Note: This is a good, bad, and ugly post. We took some chances, got some things right, got more than a few wrong. In the end we launched SQLRally, had a good event in 2011, and have one coming to Dallas in 2012.  

Way back in July 2010 as we started on the SQLRally journey towards Orlando in May 2011 we also tried to put some thought into how, when, and if we would proceed with picking the site for 2012. What we came up with:

• When: As soon as possible after the November 2010 Summit
• How: An Olympic style bake off between the top three cities/chapters in specified time zones
• If: Regardless. We simply did not have time to wait for the 2011 results to  begin planning and executing on 2012.

We started work on the “process” in October and then in January this year we were able to finally start the selection of the 2012 site. The goal was to try to have the chapters showcase their strengths both to the Board and to the community, and we tried to do that via an application. We ended up giving chapters a very short (unfairly short) deadline, and then we had at least one ask to submit late because of the short notice. We originally had four cities participating and then one decided to remove their team from the race, leaving us with three.

The next step was to vet the applications,and all (Dallas,Denver, Nashville) had while I consider to be good SQL communities, each having a number of volunteers known to PASS and with a track record of doing good things. I wanted to make those applications public, because it would provide the details of their team and their vision, including ways they would support ‘after hours’ events. HQ felt that was unfair because some teams had spent more effort than others on their application, and so we ended up giving more time to everyone to polish their application. It wasn’t a good decision, or at least not a great one, because it felt like the rules were being changed on the fly and we didn’t do a good job at all of explaining the logic.

That was also the point where I understood better (not that I didn’t see some of this in the forecast) that it’s just not possible to stick to much of a process when you’re doing something for the first time. (Note: Remember that for Orlando what happened is the Board asked me to prototype an event based on having a cash budget of $25,000. I used Orlando as a convenience, and in the end the Board adopted the plan in order to move forward quickly.) We had a vision of the bake off, but the things we didn’t know would soon cause pain. For example, some teams saw it as a competition to win (push hard, don’t share ideas), others saw it as an informal contest (enter, share, enjoy). Neither is wrong, but we didn’t do a good job of setting the expectation of what it should be (or hoped it would be at least).

The concept was also that the local teams would suggest some venues, and then HQ would leverage their experience in event venues to qualify the venues and get some initial pricing. This idea sounded good, but it ended up feeling like both shared and contested ownership of a task. Teams expected to be in the loop during the process, HQ wanted to compile all the results and send back in one batch. HQ sent back different results than the teams expected, and we had to dig into that as well.

Why different results? Dealing with venues is hard. They speak their own language. For example, I know from seminars I’ve done here in Orlando that when I discuss lunch costs it’s typical for them to say something like “that will be $23.75 plus plus”. If you’re not listening it’s easy to dismiss plus plus as blah blah, but what it means is $23.75 + 20-30% service charge + sales tax, which can be a surprise when get the bill.

Venues make their money on food & beverages (F&B) and rooms. For example, for SQL Rally 2011 the meeting space was $12,000 a day, but we could get the space for free with a minimum daily spend of $12,000 in F&B. Not hard to get to when you’re paying $35++ for lunch per person!

Hotels typically require a room  block (a room block is a number of rooms that you guarantee to use, or to pay the equivalent cost) because most large groups want meeting space AND rooms. The hotel doesn’t want to leave rooms empty because one group took the meeting space and another group needs space plus rooms. We were very lucky in Orlando that the contract we negotiated did not require a room block. For Orlando we had a hold on 50 rooms (Orlando is a great market for hotels, we expected most attendees to stay in other places) but we didn’t have to use them, we could just release them back. Normally if you take rooms you get some allowance (“attrition”) of 10-15% that you’re allowed to not use. You can block 250, but as long as you hit 85-90% the rest aren’t charged to you.

By the end of the journey I learned a couple new lessons. One is that going direct to the hotel tends to yield different results than going through the local hotel/convention bureau. In general going direct seems to offer more flexibility. The other is that blocking 200 rooms isn’t the same as “200 room nights”. 200 rooms means 200 rooms per night during the main days of the event. For a SQLRally that works out to anywhere from 250-600 room “nights”.

