Category Archives: measurement

Reponse to Commenter Jaffer Ali’s "Driving In The Rear-View Mirror"

Leave a reply

Jaffer Ali, the semi-retired CEO of EVTV1, and a jovial industry colleague is usually good for some creative commentary and periodic fire-starting.

Today’s piece “Driving In The Rear-View Mirror” published in Mediapost, required a response as I completely disagreed with it.

http://www.mediapost.com/publications/?fa=Articles.showArticle&art_aid=113117

This notion is nothing new. In fact, I know it has been on the radar since the Web 1.0 days; check out the original 1997 Lilypad white paper discussing time-shifted response-behavior and measurement:

http://www.seicheanalytics.com/consulting/lilypad.html#bapm

(See Figure #C.)

TwitClicks drops Twitter user feature?

Leave a reply


TwitClicks is a good URL shortening and analytics service. At free it is hard to complain too much.

However, there has been a noticeable absence of a key and unique qualitative reporting feature of TwitClicks – “Who Visited – See Best Guess”. While the domain of the traffic source shows up (Twitter, Facebook, LinkedIN even Ning show up fine) since March or April 2009 there is no more data on specific Twitter users. Previously, this worked great and revealed which users were responding to which messages – essential for understanding response and viral propagation.

A simple test was performed on a recent post on this very blog to determine the source of the problem. A TwitClicks shortened URL was posted on Twitter and functioned fine.

TwitClicks results.

However, when it came to reporting the previously available user name feature TwitClicks consistently provides no data over several more tests. More specificlly, there are absolutely no user names of Twitter clickers anymore. Possibly, related to this it was determined after a referral monitoring test that Google Analytics was being used on a redirected TwitClicks page – basically in the middle between Twitter and the target landing page. While the redirect process is essential, it unclear what value the JavScript-based Google Analytics system brings in the context of TwitClicks.

With all the compeition for URL-shortners and Twitter analytics tools, it is odd that this feature was dropped at this time. Inquiries to TwitClicks have not been successful – would love to hear from other measurement-oriented folks if they are having the same experience.

Chicago Analytics?

Leave a reply

I’m always interested in connecting (and reconnecting) with colleagues in the Chicago area; especially those working in the crazy field of online measurement. The other day I received an urgent request from Meetup.com, warning that the Chicago Data and Strategy Consortium meet-up was about to be canceled! So, I volunteered myself to prevent that from happening…

Lincoln Park Lagoon looking southeast.

Why? I personally, would like to see an informal group of professionals working in this nascent field that isn’t always served by other local groups like Chicago AMA, CIMA, national groups like IAB, OPA and the more software-oriented or social/networking groups. Ideally, less drinking and more learning. Thin overhead and easy to manage…maybe even invitation-only?

In what areas of analytics and research are people interested?

Contact me and let me know what you think…we’re all busy people and all at different stages of our respective careers.

http://www.chicagoanalytics.org

Measurement Gobbledygook

Leave a reply
With the proliferation of analytics, there has also been a boom in jargon, semantic-driven miscues and misuse of terminology essential to business management.

Goal – the singular and primary winning state that is desired.

Objective – the measurable milestones that contribute and define the goal.

Strategy – the actions, tasks and/or steps needed to realize the goal.

Tactic – the tools or implements used to support execution of the strategy and reach the objectives.

SM (Success Metric) A subset of metrics or KPIs that communicate the efficacy

KPI (Key Performance Indicator) An important measure that should be tracked and reported.

MVA (Most Valued Action) Like an SM, but generic enough to include clicks or other non-page serving activity.

Precision Exactness of a measurement, e.g. 1.2 Page Views/Visit versus 1 vs 1.2109.

Accuracy Indication of how close a measure is to the generally accepted; it is about the relative difference.

Efficiency Measure of how well desired outcome is achieved per given unit of input, often expressed as a ratio

Efficacy Measure of the ability to produce a avery specific outcome, often a binary determination under given conditions.

Effectiveness Measure of the capability to produce the desired outcome, used more generally and as a relative measure.

Why the Click Is the Wrong Metric for Online Ads

Leave a reply

Story in AdAge by Abbey Klassen. Amazing that in 2009 this is still being written about….the comments even more telling. Here’s mine:

Wow. It’s great that this issue is being raised. However, this is nothing new. Same story since the Web 1.0 days as this could have been written over 10 years ago…actually, I did in the Lilypad white paper!

