Bloggar mina anteckningar från Joshua Porters utmärkta och högst rekommenderade bok ”Designing for the social web” (Bokus, Adlibris). Här kapitel 3 och 4. Se också mina tidigare anteckningar; introduktionen och kapitel 1–2, kapitel 3–4, kapitel 5, kapitel 6–7.
Chapter 8: The Funnel Analysis
- 164. Having solid metrics for each major milestone in the progression of use is crucial. (Interested, First-time use, Regular use, Passionate use)
- 165. You want to get more fine-grained. Each site a different funnel.
- 165. Figure out your metrics. Eg: (1) Site visit, (2) Trial sign up, (3) Active use, (4) Paying use, (5) Recurring use.
- 166. Active eg 5 times/mo
- 167. · Establish a baseline – what is normal traffic – for all steps in funnel. · Find leaks. · Make changes. · Measure against baseline.
- 168. You can get even finer-grained – funnel steps per each page in sign-up process [for instance]
- 169. You can also test smaller parts of the app, such as sharing of objects.
- 171. Users act differently depending on where they come from. Digg users, eg, can disrupt your baseline.
- 171. Account for people jumping back and forth when setting up measure points for the funnel. Don’t expect them to linearly move through the site.
- 173. Perhaps use different landing pages for different traffic sources, such that you can measure with greater accuracy.
- 173. Perhaps measure organic, referral, and direct traffic separately.
- 174. Page views not that useful a metric.
- 175. Uniques, Repeats, Time on site, Pagerank, Sign-ups, Feed subscribers, Click-throughs to other sites
- 176. Social metrics: comments, items shared, # friends, # posts, # feedback msgs, # favorited, # bookmarks.
- 176. One or a few core metrics, defined by Activities.