In order of relevance:<p>* Doing e-commerce, funnel analysis is the metric by which your startup will live, or die. Mixpanel might help, but if your app has non-trivial business logic, implementing it by yourself might be better<p>* A good extended scope over that is presented in Startup Metrics for Pirates ( <a href="http://www.slideshare.net/dmc500hats/startup-metrics-for-pirates-long-version" rel="nofollow">http://www.slideshare.net/dmc500hats/startup-metrics-for-pir...</a> ), which takes the whole lifecycle of your customers into consideration<p>* A/B testing, and their related metrics are important to figure out, and validate learning about your customers. A/B is best implemented server-side, directly within your app's framework<p>Once you have these basics down, and even before you launch your MVP, you will need to come up with hypotheses about most columns in the business model canvas ( <a href="http://en.wikipedia.org/wiki/Business_Model_Canvas" rel="nofollow">http://en.wikipedia.org/wiki/Business_Model_Canvas</a> ); for consumer B2C apps, the most important will be your Channel hypothesis -AKA customer acquisition. Depending on your hypothesis, you will need to measure, and validate/invalidate it, for which you can use:<p>* Adwords has an extremely cool UI by itself, doesn't need anything extra<p>* For SEO metrics, check out seomoz, or competitive SEO tools. You will need to keep track of best performing keywords, general trends, and competition. It's wise to complementer your SEO strategy with promotional material (see patio's writeups about the minisites tactics), at which point the number of potential customer acquisition keywords can grow geometrically.<p>* There are a couple of tools for keeping track of Social/viral growth, but facebook's built-in FB insights is pretty good by itself<p>Your attention is zero-sum, and should be treated as such. Therefore, the most important factor in each metrics you should ask, is whether you can take action upon it (as opposed to being a vanity metrics [ <a href="http://www.fourhourworkweek.com/blog/2009/05/19/vanity-metrics-vs-actionable-metrics/" rel="nofollow">http://www.fourhourworkweek.com/blog/2009/05/19/vanity-metri...</a> ]), and if so, what kind (split testing? more cash into adwords? building new niche-specific site? etc).<p>Godspeed!