The twist today is that too many so-called online “marketers” either protesth too much (like the effect of branding but want to pay like performance) AND/OR opt for the easy way out and fallback on search – a FUD mentality. The former is solved by better media-side negotiating and the latter by training and education.

It is called branding and it’s about way more than measuring clicks.

RSS Advertising Part I – Cat & Mouse

Leave a reply

Why RSS Advertising?
So, it has been decided that you want to target that hard-to-reach segment. The one that hates advertising and doesn’t click anyway. RSS advertising with in-feed ad networks like Pheedo show great promise at reaching these folks; more interesting is that they seem to be getting them to respond at significantly higher rates. Measurement on the other hand is still a bit dicey.

First a quick marketer-friendly primer on RSS. Typically, RSS feeds are accessed or consumed with either a dedicated standalone reader application or through a standard Web-browser accessible service like Google Reader; dedicated reader applications can be used on both desktop/laptop computers and when reading feeds while mobile. Using a special version of XML, RSS (Really Simple Syndication) is published and syndicated out to subscribers that opt-in to receive your regularly updated content.

They’re Just Not that Into You
RSS has taken off and in a sense has enabled publishers to cannibalize themselves by allowing access to their content in a largely ad-free environment. Almost all major media now have at least one feed and many have multiple feeds – some even personalizable. Being able to avoiding ad-cluttered Web sites is part of the RSS appeal: think 100% signal with 0% noise. Clearly,

many are very comfortable receiving information this way.

At the same time, media companies are clamoring to sell advertising against this new platform and online markters are eager to reach these consumers. Anecdotally, the RSS audience constitutes a very desireable market segment: influential, tech-savvy, affluent and naturally early-adopters. From a behavioral standpoint, these folks are known to be much less responsive to display advertising (wrong target audience, see Natural Born Clickers by ComScore). In addition, they are more likely to actively delete cookies, opt-out of email/ad targeting and employ ad blockers to avoid advertising. With these very media-literate people – it is a game of cat and mouse.
With so much going for RSS Advertising and promising results – the challenge then is, how to definitively measure success.
To Be Continued…
RSS Advertising Part II – The Measurement Crater

If a Tree Falls in a Forest…

Leave a reply

…and they don’t see your ad. Did it impact your brand?

Back at it in SF.

Which tool is being used for ad effectiveness research can help answer the age-old, “If a Tree Falls in a Forest” riddle that relates to ad effectiveness in online media. Such online research has come along way since studying it with Professor Stasch back at Loyola’s GSB.

A question from Cindy on the WAA forum that struck close to home:

“Can anyone suggest a product that is similar to Dynamic Logic. Dynamic
Logic is a good product, but there may be certain campaigns or
initiatives that would be better served by a similar vendor.
Any ideas are welcome.”

At GSF, we found that recruitment for media research is an ongoing challenge spanning study design, questionnaire development, funding, media partner alignment, statstical significance, recruitment technicalities and results preparation.

Especially thorny is achieving statistical significance when recruiting a target audience that is the least amenable to being surveyed. However, clients need this information often for internal financial modeling as well as the marketing value. Some findings:

  • After three quarters of Dynamic Logic (a Millward-Brown and therefore WPP company like GSF) and a pure-intercept basis; InsightExpress offers a similar approach; it was decided that we pass on more of the same.
  • ComScore offers a combined panel-based and intercept service leveraging similar control vs. exposure methods (BrandMetrix & CampaignMetrix). Sounds very promising.
  • Nielsen offers a solution in this vein also but has a somewhat smaller (but growing) panel; they did offer an interesting post-exposure email option that was novel.

ComScore shows alot of potential for a number of reasons: one interesting by-product of the panel-based approach is the potential to understand the context of what users were doing before and after being exposed to an advertising message – in addition to clickers. For example, what is the likelihood of a trademarked or category search or visiting the target brand’s Web site after being exposed? Turns out quite high.

For some background on what you can do with this (and handy industry benchmarks) be sure to check out their “Whither the Click” white paper from last December.

Good luck Cindy!