Analytics Shamalytics

As you might expect, I regularly check out the data on how many people read talkRA, and how popular it is compared to other, similar sites. Not that there are any similar sites, but you know what I mean. Of course, the problem with analysing such data is that a lot of the data is garbage. For example, I went to one well-known web analytics site, and found that for one of the top revenue assurance sites:

We estimate that 90% of visitors to the site come from Brazil

What!?!?? Which well-known RA website is amongst the top 50,000 sites seen in Brazil? Is it one written by somebody in Portuguese? Is it written by an award-winning ‘expert’ with connections to Latin America? Nope. The quote refers to talkRA! Which proves that not everything called ‘analytics’ should be trusted. Whilst talkRA has plenty of readers in Brazil (obrigado!) it has many more readers in other countries. The top five countries for talkRA, per Google’s reliable stats, are as follows:

1. India
2. United States of America
3. United Kingdom
4. Israel
5. Portugal

Brazil comes 14th on the list. Analysing Google’s list is pretty straightforward. It hardly takes a genius to work out why English-speaking countries, countries with large populations, and countries with major RA vendors feature above Chad (180th) and Tajikistan (181st). You are all smart people, so you have no problem looking at this data and concluding that talkRA probably does receive most hits from India, and not as many from Brazil. You can also tell when people come out with lousy data, and ridiculous analysis. If I wrote that talkRA was the 48,153rd most popular website in Brazil, you would know that must be wrong, irrespective of whether that was my analysis, or somebody else’s. You spend your lives looking at data, separating the suspicious from the auspicious, the mendacious from the meaningless. So why do people, working in this sector, still try to mislead you?

Honestly, I have no good answer to my own question. I suspect the answer lies somewhere between money-grubbing stupidity (“hey, if I lie, then idiots will buy more of my crap!”), tolerance (“better not upset the liars… I might be working for them in future”), and apathy (“it’s not my job to kick the butts of liars, that’s why we pay talkRA… whaddaya mean, those guys do it for free?”)

So based on the data available to me, here are a few more rhetorical questions, that should all be less rhetorical…

Why does Papa Rob refuse to acknowledge that talkRA exists?

Papa Rob Mattison, Grand Poobah of RA, we know that you are reading. We can see you doing it! You obviously have no idea how the internet works. Allow me to explain.

Per our magic internet map, it is easy to see that talkRA has a loyal follower somewhere around Algonquin, Illinois, USA. Which is right next to Sleepy Hollow, Illinois. Which is the home of Papa Rob, and GRAPA.

I recognize our Algonquin fan from the analytics. He has been a regular, since the early days of talkRA. This month he has racked up over 16 pages per visit. The average Joe reads the talkRA home page whilst having a coffee break. In other words, the average recurring visitor reads one page per visit. 16 pages per visit places our Algonquin reader at the very top of our global charts. And nobody else has bookmarked our classic 2009 post – ‘Putting the Ass into Revenue Assurance‘ – like this fan has. And the saddest part is, despite the searches for GRAPA-related material, nobody on talkRA has written about GRAPA since January. We mostly ignored GRAPA in 2012, because we heard reports that Bigfoot was nearing extinction. But because we have such a big fan in Illinois, and he is so interested in our views on GRAPA, I promise to feature lots more GRAPA-related content in future.

To be fair, I can understand why Papa Rob visits talkRA. Where else is he going to get his training course material from? From online town hall meetings with his ‘active’ membership? But it is a bit creepy that he keeps re-reading my profile page

Why does cVidya refuse to respond to our questions?

Come on, you not-so-mysterious talkRA fans in Israel. If I can tell that Papa Rob reads talkRA, do you really think I cannot identify why 8mn Israelis generate more hits than 197mn Brazilians? If it was not obvious already, every time a cVidya employee visits talkRA from their office computer, I can see the name of your company. Although somebody misspelled it. Unless there really is a cvidia ltd, in which case I sincerely apologize for confusing your company with theirs. I also saw how lots of cVidya employees spent time reading about WeDo’s WWUG, and that somebody searched for ‘Amir Gefen’. (Hmmm… I wonder who would want to search for ‘Amir Gefen’.) I can even see that one of you prefers British English to American English. So stop kidding around – I know you are reading this!

cVidya already gets a lucky break. We never get to analyse their financial performance, because they are privately owned. The only time we get to do some interesting financial analysis is when cVidya makes a statement about revenues and growth which contradicts an earlier statement about revenues and growth. So I can understand why cVidya is sore about that. But hey, if you are going to work in a sector full of numerate people with advanced analytical skills, maybe the solution is to avoid such obvious inconsistencies. And it could be worse. Subex has to publicly share audited accounts, letting people like me crawl through lots of bad news they would rather not share. cVidya sell analytics. Is it right that they should get so upset when people are analytical?

