I have previously posted some of the “raw data” on Six Weeks of Windows Azure but I wanted to also share some thoughts about “how it went” and “how we might do things differently” in the form of a mini retrospective.
What went well?
Achieved a lot for very little money
Our budget was £0. Given that, we achieved a heck of a lot
Variety and depth
30 sessions and 2 streams (technical and commercial) delivered by a combination of Microsoft, Partners and Early Adopters allowed us to get through a lot of material. With that comes:
- complexity – managing so many different people/organisations
- inconsistency – its easy to end up with overlap or indeed gaps
But overall we are very happy with how it worked out. Feedback has been universally positive – 95% would recommend to a colleague or another company.
The weekly one hour surgery allowed companies to ask questions and get answers. We felt these went extremely well given we had not tried this format before.
The week 4 labs were an untried format for us. 12 companies came in over the 5 days (max of 3 companies on each day). The use of the day varied enormously from reviewing architecture through to migrating code and databases to the cloud. All the companies found the lab time extremely useful.
We easily got to over 500 individuals registered, primarily through publicising through blogs, twitter, email.
UK Partner/ISV/SaaS participation
Our primary audience was UK partners although we didn’t turn away any organisation. We had over 90% partner sign up and vast majority were UK based.
What could be improved?
Reaching none .NET companies
90% of the companies were doing .NET. Which is fantastic … but Windows Azure is great for OSS as well. We would have loved to have seen more none .NET companies joining us.
Reaching commercial decision makers
Vast majority of the attendees were technical – yet we had a great set of commercial sessions. Our advertising just didn’t reach the commercial folks.
Live Meeting Audio
Was at times very frustrating – mainly around audio dropping/failing to connect for speakers. We would consider Lync or even 3rd party solutions for a future delivery.
Product team involvement
We tried but didn’t get product team involvement. However I think it would be much easier to do so in the future.
More Week one content
Some technologies (e.g. SQL Azure Federations, Windows Azure Connect or HPC on Windows Azure) were not covered until very late in the 6 weeks. I think we could do a better job of demonstrating the rich breadth of Windows Azure in Week one whilst we also have the most attendees.
What still puzzles us?
No shows in Week One
We had 190 people join us in Week One out of over 500 registrations. Many of these “missing 300” never went on to join us live but presumably some did watch the recordings. However it puzzles us why they never showed.
The right length for this activity and the right level of commitment to ask of companies
We wanted to take companies on a journey with detail and depth, hopefully leading to clear plans for when and how they would adopt Windows Azure. But… by week 4 we had significant drop off in numbers despite extremely good feedback on the first three weeks of sessions.
Potential Tweaked Version
Based on the above we would likely tweak a future delivery as follows:
Maximise the Week One experience – target hundreds of individuals and organisations
- We would make sure that in week one we covered the entire breadth of Windows Azure at some level, whilst as many folks as possible are “still with us”.
- This could be done as a one day online conference or potentially a large in-person event.
- This would be delivered as a “stand alone” activity which in itself was extremely useful. However it would also signpost a journey for the most interested/committed organisations to follow over coming months.
- All commercial content would be in Week one.
Shorten the journey with a “Month of Windows Azure” – target tens of organisations
The kick start event would signpost to a “month of Windows Azure” no closer than 2 weeks away. 4 weeks vs 6 is a good compromise
Simplify the commitment from two days to one day
We spread our 5 sessions per week over two days. Instead we would concentrate everything into one afternoon with signposting to further detail on each topic. We would retain surgeries.
Labs – targeting handfuls of organisation
The most committed could then subsequently join us in labs to unblock/validate. Rather than embedding these into the six weeks we would scheduled them for much later
A potential timeline would therefore be:
- Month 1: One day conference – left to right of Windows Azure covering technical and commercial
- Month 2: Four weeks of one afternoon webinars and surgeries
- Month 4: Labs
We did run some of these tweaks by our companies who made it to week 6.