I can’t really do anything about what he signs up for, but I could block that traffic from being reported so it doesn’t pollute my data. Would you recommend that?

It’s nice to know that I’m not the only one who has to deal with this kind of thing.

Thanks for the blog Avinash.

Im always amazed at your analysis skills which you use to give effective practical solutions. Your suggestion of using the Top Landing Pages to drill down further to find the source of “unqualified traffic” was a prime example.

Am looking forward for more. Please keep them coming :)

Jahangir

As always, another brilliant post. Thank you. The “dear avinash” examples were extremely helpful.

I’d like to share my 4Q experience with ‘bounce rate’.

I run a few different Ecommerce sites and I ran the 4Q survey on all. One of the sites is relatively small in terms of traffic and content and although didn’t have a high bounce rate - it didn’t get enough (by my definition) page views. Some I attributed to not relevant keyword traffic (we had products related to the keyword while people were searching for information related to the keyword). I always wondered why they wouldn’t stick around and at least look at more products. When 4Q was released - I ran the survey and had my answer within a few days from one of the survey respondents. It had to do with navigation. I changed it up the same day and boy the page view graph looks so much sweeter now. :) (I understand higher page views aren’t necessarily a better thing - but for this particular website higher page views are important.)

The navigation being the issue is something I would never have guessed. I run another Ecommerce site which has higher traffic and a similar navigation structure and 4Q revealed navigation was working very well for that site.

I was amazed at how 2 ecommerce sites with similar navigation structures were working so differently. It had to do with the subject matter and the keywords that were driving the traffic. No amount of number data could have solved this for us so quickly.

4Q is brilliant and very useful and I just had to share this in relation to the second example and bounce rates.

Can’t thank you enough for your contribution with 4Q. Thank you.

Manjula

Sorry about missing last week’s SM BLinks. I was really swamped and was not able to make the post in time. Anyway, the BLinkings continues. Here are top posts of the weeks. Enjoy!

What is a #1 Google Ranking Worth?

Place ads on television with Google TV Ads

The Main Factor Necessary to Convert Visitors to Customers

“Dear Avinash”: Stressed Agency Analyst & Robots Are Out To Get Me!

Flavacado: Your generosity is greatly appreciated, you have given some brilliant advice. I think you helped Mr. / Ms. Stressed Analyst way more than I did!

Thank you so much, it shows that you have had lots of experience (and dare I say battle scars :)).

Stephane: You are exceedingly kind but I think you have your limits totally wrong.

Sam: I have thought of a forum but I am already running on fumes and could not possibly do justice to such a worthy cause. I do appreciate the suggestion.

Also as every reader of this blog knows, the awesome Web Analytics Association has a great forum on Yahoo groups. Please check it out:

http://tech.groups.yahoo.com/group/webanalytics/

Jeremy, Nicole, Jen: Thanks for the encouragement, I am going to try and do more future “dear avinash” posts.

I honestly did not think they might be of value, this post was something I published in a pinch (12 speeches in five countries in the next four weeks!).

-Avinash.

I value every post that you include on your blog Avinash and the comments from other individuals.

So thank you!

Have you considered starting a “ask avinash” forum where we can all benefit from your wisdom when you answer the questions you get? If two random emails you shared here are of so much value I wonder how much more we can benefit from having visibility to all of them!

Keep up the great work for the web analtyics community.

Sam
(PS excellent presentations at the eMetrics summit, the one on tuesday on Actionable Web Analytics was the highlight of the conference.)

At eMetrics there was a great session called Advanced eMetrics Management with June Dershewitz (Semphonic), James Gardner (Aquent), Dylan Lewis (Intuit) and Mark Brooks (Mottley Fool). While most of the conference focused on tools, technologies, processes and such, this time it was all about the most important KPI: people. How to manage and grow your career. Really great panel and I wish they will make their presentation available.

Working for an Agency myself, I completely relate to the first questions and can see how having a guru like yourself weigh in.

Then there was the second question that was very helpful to simply walk through the question and proposed deconstruction of the problem.

Both answers are excellent. Really good thinking on answering the second question. Keep posting more.. not one per week…

Be business savvy, understand the nature of the business/client that you work on. Ask yourself, what is the goal of the business, how does it relate to the advertising- this should answer what kind of metrics you want to measure. If the purpose of a website is to sell furniture, but the goals of the ads are to build awareness, you’ll want to look at engagement metrics on the website for users who were directed to the site from the ads. you can look at how well the ads affect sales too, but online conversions may not necessarily be the goal of the campaign (people may want to see furniture on the site, but not buy till they see it in person in the store)

Good grasp of the marketing campaign structure itself (how else would one prove good integration of search and banner/flash ad campaigns with the numbers to back it up) . Could be interesting if you test the results to see if the integration works as according to planned (does it really look as good as it does on paper?).

Know the differences in types of metrics that different data sources can provide (example: should you use comscore or hitwise), which should save you time in knowing where you can find the data. Also, stay on top of new reporting capabilities the vendors will offer. Requires some research on your end, but could be time saving in the long run.

Data cleansing - data shown you reports you retrieve may need to be modified, or is inaccurate. Lobby for dba help if you find yourself consolidating/revising massive amounts of advertising response data from different vendors (happens if you cannot track everything out of the ad-server- and also, the limitied capabilies of such tracking systems- which deserve a post in of itself which leads to my next point)

Customize reports with vendors if possible. Are you using doubleclick/atlas/pointroll/etc to retrieve reports? Is it providing you data that you need? If not, can you work with them to customize reports that will be tailored to your campaigns.

Know basic online marketing and site tracking concepts- probably doesn’t apply to people who are taking time out to read this post :)

Thank you for your time :-)

P.S. : Great post by the way.

Agency Analyst life is fast-paced, high-stressed, chock full of various different types of clients and “mysteries”, like Avinash likes to say. In a way, you’re also a detective / federal agent of the web, discovering clues and solving puzzles.

There are probably dozens of recommendations one could give Mr. / Mrs. / Ms. Stressed Agency Analyst, but the one I’ll give is to not rush, to take your time, and focus on the quality of your work and not the speed of it (I find that sometimes the demand for “ASAP” can become overwhelming - managing your level of “ASAP” can go a long way for you).

Thanks AK!