« Know your data 4: Facebook is everywhere | Main | Keeping one's appetite after touring the sausage factory »


Feed You can follow this conversation by subscribing to the comment feed for this post.


Pfizer would be expecting to get something from the non-clicks, in the same way that they get something from all their other advertising. In other forms of advertising effectiveness is usually evaluated by restricting the advertising to certain geographic areas and looking at the effect on sales.


Rightly or wrongly, the impressions/GRP argument will be made for it. However, targeting has always been a game at the margin(see Precision/Recall).

San Nayak

This is a wrong image. Last item is "Tabloids & Blogs", CTR should be .165% instead of .0165% I guess..


Am I reading correctly: 1 milion ads, minus 100 clicks equal 999.899 non-clicks? Does that mean 1 undecided? It's probably a real-life example, since in my WA tools, I find glitches like this.
Great post, though. And it's probably what Ken means: Pfizer will be trusting/hoping for 'post-impression' converions.

Staci Burruel

Media and entertainment-related ads being on top of the CTR list was not a surprise for me, since most people go to the internet exactly for these things. Local search engine marketing firms are also getting a lot of clients wanting their entertainment-related campaigns to be advertised, meaning these people are definitely in-the-know about the advantages of advertising media and entertainment-related campaigns in social networking sites like Facebook. Local internet marketing, specifically in Facebook, requires a huge amount of R&D before the actual application to ensure that your money won't be wasted.


OK. So, I'm not in the industry but unless the industry uses a different system of mathematics your maths is out by a factor of 100.

0.01 = 1 click per 100 impressions

or 100 clicks per 10,000 impressions not 1,000,000 as you state.

Please feel free to correct me but if I'm not wrong I certainly won't be buying your book.


Noodle: the chart states 0.01% not 1%. The fact that you're looking at miniscule numbers is the reason for this post. And my book is not an arithmetic textbook - it's about statistical concepts.

Stevanef: you caught a typo of mine. In this industry, missings or data errors are treated as non-clicks. Only a positive click event is counted as a click. And while there should really be undecided events, say when an ad did not load completely, this type of situation is not typically tracked.

Ken: you're pointing out a hot-button issue in the advertising industry, which is that almost no one puts real value onto the value of "branding". Given that Facebook uses a cost per click, they have already conceded that only clicks are worth anything. I don't agree with this but what you infer from their pricing schedule is this.


You still have an arithmetic error: 1 million ads - 100 clicks = 999,900 non-clicks. Even a statistical conceptualist should get that one right...

The problem, though, reminds me of looking for genetic markers for rare disease -- one has very few cases but huge variation within the cases. No-one has figured that out yet, either.

Out of interest, what are "normal" CTRs? (say, for the nytimes?)


David: That's fixed now. Thanks for your comment. There is certainly a similarity with genetics. The difference is that people put a lot more trust in the genetics findings just because of the subject matter, which in my mind, is misplaced. That is the subtext for my post on genetic testing here.

All: While I appreciate errata, I don't understand why they are being delivered with a smack on the face. After all, this is a blog and not a refereed journal.


It was meant more as a wry prod rather than a slap in the face... Surely there is some humour in quantitative pros making basic errors of arithmetic. I fear your friend Noodle didn't find it quite so amusing.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)


Link to Principal Analytics Prep

See our curriculum, instructors. Apply.
Business analytics and data visualization expert. Author and Speaker. Founder of Principal Analytics Prep, MS Applied Analytics at Columbia. See my full bio.

Next Events

May: 2 New York Marketing Association Big Data Workshop, NYC

May: 5 NYPL Analytics Careers Talk, NYC

May: 8 Data Visualization Seminar, Denver, CO

May: 15 Data Visualization Seminar, Cambridge, MA

May: 17 Data Visualization Seminar, Philadelphia, PA

May: 22 Data Visualization Seminar, San Ramon, CA

Past Events

See here

Future Courses (New York)

Summer: Statistical Reasoning & Numbersense, Principal Analytics Prep (4 weeks)

Summer: Applied Analytics Frameworks & Methods, Columbia (6 weeks)

Junk Charts Blog

Link to junkcharts

Graphics design by Amanda Lee


  • only in Big Data