On Tuesday I was at SES London. I was also at SES for a day last year and I thought that this year was much better in terms of the quality of the presentations and speakers. Hats off to the organising team for making these improvements.
The keynote was by Jeff Hayzlett, the former CMO of Kodak. He was an amazing speaker. In fact he was so good that he could have been talking rubbish and I would have sat there amazed. If Jeff had said “what organisations need to do these days is throw money down a big hole”, I would have been thinking “have we got room in the office for a money pit?”.
I think the content was pretty good but, as I’ve said, I was so overawed that I’m not really in a position to judge.
Session 1: Basic Analytics
I thought this would be a waste of time for me (I’m arrogant enough to assume I don’t need basic Analytics training), but I’m ashamed to say I did come away with a few things to think about. John Marshall gave a good presentation from which I made the following notes:
- KPIs need to be understood by the C-level. This mostly means they need to have a $ sign in them somewhere.
- John really loves AOV as a metric. He prefers it to conversion rate. I disagree here. Having done conversion rate work on sites where they changed their prices midway through the experiment, I’d say that the link between AOV and conversion rate is too important, so it is a bad idea to look at either in isolation.
- He also says that bounce rate is a great metric. I’m not so sure on this. Think about it this way: What is the bounce rate for Wikipedia? Would it be good if that increased or decreased?
- The main point for me was how important segmentation is. This is something that I know how to do and that I use regularly, but I need to be much more proactive about pushing this at clients.
- Surveys are also something that I’m going to attach more importance to now.
Session 2: A Powerful New Choice in Search
I was disappointed by this session. Somehow I had got it into my head that it would have been more of an open forum, with Binghoo asking the room “How can we improve our advertising” and us agency guys yelling back “Give us an MCC!” Instead it was a presentation about how the transition will take place and how once it is done the world will be full of sunshine and rainbows for all who have to work with them.
- Yahoo and Bing use quite military language. I made a note of “join forces”, “alliance” and “retreat”. I might be reading too much into this, or it could be that they are on a serious war footing internally.
- They talked a lot about improving ROI for advertisers, which is a personal hate for me. What about profit?
- They estimate that the time saving caused by the merger will be around 20%. I need to get a job working solely on Panama/AdCenter accounts before the merger is complete so that after it is done I can take Friday off work.
- Jon Myers talked about some advertisers getting a “great increase in CPA”. I assumed this was a blooper.
- Another message that was used throughout the presentation was “efficiency”. Which makes it ironic that they’re reducing market efficiency by forcing advertisers to bid the same on Yahoo and Bing.
- One of the advantages of the search alliance is “bringing two businesses together to innovate faster”. I thought the general opinion was that smaller unites were better for fast innovation.
Session 3: Meaningful SEO Metrics
Pete Young gave an excellent presentation on SEO forecasting. Unfortunately (for you), the main point I got from this is that I will have to look at his presentation and take my time going through it. I think the rest of the audience felt the same because, given the interesting topic, there were very few questions for Pete afterwards.
Matthew Bailey’s presentation was on some of the methods his company uses when doing Analytics. Some of what he said was, in my opinion, fundamentally flawed. He talked about dragging in as many data points as possible and then looking for patterns. Unfortunately people are very good at spotting patterns where no pattern exists. He didn’t really have a good answer on how to prevent this happening, apart from that any conclusion you make should be immediately obvious from that data.
If I study the moon for long enough then I see a face. Excluding the hypothesis that the moon is a face and not a lump of rock I have to conclude that the face was made by aliens because I know it wasn’t made by us. Therefore aliens exist.
Going too deep in the data has dangers.
Session 4: The Ultimate Search Marketing Battle
I am a PPC guy. Nothing that was said here was ever going to change my mind, but the session was fun. I probably should have gone to the Analytics Deep Dive instead; that might have been more productive in terms of learning new and interesting things.
As I mentioned in the introduction, I was impressed with the quality of the conference compared to last year. The speakers were excellent and there was enough interesting content to make the day worthwhile for me.