We’ve just wrapped up the interviews and I wanted to jot down a few thoughts on what I’ve seen from using the process that we have today:
- The time requirement for the NomCom members seems reasonable
- We still have a lot of work on explaining the “why” of the tasks so that we get year over year consistency of execution. HQ has/will continue to pay a key role in that.
- I think the process does a fair job of helping the NomCom assess candidates. I rate it fair because I wish for better (and we’ll look more), but it’s always analog.
- I’d like to see the process revised to include a mock app review and a mock candidate interview. Particularly the latter, we found as we went that we’d hear someone ask a question and go “that’s a good question!”, and sometimes we’d hear a good question badly framed. One mock would smooth a lot of that out, especially with an interesting “candidate”.
- The current application plus an hour interview is enough to assess them. Not sure longer/more would be worth doing.
There was one part that surprised me, and that was how hard it was for me – who knows better! – to remember that I’m not hiring candidates, I’m screening them (you the voter get to do the hiring). What’s the difference? Screening means we pass on qualified candidates, even if in some cases we heard an answer we disagreed with it on direction or style or substance. The difference is subtle, but it’s there. The process we have mitigates that problem nicely by scoring candidates across different categories, not just a thumbs up/down. I’ll add that if we have a weak point its that we don’t have a lot of good data points on who will work out to be a good Board member. We should work on that, not just for vetting, but for growing future candidates.
Thinking about the rest of it, I am concerned about the drop in total candidates this year. Why so much lower? One theory I’ve heard is that with one current/three former Board members on the NomCom potential candidates were intimidated. I hope that isn’t the case. Not enough marketing? If you have thoughts on any part of that I’d love to hear from you offline/confidentially.
The NomCom isn’t quite done. We still have to recommend the slate, and then revisit the entire process to see if more changes are needed.
I’ll close by thanking the candidates for this year for stepping up and taking the challenge. Win or lose, your participation means a lot to the organization and its members.
- Reg count at 479. Looking very good for hitting 500!
- First sponsor sent out a geo-targeted message for us, another one working on it, more to go. We wrote a ‘suggested’ message to decrease the friction. First vendor even sent back open rates on the email!
- Karla/Shawn dropped in at ONETUG meeting last night to hand out flyers and coordinate ONETUG sharing the PASS table
- Just scheduled sponsor mention tweets for next two weeks plus day of event (need to do same for speakers/sessions next week)
- Signed up for trial of Hootsuite Pro to do the scheduling, works reasonably well
- Our half day student seminar seems to be resonating. Our initial target was 30, we hit that and asked for a bigger room. Today we hit that cap of 60, so we’re moving to a room that seats 100! We’ll see how many how, but still, this is a much bigger response than we’ve seen with our previous efforts to engage them.
- We’ve tentatively decided to include them in the main event lunch, but not to register them or give them event raffle tickets. That keeps the sponsor list fairly targeted. Students will be welcome to attend sessions or visit sponsors after 1 pm.
- One of the things we talked about this past week was a photo/video plan, need to make sure that covers the student seminar (is now in a different building)
- We need a couple raffle prizes for students (most will leave at lunch), an eval, what else?
- Wrote the ‘you haven’t paid for lunch yet’ message that goes out next week, still a lot more messages to write
- Starting to get news on raffle prizes, will try to add that to messages too
- Networking landed us a local gold sponsor, working on a second one. Pays to maintain relationships!
My goal since Part 2 has been 500 registered and 350 attending, it’s exciting to be close! But if our projection holds, we should still add another 100 before the event. The real question is what our drop rate will be? I’m expecting it to be higher than last year due to the longer sign up cycle, but that’s only a guess. We’ve set the waitlist threshold at 600, can we do it? We shall see.
Ah, an interesting week for registrations. We’ve jumped to 426 registered, making my goal of 500 doable. The jump is in part due to timing, in part due to only having mailed to the “already registered” list only last week (a goof on my part). This week we’ll have two messages going on, one to the registered list to work on bring a friend, the other to the not yet attending list to continue to plug the event. Besides that, here’s a few more things going on:
- Emailed a reminder to the Florida chapter leaders asking for one more plug at their September meetings
- Drafted an email that ONETUG has agreed to blast to their list (to go out after their Sep meeting). This is huge, it will be a ‘one topic’ email just about SQLSaturday.
- Finished scheduling the last of my canned auto-tweets through this Friday, then we’ll change the mix (interestingly, I see relatively few registrants opting to post their attendance to Twitter)
- I’m trying to finish the rest of the email messages this week, I don’t want to risk life/work hitting me just when I need to be on my game the most in the last couple weeks
- Discussing if we can have someone staff Twitter all day at the event and what we can do with it if we can
- Also discussing our photo strategy, I’d really like to have those for next year
- Asked Kendal to also chart the number of emails sent out to reg/non-reg lists by week
- Next week I’ll send the email to the host venue (college) advisory council as part of the final push
- I still need the 2007 & 2008 registration numbers
- I need to think more about what I can do at the event to help us market better next year. Focus group, surveys, contact gathering, photos, what else?
At this point the strategy is simple – keep pushing!
