E-mail Permission Marketing: it works!

In 1999 Seth Godin wrote an incredible book called “Permission Marketing” (you can check out part of “Permission Marketing” for free at Amazon or could read the entire first four chapters if you e-mail me at damian [at] idaconcpts [dot] com, and yes, I am authorized to forward the first three chapters of this book via e-mail, as long as I don’t make any profit out of it). Seth is a truly great author and you should check him out. He’s such a good author that he often gives out part of his work for free (as in the case of “Permission Marketing”) or even complete books.

The concept of permission marketing is best exemplified by the smart use of e-mail.

skitch-permission-email — Use of permission marketing from Skitch.com.

Recognize this little checkbox? Every time that you are interested in an online service or product and you need to register to be able to use it, the makers will ask you this question. Do you want to hear from us? This little question is very powerful because you are already engaging in a conversation with your users. You’re letting them know that you want to keep in touch with them and likewise you are asking them, if they are ok with it.

Now, I understand that you might challenge this proposition: people are ok with an organization reaching out to them. Well, consider this survey from eMarketer:

Do you notice the change in user trends about e-mail messages from companies from 2005 to 2008? People care about these e-mails because 1) they are REGISTERED, 2) they checked the little “it’s ok to contact me” box during registration, 3) the e-mails that they are receiving are personal, relevant, and ANTICIPATED messages (the three pillars of Seth’s permission marketing), and 4) they can choose when and where to check these messages.

“It’s about me, it’s about what I’m interested right now, and it’s delivered in a format that I want to get it.” – Seth Godin (“All Marketers are Liars” presentation at Google, 8:41)

The beauty behind e-mail permission marketing is that people do not have to be interrupted, they choose when to review the information (if, at all).

However, any organization can mess up this priceless, golden permission that its users have provided by abusing this permission and turning its e-mail messages into impersonal, irrelevant and unanticipated.

In conclusion:

Provide your users a check-box during registration so they can decide whether or not to give you permission to contact them.
Make your e-mail messages personal (ask for feedback, provide various channels for communication, thank them for giving you a job), relevant (tell them about how you are making their experience better, tell them how other users got in touch with you and they made a difference in the new release) , and anticipated (bi-weekly, monthly).
Don’t abuse the permission your users they have given you: NEVER sell your e-mail list to other vendors, ALWAYS provide the option for 1-click, easy unsubscription, and ALWAYS respect if they decide to unsubscribe from your e-mail list.

Thank you for your time. In the next post I will discuss about web metrics of permission marketing e-mail campaigns.

"Google Devalues Everything It Touches" – Wall Street Journal Chief

I have chosen this title for today’s post from an article by Tom Foremski at the SiliconValleyWatcher because it touches my experience with Google Adwords both for a real state broker’s website in Guayaquil, Ecuador and a Mac photo software developer here in Hawaii.

The gem from this article is the quote:

Mr. Thomson (Managing Editor of the Wall Street Journal) said, “Google devalues everything it touches. Google is great for Google but it’s terrible for content providers.” He said that Google doesn’t distinguish between the quality of the content around which it serves up ads, it is concerned with quantity rather than quality.

I have to agree with Mr. Thomson on this comment because Google Adwords currently lacks a lot of features that are necessary to truly create ads that are segmented towards specific niches. I know that this last statement might be challenged very quickly by talented SEO experts.

Let’s take a look at two cases, one positive and one negative.

Positive:

Visits to srbienesraices.com from August 1, 2008 to February 13, 2009

As it becomes quite clear from the graph above, having included a Google Adwords CPC campaign into srbienesraices.com was a great idea because it create up to 20 times more daily visitors. More than 80% of the site visits were provided by this campaign and the bounce rate from these visits was not that much higher than the bounce rate, about 40%, of direct traffic during the same time period. Once the CPC campaign was finished, the number of daily visitors drop back to previous levels (this means that extensive A/B Testing is required! However, that is beyond the objective of today’s post).

