Notes and Thoughts on Grocery 2.0

In my opinion, understanding web analytics sometimes feels like predicting the weather. There are some standards in place already. Like 90% chance of rain probably means it’s going to rain. But, more often than not, some level of disagreement exists. Metrics definitions, the validity of measurement and third party reporting/accreditation are sources of debate for web analysts, salespeople and advertisers. So, when everyone’s prediction differs, how do you know who’s right?

One of the buzz words that attempts to answer this predicament is transparency. Transparency means defining how you will measure something, applying industry standards when possible and being consistent.

With that in mind, I have defined the most common online metrics. Let the clarification begin!

Rachel
______________________________________________________________________

Clicks: A metric which measures the reaction of a user to an Internet ad or linked editorial content. There are three types of clicks: click-throughs, in-unit clicks and mouseovers.

Click-through rate (CTR): The average number of click-throughs per hundred ad impressions, expressed as a percentage.

It is important to understand the limits of click-through rate as the sole metric by which an advertisement’s effectiveness is measured. CTR measures the percentage of people who clicked on the ad to arrive at the destination site, but does not include the people who arrived at the site later as a result of seeing the ad. CTR is a measure of the immediate response to an ad, but not the overall response to an ad.

Clicks had more value when Web site traffic was accepted as a measure of success regardless of where users went after they landed on the site. The trend towards profitability, along with advanced tracking capabilities, has resulted in less interest in click-through rates and more interest in conversion rates.

Conversion rate: The percentage of visitors who take a desired action.

Examples of desired actions include sales, registrations, downloads, and virtually anything beyond page browsing. A high conversion rate depends on several factors: the interest level of the consumer, the attractiveness of the offer and the ease of the action to name a few.

The interest level of the consumer can be maximized by reaching the right visitor, in the right place, and at the right time.

The attractiveness of the offer depends on relevancy, includes the value proposition and how well it is presented. Small, impulse items (like groceries) typically have a higher conversion rate than large, shopping items.

The consumer’s ease in completing the desired action is largely dependent on site usability, intuitive navigation and fast loading pages.

Demographics: Common characteristics used for population or audience segmentation, such as age, gender, household income, education, marital status, etc. A network with a strong demographic that is in line with an advertiser’s desired target audience will be more effective than a broad national network.

Impression: A single instance of an online advertisement being displayed. Ad impressions indicate how many times an advertisement is seen in total.

Page View: A request to load a single HTML page. Page views indicate how many times a webpage is seen in total.

Search Engine Optimization: The process of choosing targeted keyword phrases related to a site, and ensuring that the site places well when those keyword phrases are part of a Web search. For more information on SEO and Google, check out Google Papers , a list of publications and white papers written by Google employees.

Stickiness: The amount of time spent at a site over a given time period.

Unique Visitors: Individuals who have visited a site or network at least once in an allotted time frame, often a 30 day period.

Most measurements of unique visitors are estimates. Sites often calculate unique visitors based on IP address information and through cookies. However, many factors may skew the results.

Traffic rating companies typically calculate unique visitors by monitoring actual usage of a group of volunteers (called panel data), then applying the results to a network’s population. Some, notably Quantcast , use direct publisher data and panel data.