Here is the current chart:
Here’s something I haven’t talked about much yet. Over the past five years we’ve seen an incremental decline in the number of registrations, from a high of 384 in 2009 to 344 last year. Not a huge variance. Enough to worry? No, yet at the same time I’d always rather see it static or growing. It’s hard to attribute the changes year over year to any particular cause, but my guess based on what we see this year is not enough marketing messages sent in the last two years. By “not enough” we might be talking a difference of two or three, but it takes time for the message to get through the ‘clutter’ of work and life in the inbox. Kendal created a big win by starting the process of visualizing what little we know about our marketing efforts and it’s easy to see this was a big miss when we built SQLSaturday (a combination of not enough time and not enough vision). Having charts like this built-in would make it easier (and more likely) that whoever owns marketing is competing to match and beat the previous years number.
I wrote Titles Matter – Part 2 based on lessons learned the hard way not long ago. Titles matter because they help the rest of the team understand where you fit in. Titles are the ultimate elevator pitch – this is what I do. I was pleased to see one comment where someone changed their view about titles based on what I wrote. Not often that happens (or that you know about it at least) and that made my day.
Last week we were at 332, this morning at about the same time we’re at 356. Not bad for a week that included Labor Day. Chart makes it look a little worse than it is, it shows the actual against the goal for the next report period.
Here’s the last 4 weeks of trend data. If you look at the delta between week 3 and week 0, we’ve always added at least a hundred and twenty. I’ll still – still! – confident that we’ll hit our goal of 500 registrations. The next email will go out tomorrow.
Registration is 332 as of 8 am on August 26. We added a whopping 40 registrations last week – a very good week! Our standard Tuesday email will go out about 10 am this morning with topics on MVP’s in attendance and lunch. On Friday we’ll have our first “special” email go out to those who have registered, giving them first crack at a seat at our Oct 15 meeting featuring Mark Souza. Looking ahead, next Monday is a US holiday so we’ll push our weekly email to Wednesday morning. First day of the week anything that looks ‘noisy’ gets ignored or deleted all too often.
The oPASS meeting is this week and that will be a chance for Kendal to plug the event, good to hear it from someone different.
Going into September, oPASS & MagicPASS won’t have their normal meeting due to SQLSaturday, so they’ll send out at least one email to their lists. That will be a nice change from the quasi-official email we send out from SQLSaturday. Different tone, different font, different sender – any one of those might drive them to open the email. I’ll be sending a reminder to the Fla group leaders next week asking for a mention at their September meetings, and ONETUG meets again on September 11th, we’ll have at least one person there to talk about the event and hand out flyers.
I’m still – still – trying to figure out a good way to use what I have to drive attendance or referrals. Here’s my list:
- Tshirts. Sized items and we like to order early, so not the easiest thing to work with
- SQLSaturday Orlando mini towels. We’re ordering a bunch and not sized, so maybe
- Lunch/Dinner with an MVP (takes coordination of schedule/interest, not easy)
- Ticket to a seminar (has a hard cost)
I think what I want to do is send another “special” email next Friday that targets those registered to ask for referrals. There’s no support on the site, so I think we’ll use a survey (your name, their name, your email, their email) and try to incent it somehow. Still thinking.
I paused the LinkedIn campaign this morning. The ad targeted by job title generated 5000 impressions and 11 clicks for a total of $14.63 spent. Overall we’ve spent $65.46 across the the two campaigns. I think we’ll turn it back on the week of the event and see if a ‘just in time’ message delivery helps.
I just checked in again on our LinkedIn campaign, had a couple days with good impression counts. We can still only tie 1 registration to all of the LinkedIn effort so far. Keep going, or ? I think I’m going to see if I can change the URL to a tinyurl, don’t know if that will pass the ad approval process or not.
I wanted to take a look at the effectiveness of each email, but because of a data problem graph is really skewed. There is the idea of a “bulk loaded” list so you can work with all your data in one place, but there is a bug in the registration proc (which I think I wrote, way back when) that doesn’t update the registration date. Kendal is compensating in the weekly graph, but the data on a daily basis isn’t great. I see the expected “Tuesday spike” most weeks except the week of Aug 5th. It could be we had a bunch of people register from the bulk list, but I think more likely the subject line (SQLSaturday Orlando – Sign Up For Our Dimensional Modeling Seminar on September 26th) didn’t drive enough people to open the email.
It’s a nice steady trickle of registrations, we’re at 321 as of today.
Got this in morning email from Khan Academy. Not sure if I agree that we all can learn anything, but I love the video anyway because the best lesson to learn is that learning takes time, effort, and stumbles. I watch my own kids struggle with this, and think how long it took me to understand it. Worth 90 seconds to watch it.
I happened to read When A Flight Vanishes From The Sky, Amateur Trackers Know It Instantly on http://fivethirtyeight.com/ and that took me to http://www.flightradar24.com/. It varies from real time to a 5 minute delay in the US, showing all the planes that are in flight with a transponder active. They use FAA data, but also lots of local users with a $40 device plugged into a USB port in return for premium access to the service. It’s interesting, really it is, to see all the planes in flight on an average afternoon. You can see the flight path, speed/altitude, plane type. Great idea, and it has me thinking about the data. Not complicated data, it’s the crowd sourcing that makes it work. What could I apply that to?
And last, I also read a post about Getting Images for Every Kilometer of the UK, also at 538. Very similar idea, except this uses photos.