Google Adwords was able to drive visitors to a small website for a real state broker in Guayaquil, Ecuador because the concept of the website was simple: 1) visitor wants to buy real state in Ecuador, 2) visitor can take a sneak peek at some properties (very few pictures are uploaded ON PURPOSE), and 3) visitor have to contact real state broker to get more information. The funnel strategy is plain vanilla.

There are no visitor requirements. Neither technical (e.g. browser type, operating system) nor demographic (e.g. age, income).

So what happens when technical details and demographic are included?

Negative:

I cannot go into specific details of my Google Adwords campaign at my past position at iLovePhotos but I can tell you that when you add technical and demographic requirements for your visitors from a Google Adwords campaign, things are not that smooth.

For example, if I am targeting a male in the 18-25 year old range that uses Firefox on Windows Vista and that is interested in photography, I am at a serious disadvantage with Google Adwords. Remember the quote from Mr. Thompson above: Google is good for quantity, not quality. Google Adwords will allow me to target a lot of people (I had about 3 million ad impressions during a 2 week period) but the number of clicks from the people with the technical and demographic characteristics that I required was about 0.0026%).

A/B Testing of ads? At Google Adwords I did plenty but there is no way I can customize the ads to target users with specific types of browsers or operating systems (if you know how to, then contact me at damiandavilarojas [at] gmail [dot] com right now). The only way that I see around this is to do extensive research on the demographics and technical background of the users at specific websites and then target these specific websites using Google Adwords.

In conclusion, Google provides useful customization options such as region, ad location and target content but still lacks more ad segmentation features. If you have a very simple funnel strategy and just require quantity of users with very few (or no) demographic and technical requirements, then go ahead use Google Adwords! But if your funnel strategy is more complicated and/or your users have specific technical and demographic requirements, then you will have a tougher time using Google Adwords. Google Analytics is doing a great job at advance segmentation, how long is it until Google Adwords catches up?

Viral marketing

Take a look at the AARRR model from Dave McClure and tell me what you think is the hardest step?

The AARRR Model from Dave McClure (Master of 500 Hats)

I asked this same question to a class of MBA students here at the Shidler College of Business and the answer of choice was the last step: Revenue.

Yes, conversion is very, very hard to do. However, I think that what my MBA colleagues missed is that you cannot get to the Revenue step without Retention and Referral. Even though viral marketing only appears under the Referral step, I have found from my personal experience that viral marketing involves both Referral and Retention.

The current state-of-the-art of Acquisition is so advanced that acquiring users (more than 30 seconds on your site and at leat 2-3 pageviews) is relatively easy.

There are plenty of techniques (refer to the orange square above) that are pretty effective of practicing interruption marketing (as Seth Godin calls it and he provides quite a sad example of it). A highly effective of acquiring customers is through Facebook Advertising, if you want to find out more about it read this post on how to target your audience using Facebook Advertising.

Currently web marketers are masters of the Acquisition step and MBA students (future web marketers) are focusing on finding out how to excel at the Revenue step. The best example of this sad business model is the thought that Twitter is a Cash Cow in the Making (derive a funny @name, horde tons of followers, and reap the CPC rewards). In a nutshell, the thought is that Retention and Referral are going to happen automatically somehow in any startup model. During the dot-com era, and some still today, Internet startups fail to understand that the most common source of failure for startups is a lack of customers and not a lack of product development. Often startups are good at managing its product development, but terrible at managing its customer development.

The gold (a.k.a traction or conversion) is to develop effective, scalable, contagious, ADDICTIVE Retention and Referral steps.

Viral marketing is essential for the success of any business enterprise. Word-of-mouth beats any marketing concoction any given day.

Really good examples of viral marketing are:

1. Photojojo’s Scavenger Hunt: This little forum post has created 504 responses from Photojojo’s readers. It is a very, very simple idea, yet very, very, very A-D-D-I-C-T-I-V-E.

