Alexa, comScore, Demographics measurement, Google Trends, Hitwise, Internet Audience Measurement, Metered Tracking using panels, Neilsen Netrating, plan, qsearch, Segment Matrix

comScore and Audience Measurement

One of the most frequently asked question in the net space is what is size the internet universe and how much traffic can I target (share of Traffic). There are a few free resources that we usually look up Alexa, Google Trends (to see the relative popularity of a term on Google search engine) . Though these are free, the idea of bribing our friends who have access to comScore, Neilsen NetRating, Hitwise or any such panel information, might be a good. The information provided by these audience measurement companies is far superior to the Alexa.

Fortunately, since my organisation subscribes to comScore data, we don’t have to resort to any unethical favours (I’m infact open to lucrative offers). Jokes apart, comScore data helps to access the growth of our site traffic vs the competition and the universe and is a regular feature at the end of the month / board presentation. There were some occasional, where we had to measure the cross visitation/audience duplication matrics to find out our standing in the net space.

More importantly, my world (read universe) was limited to 26 Mln Indian online population, which is the figure that comScore reports,. All this while I considered it to be a better estimate and ignored IAMAIs and NASSCOM). I also had the notion that based on these numbers penetration of eCommerce was only 10% (roughly 2.6 Mln internet users).

And then, I happen to meet our comScore Account Director last week and had lot of my fancy notion changed. This post is an ode to that meeting and I hope you will find it useful. Please note that comScore is the only source for such rich panel information, as of now…

1. Since comScore uses a panel, and recruits only for household and office usage, 26 Mln number is a representation of this community. This doesn’t include cyber cafes (which was 40% in IAMAI 2006 report), Schools & Colleges (6% in 2006) and other sources (which may include net card). This will significantly increase the numbers (and which also suggests that our eCommerce penetration is very low). In some countries, they quantify that the internet usage originates from outside the universe (e.g persons under 15; from public machines (like internet cafe), they do mention that the projections do not reflect this usage. These countries include China, India, Ireland and Mexico.

2. It is not just a traffic estimating tool. It has lot more data to offer. Reach & Frequency Planning (will be a great assets for folks associated with branding), psychographic segmentation, and identification of Heavy/Medium/Low internet users.

3. It has a panel size of 16,000 in India, which is the highest in the region. It is significantly higher than China (8,000) and the next closest panel size is in Australia (15,000) members. Notably, it has a million opt-ins for the application, however most of them are filtered to include Home/Office. Only in US they offer the Office/Home split.

4. comScore collects data for only 15Years+ individual. They identify this through the mouse-use patterns and keyboard usage pattern.

One other objective of this post is to elaborate on the features that comScores provides, which will empower emarketers such as us, to look beyond Alexa and Google trend Data. The idea is look beyond numbers and demographics profiles and get a psychographic rich profile of our site visitors. To some of us who use site analytics, this data might just corroborate the facts. For other, it will define avenues, as all users are not always the same.

comScore’s Segment matrix H/M/L serves exactly that purpose. It acknowledges that web behavior varies greatly by segments. Heavy users might have an unequal impact on the overall statictics- they are 20% of the online population, but consumer 2/3 of total online pages and minutes. The light users on the other hand account for only 6%. Using this feature you can understand and take appropriate action points for heavy, Moderate and Light internet user segments. These are key benefits…

1. Behavior analysis: Heavy users are the easiest to target, but light and medium users would be necessary to build the reach. The basis for this is the content they consume. With this report you will be able to get an idea where your consumers go apart from their category (say travel/retail)

2. Reduce waste of online ad spends: comScore Segment matrix H/M/L provides advertisers and agencies the means to identify sites, where advertising delivery is more likely to reach light internet without overreaching heavy users.

3. Get valuable inputs for site development: Publishers can prioritise development of content and features according to different requirements of H/M/L user segment.

4. Increase ad sales/revenue- Publisher can demonstrate the site’s ability to reach premium segments. Alternatively a marketer can look at psychographic profile of its site visitor.

comScore has one more good product for search related statistics- qSearch. This tracking provides weekly and monthly views of consumer search activity by search engine, which answers the questions…

1. How large is the search market and how fast is it changing.
2. Which search engine has the greatest reach among all searchers and how is it changing.

qSearch also reports on searches conducted on sites such as eBay, Amazon, Expedia, etc. Furthermore, it also includes search activity from following user activities:

– Searching on Maps and yellow pages
– Search conducted through hosted or affiliate search relationship
– Searches conducted across that various tabs found on most search engine e.g web search -> image search -> Video search. The definition of such fall under
– A user interaction where the user is presented with a search result page
– The search result page allows the user with the ability to refine or change their search parameter
– Search can be initiated from a drop down or clicking a link, as long as first two rules are satisfied.

All push-traffic and non user requested activity is filtered. It reports following searches in addition to all the searches

1. Auto Search: Refers to the search that is conducted by entering a search string into the address box in the browser. DNS error pages are not counted.
2. Toolbar search: Search executed from the Toolbar
3. Affiliate/Hosted search: where the search engine is hosted on an affiliate site.

comScore is at an advantageous position in India with its first mover advantage. Nielsen NetRating is still to start and gain ground. Comparatively Neilsen NetRating has a panel size with lower numbers as compared to comScore. However, they are known for their descipline and method and use the same metered tracking using panels and telephone surveys. Their products are called @plan, which helps in media planning (like H/M/L Segments of comScore). Sooner than later they need to have a foothold on one of the fast developing Internet Markets in India.

Hitwise is another such product, which given share of traffic data. Hitwise studies the browsing pattern at ISP level (in Europe, they have tied up with orange) and has more realistic estimate of share of traffic. Their downside, is on getting the Demographic profile of the audience. They too have to start their services in India.

For any emarketer, Brand assessment, buzz quantification and cross media measurement is a challenging task. Like offline disciplines, the need of the hour is to get used to measurement tools like these, which will help to define and shape up realistic business goals.

Do share your comments.



Add to Technorati Favorites

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s