing their research, they're coming out behind. 

Other Direct Benefits: Trade and Media Outreach
Of course, wineries do not live by tasting room traffic alone, and a region's or winery's reputation can be built faster with help from the trade and media.  Outreach to these groups is another function of a membership organization.  Let's look at media outreach first.  I'm always skeptical of documents that point to "ad equivalency value" of editorial pieces, but anyone who has investigated the cost of an advertising campaign knows it's staggering: placing a one-page ad in any of the big food and wine magazines is more expensive than annual membership in the PRWCA for our largest wineries.  And the value of advertising is in repetition and duration, which puts it out of reach for any but the largest wineries.  This leaves smaller wineries to work to get editorial coverage, an effort made much easier with exposure to the dozens of writers that a group like the PRWCA brings into town each year.  Is this something that every winery will get equally?  Of course not.  The writers don't write stories spoon-fed to them, and most come to town with agendas of their own.  But it only takes one success to reach an ad equivalency of equal to or greater value than the membership cost.

There are similar equations to calculate in the trade outreach that a marketing group does, in trade education, in buyers tours -- where the trade is brought to the region -- and in road shows, where the region brings itself to major markets and invites trade to see what's new.  What is the value of a by-the-glass placement at a high-profile restaurant in Indianapolis (one concrete result for Tablas Creek from last year's buyer's tour)?  Or an agreement with a new distributor to represent your wines in a market you're looking to break into?  Or a feature in the newsletter of a retailer?  These, too, are occasional but powerful additions to a winery's business.  Of course, these benefits aren't of much value if a winery is only selling out of their cellar door, but if that's the case, they probably get all the benefit they need from the additions to their tasting room traffic.  In fact, it's the added work that an organization has to do to reach the trade, and the disproportionate impact that has on larger wineries, that is the best justification in my opinion for the sliding scale of winery dues.

Both trade and media outreach are benefits that are only available to members, as non-members aren't included in visit itineraries or group events.

Conclusion
In a comment in a recent piece on the Hoot N Annie blog, Gary Eberle says that membership in the PRWCA is "the cheapest investment a winery can make in direct to consumer sales".  And I agree... whatever your size, if you have a tasting room, being a member of the organization pays for itself many times over. 

When I moved out to Paso Robles, I thought that membership in the PRWCA was something of a civic duty: an investment a winery did for the long-term growth of the area.  I still think that's true.  But I think it's equally true that membership brings direct returns to the winery many times greater than the investment.  Good for the region, and for our bottom line?  Why yes, please sign me up.

How do we get back the handful of wineries that drop out each year?  And how do we minimize the number that we lose?  Stick around for part 3.

Footnotes
¹For the stats geeks out there, I had my wife dust off her graduate statistics work and run a t-test on the changes in our weekly traffic data since the beginning of last year.  Of the 76 weekly data points, there were 30 weeks where the PRWCA's ad campaign was going on and 46 weeks where it wasn't.  With that few data points, I would have been surprised had the data come out as statistically significant. It didn't, but the p-value (the measure of significance) was .189, indicating that there was a less than 20% probability that the difference between the result sets was due to chance.