So, cVidya, do us a favour. And do yourselves a favour whilst you are at it. If you want to be treated with more respect, stop pretending you cannot hear what talkRA is saying. I can get answers from Subex. I can get answers from WeDo. So cVidya should expect the same treatment, which means answering questions. You can start with this perfectly reasonable question about the way the numbers were calculated in your ‘TM Forum best practice standards-based risk simulation’. (Maybe somebody should search around the office for Amir Gefen, and force him to answer the questions raised by his marketing spiel.) I promise that if cVidya gives me an answer, it will be published, without any editing by me. I may comment on the answer, but I will not change it. And yes, I know that you have already read that post. And somebody has been re-reading my profile page. So you know me. And I know you. So maybe you should be less like Papa Rob, and you should stop ignoring me when I ask a reasonable question. After all, I used to like your CTO who runs the TMF RA group. He talks a lot about answers and questions, like he believes in open dialogue. But I stopped liking him when he started refusing to give serious answers to serious questions. So who needs to change – you or me?

Is it ethical for talkRA to ‘spy’ on its readers like this?

On the one hand, the answer to this question is ‘yes’. There are ways for people to hide their identity on the internet, if they want to. This is a website for people in the telecoms sector, so they should be conscious of what data is being captured on the internet. On the other hand, the answer is ‘no’. Why should I want to know anything about you? But this post illustrates how easy it is to piece information together, if you are prepared to do the analysis. I even feel a little bad about sharing the fact that somebody was searching for ‘Amir Gefen’. (Hmmm… I have a theory about which cVidya employee might want to do that.)

On the other other hand, the answer is ‘yes’. talkRA is given for free. So it is bad manners for people to consume, copy, and repeat all our info, and then pretend talkRA does not exist, or that they never heard of us. So on the other other other hand, the answer is ‘yes’! If more people visited, it would be harder to analyse the data, and I would have less reason to do so. So if you read this, do not pretend otherwise. Do the decent thing – share a link and a recommendation with a buddy who has never read talkRA before. And on the other other other other hand, the answer is still ‘yes’. If the NSA is spying on the internet, then why not talkRA? I think you know who you can trust to provide robust data, share honest analysis, and do the right thing. And if not, then why are you still reading this?!?!

Eric Priezkalns
Eric Priezkalns
Eric is the Editor of Commsrisk. Look here for more about the history of Commsrisk and the role played by Eric.

Eric is also the Chief Executive of the Risk & Assurance Group (RAG), a global association of professionals working in risk management and business assurance for communications providers.

Previously Eric was Director of Risk Management for Qatar Telecom and he has worked with Cable & Wireless, T‑Mobile, Sky, Worldcom and other telcos. He was lead author of Revenue Assurance: Expert Opinions for Communications Providers, published by CRC Press. He is a qualified chartered accountant, with degrees in information systems, and in mathematics and philosophy.

2 Comments on "Analytics Shamalytics"

  1. Avatar Hans Eismann | 1 Jul 2013 at 2:47 pm |

    Hi Eric,

    A couple of remarks:
    1) Throughout the article I kept thinking of the “NSA-type” analysis you had embarked on, only to have that admitted at the end of the blog post ;-). touché
    2) I am also based in Illinois – just outside of the beautiful city of Chicago. I wonder what town shows up when I navigate TalkRA – Lisle, Naperville, Lombard?
    3) I do share TalkRA with others

    enjoy the summer!

  2. @ Hans, thanks for leaving a comment on the site – normally we can’t drag you away from the Twitterverse, though your retweets are always much appreciated!

    To respond to point 2, I recognized the name of Naperville, so that was my first guess. But then I checked the data and saw how Twitter was a popular inbound source for readers in Lisle, so I think that answers your question ;)

Comments are closed.