Back to the story. We were trying to build our 2012 site based on the 2011 budget, and that budget did not include a room block. From the start that presented serious challenges. We were never able to find a satisfactory location in Nashville that didn’t require us to commit to rooms, and the only location we found in Denver was the week prior to Memorial Day. We worked hard on this, but over the months of effort I realized that picking Orlando as the prototype had shielded us from some hard lessons.

Over the course of the selection process I talked to the Board several times about amending the budget by adding an estimated $40,000 to cover our liability for rooms. The first two times it was a solid “no”. The intent was to build a low cost/low risk/low profit model, and increasing the risk meant changing the formula. We went back and wrestled with hotels some more, and had the same conversation again at the Rally with the same result. It was a frustrating decision, but far from a bad one. My job was to make it work and ask for things if I needed them, their job is to keep an eye on the big picture and not let things get out of control.

In March/early April we were at the point of being ready to try to do a vote, but the Board asked for a delay, wanting to wait to see what registrations would look like after the April price bump. This was the first of a few times when we veered into chaos. We had decided last year to go forward without regard to 2011 results, now we were trying to reverse that without reversing it. That date passed, and then I was asked to delay again until after the Rally so that the Board could assess the ‘look and feel’ and think about room blocks in context.

We discussed the room block issue again during our Board meeting there, and with the answer still “no” I had several on site meetings with the team leaders.  At that point our site selection was far from done, and we decided to extend the search for hotel/other for a few more weeks, with the condition that if by the cut off date they didn’t have a location, they were out of the picture. We planned on starting the voting on June 15th unless somehow no one had a viable option.

I did some 2011 wrap up and headed off on vacation. Mild chaos while I was gone, by the time I got back into the conversation we were down to this:

• Dallas, two locations that didn’t require a room block, budget still uncertain on one but doable even in the worst case by doing offsets in the budget for other items
• Denver. One location that didn’t require a room block, but it was the week prior to Memorial Day.
• Nashville. No really viable location

We had a Board meeting last week and part of the agenda was to review the site selection. Nashville was, unfortunately, out of the running. The question was, do we consider Denver an option given the date conflict? We wanted to do a vote, but the week before Memorial Day was a big risk, at the least we’d have to forecast lower attendance and even then, no guarantees. Eliminating Denver meant Dallas would win by default. Ultimately with time running out as we get into Summit season we decided that was the best we could do (and that is not meant to insult Denver – we wanted the community vote).

I did a call with the leaders and Bill Graziano last Thursday to explain the results. Denver and Nashville were incredibly gracious about the decision, all the more meaningful to me given the long, tedious, and bumpy path to a decision that went against them.

We’ve still got work to do in Dallas. HQ will be doing a deeper assessment of the two sites and sending that back to the Board along with a detailed budget. Our hope is that by July 15th or so (please don’t hold me to that!) we’ll have the contract done and can set up the event site.

As you read all of this it’s probably easy to see villains, or to see us as bumblers, but my view is that neither is true. PASS is often viewed as a slowly moving organization, and one that takes few risks. For the SQLRally effort neither has been true. We had an idea, asked for a proposal, and without endless debate said “do it”, and we also made the decision to launch into execution of 2012 with no idea of how 2011 would work. An aggressive strategy, one that at a high level worked out as a big win.

After taking some time off to recharge a bit more I’ll be working on the 2013 site selection. As I write this I’m not sure how it will change. The venue pain would go away if we budgeted for a room block, but that means going through the budget to see how much it would raise registration and adjusting other costs, and I think it also will require another look at ways to do it at lower cost, even if it means capping attendance at 350-400 (but which might enable us to do 2 a year in the US).

I’d like to preserve the idea of a community vote on the site, but we’ve got to do more to make sure that we are voting on good and financially responsible sites, and that may be at odds with the way we tried to do it this year.. We’ve got to define the scope better too, as a minimum I believe we will define the date we want (no spread across Apr-Jun) and I think we’ll require it to be at a hotel or convention center with attached hotel because we believe the on site networking is a very big piece of what we want to accomplish.

To the teams in Nashville and Denver, thank you for participating and doing all we asked of you. You’ve earned some karma with me for sure. To Team Dallas, congratulations on winning, even if it wasn’t quite the way you wanted, and now that you’ve won, I challenge you to do it better than we did in Orlando!