An e-retailer, for instance, may be more interested in sales by product category or at different price points, though a business-to-business (B2B) service company might want to look at which sources produce the most qualified prospects. The trick is to select five to ten relevant metrics for your business.
If something isn't measured, it can't be evaluated. If it can't be evaluated, it isn't considered important.
You can establish your own KPIs. Then you can combine them with other information about how your various marketing efforts contribute to sales and leads, to your bottom line, and to your return on investment (ROI). Armed with this information, you'll be in a position to make strategic business decisions about your marketing mix, no matter what size your company.Enter at least one KPI for each business goal on your Social Media Marketing Plan. Some business goals share the same KPI. Schedule a review of the comparative metrics on your Social Media Activity Calendar at least once per month — or more often if you're starting a new endeavor; you're running a brief, time-constrained effort; or you handle a large volume of traffic.
Overcoming measurement challenges
Measuring success among forms of social media, let alone between social media and any other forms of marketing, is a challenge. You're likely to find yourself comparing apples not only to oranges but to mangoes, pineapples, kiwis, pears, and bananas too. In the end, you have to settle for a fruit salad or smoothie.Install the same statistical software, whether it's Google Analytics or another package, on all your sites. Your sites may not have identical goals (for instance, users may not be able to purchase from your LinkedIn profile or request a quote from your wiki), but using the same software will ensure that metrics are consistently defined. In fact, the availability of compatible analytics packages may influence your selection of a host, development platform, or even web developer.
Using A/B testing
You may want to apply A/B testing (comparing a control sample against other samples in which only one element has changed) to your forays into social media. Just as you might use A/B testing to evaluate landing pages or emails, you can also compare results between two versions of a blog post or compare performance of two different headlines for an update on a social media service, while keeping all other content identical.If you're comparing performance (click-throughs to your site) of content placed in different locations — for example, on several different social bookmarking sites or social news services — use identical content for greater accuracy.
Don't rely on absolute measurements from any online source. Take marketing metrics with a shaker full of salt; look more at the trends than at the exact numbers. Be forewarned, though, that the temptation to treat numbers as sacrosanct is hard to resist.
To no one's surprise, an entire business has grown up around web metrics. If you have a statistical bent, join or follow the discussions on the resource sites listed in the table.Online Metrics Resources
| Site Name | URL | What It Offers |
| eMetrics | www.emetrics.org | Events and conferences on marketing optimization |
| HubSpot | http://blog.hubspot.com/marketing/how-to-run-an-ab-test-ht | A/B testing how‐to FAQs article |
| Digital Analytics Association | www.digitalanalyticsassociation.org | Professional association for analytics practitioners |
| Web Analytics Demystified Blog | http://blog.webanalyticsdemystified.com | Digital measurement techniques |
| WebProNews | www.webpronews.com | Breaking news blog for web professionals, including analytics topics |
| Webtrends | http://webtrends.com/resources/overview | Resources for analytics and other marketing topics |