Social Media Marketing

Measuring the Web from ad:tech San Francisco

Bottom Shape

Krista Neher

Apr 21 2010

Measurement is a hot topic in digital marketing. Over the years measurement techniques have evolved and marketers are looking beyond the click to get a full picture of measurement.

In this session at ad:tech David Smith, CEO of MediaSmith shared some insights on measurement for digital marketing.

Visitors = Audience

Unique Visitors = Net Reach

ClickThrough = Consumer interest/action (Direct Response)

ViewThrough = Consumer interest/actions (branded).

Engagement – Still fuzzy. This active involvement with communications

Key Takeaway: There is too much emphasis on clicks at the cost of other metrics.


Audience buying on the web can be different than buying audience in traditional media. In traditional media you purchase demographics and you may have some waste to the extent that you can’t target very specifically. On the web there is very little waste because you can target more specifically. Think target audience vs. total audience – in traditional media you get the total audience and on the web you can choose to buy total audience or just your target audience. Targeting comes at a price – your CPM may be higher when you are targeting more specifically, however if there is less waste you may have a higher overall ROI. Don’t get too hung up on the CPM – focus on the overall value.

Unique Visitors

Unique Visitors is a front end metric that tells you how many people are going to a web site. You can get research from companies like comScore, Nielsen NetRatings, Quantcast and Compete. Research companies are not the same as web analytics. Web analytics derive their information from cookies (like Google Analytics and Omniture) however research companies take a more holistic approach.

ClickThrough vs. ViewThrough

Clickthrough is not a Key Performance Indicator. Over 50% of all visitors that come to a site are from viewthrough and not ClickThrough. ViewThrough is the traffic that comes to your site as a result of someone viewing your ad online (although they may not have directly clicked on it). With today’s rich media and applications, a better measure might be clickthrough, viewthrough and engagement. At the very least, look at clickthrough and viewthrough. There is a strong correlation between viewing ads and traffic – consumers may see an ad and go to the site later.

CPM (cost per thousand impressions) is not a Key Performance Indicator. Performance should be the evaluator in direct response campaigns, not the front end pricing. The front end pricing doesn’t matter – pay a premium for better targeting and response relative to the audience you want to buy. Don’t compare HouseHold CPMs with web target CPMs – the web CPMs can be more targeted and produce better result. Marketers need to focus on the actual results and quality vs. just the cost

2 Replies to “Measuring the Web from ad:tech San Francisco”

  1. Krista,

    Great summary of an important topic. Coming from the event marketing industry, which really lags in the ‘measurement’ area, we watched the web measurement pushing against the traditional media measurement terms with great interest. I agree with most of the above but would really like to understand how to better measure “engagement” especially as it correlates to similar spending in traditional media.

    With event marketing, engagement was really high (often one-to-one personal contact) but hard to measure an ROI. With web marketing, engagement can also be really high but can often be easier to measure.

    I think another way to look at things is where a consumer touches a company’s media in the sales funnel, traditional media is often very high awareness level, web is often lower, especially when generated from links from like-minded communities or search. Therefore, as mentioned above CPM is higher AS it should be because you are lower in the sales funnel with the prospect/consumer.

    Great topic – thanks for the post!

    – mark

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Read some more of our recent posts

Start My Free 5 Day Trial

Try FREE for 5 Days!