Category: Analytics

Food for Thought

Prasad AjinkyaLeave a comment

The Slow Winter

A hilarious take on Moore’s Law. If you have studied hardware and transistors, then definitely read through this article. Do note, that it’s a PDF document. What is Moore’s Law? … well

Moore’s law is the observation that, over the history of computing hardware, the number of transistors on integrated circuits doubles approximately every two years.

5 Dos and Don’ts of International SEO – Whiteboard Friday

I had earlier written about Moz’s Aleyda and her checklist for International SEO. Last week Moz had a Whiteboard Friday with Aleyda running us through that Checklist. For people wishing to target other geographies where they do not have any presence, this is a must see video.

Six Visual Solutions To Complex Digital Marketing/Analytics Challenges

Avinash Kaushik gives simple and clear answers to most challenges that digital marketers. The explanation uses Venn Diagrams to make strong points. If you are a HiPPo (Highly Paid Person with an Opinion), then do take some time from your busy schedule and go through this. If your organization is not actively working on any analytics initiatives, then it is high time that you should start!

Sexy Up Your Marketing Data

Annie Cushing shares her presentation from SMX East and drives a point home. For marketers to make sense to the top management, data visualization is crucial. People prefer looking at great looking data instead of just a series of numbers.

Why should you use Annotations in Google Analytics?

Prasad Ajinkya1 Comment

Google Analytics is a tool which often amazes me. It contains so many features packed into it that it would take years simply to understand and appreciate all of them! One such feature which I do not see being used as much as it should be is the Annotations tool within Google Analytics.

What are Annotations?

Annotations in Google Analytics

Annotations are comments which you can put for a particular date. So in this picture above, I have indicated at what dates did I publish a post on my blog. I have been busy in the months of April and June!

You can see the annotations bar in the Audience Overview of your Google Analytics profile, there is a small expandable div which you need to click on first.

The thing with Annotations like most of the Google Analytics features is that in order to retain flexibility, it is an open framework. Meaning you can use it as you see fit. Which could be one reason why its not used as often as it should be!

Why should you use Annotations?

Annotations are a simple way to Attribute certain action that you did to a rise in traffic.

Annotations in Google Analytics Peak

In the figure above, there are two trends which I want to explain using annotations.

  1. One is a clear spike in traffic which fetched me around 350 hits in a day. This I am attributing to commenting on a popular gaming blog with my link (I cross referenced that with the referral traffic coming to that site and they match)
  2. The other is an increase in the site traffic, this I am attributing to the increased frequency of posts that I am publishing on this blog

So, How does one use Annotations?

Well there are multiple scenarios how you could use annotations –

  • If you are a team working together on a site, then you can decide on a format for annotations such as Dev. Post, SEM. Having this communicated across the team helps ensure that too many annotations do not clutter your dashboard and confuse you or your team members.
  • If you are a blogger like me, then simply noting the time of the day you blog will help you maintain some regularity on the blog, with a clear sight on how this is impacting your traffic.
  • If you are a marketing manager of a site and handle additional campaigns, say a product launch or a TV Commercial, then you can correlate the increase in traffic and goals on the site to that particular campaign. Remember the first step in attribution is measurement!
  • Cyclicality and Seasonality can be tracked and predicted in the future using annotations. A simple example of this in action is when I had written about the Sunday Dip.

Enough of ideas on how one can use annotations, did you use annotations before and how did you use them?

Why hitcounters are a waste of script?

Prasad AjinkyaLeave a comment

A simple search on Google about hitcounters would lead you to literally hundreds of different sites which offer pretty looking widgets which display a counter in your site. The question is do you think that the number really means anything to you?

What are hitcounters?

hitcounterHitcounters are simple scripts solely created for measuring one thing only. The number of hits you are getting on the site. They are included in the HTML script so that when your site loads, the hitcounter records this as an increment to the total hits on your website. Yes, if you are interested in web analytics, hitcounters are a cumulative measure of pageviews. Thats it.

There used to be a time when web analytics was based purely on parsing server log files and crunching them into numbers. During this time having a hitcounter had a great value (since you would not be required to do the heavy lifting of understanding which page had which files and then parsing the server log). This is the era when these hitcounters really flourished … in fact in the pre Google Analytics days even I had tried a couple of these hitcounters. That’s around a decade back and things have changed.

