Searchquant [Mostly] SEM Predictions for 2007

I haven't seen any '07 SEM predictions on the blogs I frequent, so here are mine:

10. Growth in number of words per query not only stalls, but goes negative as SE's get better at personalization, geo and context awareness.
9. Keyword management systems vendors realize their systems can't optimize in opaque markets. Their advertisers realize this too, leading to huge churn and massive consolidation.
8. At least one startup gets funding to bridge the online/offline revenue measurement gap using novel approaches. Think Quick Response codes, Project Apollo, RFID readers in cell phones, stuff like that.
7. Y! Panama helps good advertisers do better, and poor advertisers to do worse.
6. Bay Area real estate bubble bursts, offset at the high end by Googlers buying everything in the $1.5M-15M range.
5. Ad copy and multivariate testing stampede starts in earnest, as advertisers fight to raise conversion rates and avoid extinction.
4. MSN doesn't get above 11-12% search market share.
3. George Reyes' fingerprints found on Quality Score knob at the 'Plex.
2. Domain traffic comes to be viewed as the lowest-quality component of Y! & G's networks.
1. Someone proposes via AdWords.