PASS announced yesterday the end of the SQLRally brand. It was disappointing news for an idea that started with such promise. It’s a minimal impact decision as Rally had long been dead here in the US and I imagine the international users will be able to proceed using the SQLSaturday brand (and it’s tool set) without much pain (if not some disappointment). I don’t disagree with the decision, clearly it never grew into what we hoped. The rest of this is a short look back at what we tried and what I think went wrong only – some of it my fault, with the benefit of hindsight.
We hoped to achieve a lot with Rally. Lower cost but not free, serve the opposite side of the country from the Summit, challenge chapters to submit a proposal and win the bid to host it, an intermediate step in the speaker “farm club”, give our local leaders a chance to deal with a “Summit like” event on a smaller scale to prep them for the day they might serve on the Board.
Looking back, I still love what we did with the first event. Lots of discussion, published minutes of the committee meeting, and just the feel of the event. Calmer, more intimate than the Summit, and somehow better than a SQLSaturday, in part due to the multi day layout and in part to a very nice venue. We got pretty good reviews from the attendees, enough that we decided to have a second one.
In turns out that our success was a failure because we worked too hard to make the first one work and we made too many compromises. We got – in hindsight – a very, very, very good deal on the venue because we had to do a minimal commitment on rooms at the hosting hotel. Room commitments are a big deal because they add risk. If you sell the rooms all is good. If not, life is rough and you spend a lot of money. Risk mattered because we built Rally as a zero sum financial model – we wanted to just do a little better than break even. We also kept pushing up the price point because the Board wanted everything – coffee breaks, wifi, etc, so that we were a long way from the initial goal of $99 per person and funding it with a $25,000 guarantee from PASS. We also ended up with a much heavier (and costly) involvement from HQ than we expected (but perhaps needed). We ended up with a model at the end of year one that was heavier than anticipated, did ok financially, and had financial risks that wouldn’t really surface until the next event.
The second event succeeded, but signaled the end due to the weight of the financial risk of the room commitment and was initially thought to have lost money (when the dust settled it turned out to have done ok). That was the end of it in the US. I think the model could have been fixed (and I still think about giving it another go here in Orlando – SQLReally maybe?) but there wasn’t any push from the Board to figure it out and certainly no internal champion. This is where life on the Board is hard. Do you keep going with a lot of risk and basically no financial upside, or put a fork in it? Not a simple discussion or decision. It ended up being the international brand, useful because it was important to preserve the Summit as the marquis event and a Rally allowed charging for attendance (SQLSaturday is free, though paid pre-event classes are allowed). From what I saw that worked, but I haven’t been privy to the results in attendance or financials.
Stopping the Rally model in the US was not well received, something that I’m sure contributed to the delayed end of the format. There was never any effort that I saw to revise the model for US usage.
As an aside, one of the reasons I’m ok with the end of SQLRally is that it was an event without the kind of tools we have for SQLSaturday. Tools make a difference. Back then we hoped we might work toward a universal registration system, but I’ve seen no sign that that is being considered.
Lots of lessons, many of which seem to be lost to time already. I was trying to figure out what was the most important lesson, the one thing we should write down and keep in mind? I think it’s that having a champion matters. If you care about something you assign someone – look at BAC having an advocate vs no one person working on volunteers. Having a champion doesn’t guarantee success, but not having one, well, the chances of succeeding are a whole lot lower.
It’s hard to write this without thinking of how it compares to BAC. BAC has better financials, or at least better financial prospects given the price points, yet it has struggled to find an audience when Rally clearly had one from the start. It intrigues me that BAC is a standalone brand instead of a format, we could easily have done SQLRally BI, BAC, Performance Tuning, or whatever else. Having a canned format is incredibly handy. I wonder, perhaps cynically, if the difference is that BAC allows the Board to see themselves as visionaries breaking new ground (not the worst idea), or is it just about the financials? Or neither? Anyway, enough on that.
Regardless of the path, the result is that it didn’t work out. Taking it off the map clarifies things and that’s good. I think about everyone who helped build it and I think “we did good”, even if it didn’t grow to the heights we hoped.