So what has changed about hitcounters?

With the onset of free web analytics programs such as Google Analytics, Clicky, Mixpanel, you can measure far many things than a simple pageviews metric. In fact the pageviews metric these days is almost meaningless. Hitcounters these days serve only as a source of eye candy on the site … an eye candy which is a sidebar/footer widget at the best or an eyesore at the worst.

There is yet a diminishing population of website owners who still brag about this metric and hence go ahead to put this script on their HTML.

Why do I have against hitcounters?

I do not have any personal vendetta against any hitcounter script. However, think about it … you are adding a script to your page … this impacts your page loading speed. If this is scripted correctly, then hopefully it will not add to your web page loading speeds … if this some random script you have found on the internet, then it might as well be a backdoor for a malware.

A cumulative number is just one data point, instead if you tracked how this is growing, then you could eek out some insights from this number. What about tracking how a user is visiting your site across multiple visits? Naah … this functionality is simply not there in hitcounters! What about e-commerce tracking? Well … not there either. As I said before, it only counts pageviews and shows it in a styled manner. That’s it.

In short, at best it increases your web page loading times. Is that silly looking counter then really worth the cost you are incurring?

My answer is a resounding NO.

I have a hitcounter on my site!

Well there are two approaches you can take. The first is to discount what I have said in my post and go ahead on that dark road.

The Dark Road - Matrix

If you go down that road, forget you ever read this post and live happily in the land where everyone counts the hits on their websites.

The best part about a cumulative metric is that it will never dip!

The other road from here is the road towards enlightenment, towards data and correct measurements. Towards Google Analytics and other tools which will help you measure each and every aspect of your site’s visitors. Here’s what you need to do to take this path –

  1. Remove that hitcounter widget from your site
  2. Register for a Google Account and go to the Google Analytics site
  3. Register your site on GA
  4. Include this code in your site’s template
  5. Start measuring!

In this journey you will come across many hurdles and questions, however on this road you will find other travellers who will have mastered those hurdles and they will be more than happy to help. Have fun and enjoy developing a richer understanding of web analytics!

Conversion Rate Optimization in the Longer Run

Prasad AjinkyaLeave a comment

Today a comment by Analytics guru – Avinash Kaushik made me to pause and take a re-look at my perspective on Conversion Rate Optimization (CRO).

What is Conversion Rate Optimization?

Conversion Rate Optimization is the science behind increasing the conversion rate of the pages of a website.

Let’s say that you are running an Adwords campaign for your website, then most likely you would use a specific landing page built purely for conversions (be it purchases or be it lead generation). With the interaction that I have had with the different Search Engine Marketing agencies in India, the default target conversion rate that is considered acceptable is a measly 5%. At EduPristine, we try for 30% and on an average generate a healthy 25% lead generation conversion rate.

However, the tactics that we ended up using were pretty much what Avinash has mentioned to avoid … most of the time when we conduct CRO experiments, it is with a short term focus of getting an incremental raise in the conversion rates. Here is an excerpt from the comment –

Then I realize that I’m sure someone tested this. I’m sure some “conversion rate optimization guru” was brought in to do this. I’m sure they got 10% more leads. I’m sure their lead conversion rate went from 1.67% to 1.84%. I’m sure to them this looks awesome.
But what about the other 98.16%? People who did not convert, many of whom, like me, might have thought this was distasteful? Some of the 98.16% surely thought “This is not the type of company I want to do business with?”

So where does this put us?

I used to think that 25% conversion rate is awesome (and believe me, it is! :-) ). However, what is happening to the user experience on the site … a modal pop-up here, a lead generation form there … do the visitors on my site really want to see that?

The pragmatist within me says that the purpose of the site is lead generation, and that it is doing brilliantly.

However, the idealist says that in the long run the site needs to welcome the user … it needs to provide that the same user keeps returning again and again for clear value that he gets from visiting the site. The %age of repeat visitors needs to increase … and that metric is something that a CRO specialist hardly looks at.

How about a metric which measures the number of unique visitors who visits your site more than x times? That could be a measure of customer retention.

So what needs to be done?

Instead of focusing on incremental changes on the page to get a temporary raise in conversions, focus on user experience and ask yourself this question … would you visit your own site again?

What are your thoughts on this?

What makes the Perfect Post on Social Platforms?

Prasad Ajinkya2 Comments
Perfect Posts On Social Platforms

Perfect Posts On Social Platforms

 

When it comes to maintaining your presence on different Social Media networks, the key to this is timing, consistency and knowing what your audience appreciates. As your channel matures, your audience will start realizing that your updates are according to a schedule, which means your interested audience can know exactly when to check your channel for news they can use.

The time in these platforms and the platforms themselves should also give you a cue as to which social update to put where and when.

What is the big deal about Big Data?

Prasad AjinkyaLeave a comment

Imagine a typical day at the office for any business executive. All of us would be checking our emails, getting into meetings, attending to our phone calls and planning for that weekend getaway to unwind from a stressful week. Sounds familiar, doesn’t it? These activities will lead to a source of unstructured data being generated, data which until some years back was really difficult to store, much less analyse.

What is Big Data?

The industry has termed all these disparate sets of data as Big Data. Big Data is a collection of data sets which either is unstructured and poses a challenge in storage and retrieval or is such a large set of data that it cannot be analysed by existing systems. Dan Kusnetzky of ZDNet has put it succinctly when he wrote about Big Data –

In simplest terms, the phrase refers to the tools, processes and procedures allowing an organization to create, manipulate, and manage very large data sets and storage facilities.

Big Data could be a series of images that are used in preparing a report, it could be a video recording of an event, and it could also be a behavioural data of users on your web product.

Is Big Data just Hype?

Do we at the end of this article simply write off this jargon … or are there real life applications of Big Data in organizations? The applications of unstructured data needs one to take a step back and look at the bigger picture –

  1. A customer searches for a particular product on an e-commerce site.
  2. The site has the product, but it is currently out of stock.
  3. He tweets about this to the site’s support on Twitter.
  4. They follow-up using Twitter and finally email him about the availability of stock.
  5. The individual goes ahead and purchases the product, he also manages to buy some other gifts for his family.
  6. The site sends him an SMS and email confirmation.
  7. The order is dispatched and the site sends him a separate SMS about this. The order is sent into two separate dispatches with two separate dispatch IDs.
  8. Both the dispatches reach the happy customer. He goes on and thanks the support on Twitter.

If you see this entire cycle, the interactions and data points are across a series of different systems. Some of these systems are not even owned by the e-commerce site, some of them are vendor systems, yet there is a clear need for a better understanding of these data points.

In order to be more responsive to the potential customer, to reduce the lead to cash times, to invest in the right inventory and to justify Social Media expenditures, the organization (our e-commerce site) will have to analyse this data.

Apart from the obvious business applications, big data is also being used to make society safer. For example, in 2009 Google was able to apply big data to search terms to help identify how the H1N1 virus was spreading through communities. They did such a great job at this that not only was their tracking real time but also it was more accurate than the Centres for Disease Control and Prevention (CDC).

The Next Big Challenge

Any business analyst will tell you the importance to have clean, structured and codified data in order to cull out business insights. Without proper codification on a data set, a business analyst cannot run different statistical tests to check their hypotheses.

With Big Data the challenge is that statistical packages such as SPSS, R or the giant in this field SAS cannot handle really high volumes of data (we are talking data of the size of exabytes here!). Multiple problems will have to be addressed and solved to capture the full potential of big data. Policies related to security of data, privacy and ownership of data and intellectual property rights will also need to be formulated.

Understanding the nature of the problem is the first step to solving it. The next step would be start developing methodologies and systems. In the race to develop the next set of software applications, software giants such as Oracle, IBM and even Infosys are looking at developing their own propriety systems.

Summary

For any new technology to be adopted in main stream implementations, it has to go through a chain of events. It will be widely discussed, it will be touted as the next elixir of life, it will also see some failed implementations and then written off as just another snake oil. In the course of these events, some enlightened organizations would end up doing the right implementations, then success stories will be shared and finally it will be adopted by all organizations as a main stream technology.

Big Data has a long way to go, and it is here to stay.