2. Sprout widgets: I am big fan of Sprout because it allows you to tell a story and then that story can be shared with others. Here is my stab at creating a Sprout widget for iLovePhotos. This little widget can be found in various places of Facebook and I have found that people see it as a little pin of support for a little startup from Hawaii. You can found our widget at Bacon Lettuce Photo – The iLovePhotos Blog.

3. Blogs that instead of being e-mail are ME-mail: the perfect example is Flickr. This should not be a surprise but it is still a very hard idea to push. Instead of telling people how great your company and product are, you should be telling your users how awesome they are. Build a tribe (another Seth Godin term) that is about making feel your users good.

Retention and Referral are hard to achieve and there is no magical sure-shot way to do it. I hope that this post gets you thinking about their importance.

How to target your audience using Facebook Ads

For the last week, I have been pretty busy creating and tweaking online ads at Google Adwords and Facebook Ads, so I thought it would be useful to provide a bit of advice from my personal experience using Facebook Ads.

Why did I choose Facebook Ads? Consider the following bar graph from Venture Beat’s article titled Facebook’s traffic growth leaving rivals in the dust.

For simplicity, I will assume that you want to drive visitors to a single website.

Do your research: You cannot expect Facebook Ads to do miracles for you. It will do a pretty decent job at providing impressions but the “clickability” of your ad is 100% up to you. Forget one-size-fits-all approaches, you will require to develop at least 5 ads (I am currently working with 8). You need to think about the profiles of your website visitors. For a quick video tutorial of this idea, take the quick tour of the web attitudinal web analytics firm iPerceptions from Canada. (Avinash Kaushik is on their Advisory Council, so yes, you have to listen). Is your ad audience: female? male? young? old? English speaking? Time-constrained? Etc, etc, etc. Preparation of audience profiles should be about 60% of your time dedicated to develop online ads.
Select your text and image for your ad: As you can see from the two ads below, you can have a subject line of 25 characters, text of 135 characters, and you can upload an image (there are appear no limits on the image file size because Facebook will resize it to fit the add). It is important that you have a variety of images available because you will be needing as you A/B test your ads. Notice that Facebook users can give your ad a thumbs-up or a thumbs-down to your ad! Don’t try to fully capitalize words (e.g. FREE), or use strange characters (keep to the regular alphabet, both using different languages is ok, actually I encourage it!). Remember to use proper punctuation, otherwise your ad won’t be approved.
Think of the keyword(s) of your ad: This step is critical because it will allow you to use SEO and SEM tools such as Google Trends. Let’s assume that you want to promote your great widgets at www.widgets.com. So an important keyword to consider would be “widgets”. Let’s check out what is the search volume index of “widgets” at Google Trends. Wow, this is a lot of useful information! Notice that we get a couple of relevant news that show what drove queries for this term. Company names and product names are useful because then you can go into their websites and see what are their current SEO and SEM practices. Also, you should look if the already have ads around Facebook. Don’t try to reinvent the wheel and keep an eye on the current practices of your competitors. It is also important to check out what keywords are related to your keyword, what regions (or countries or cities) provide the most queries for your keyword, and what languages are those queries made on. Another important source of information is Google Ad Planner. For a discussion on how to use Google Ad Planner, refer to this web analytics analysis of Flickr, Photobucket, Shutterlfly, Snapfish, and Slide using Google Ad Planner.
Reach the (exact) audience you want: Using the information from the previous steps, you can fill in the fields on step 3. Notice that not all keywords are available at Facebook, so its important that you look for keywords related to your own keyword(s) of choice.
Price your ad: I will skip this step for now, because it deserves a whole post of its own. If this is your first time creating Facebook ads, then I would recommend setting the price towards the lower limit and setting a total budget for 1 month. Keep track of your A/B testing and then you will have enough information to develop a more detailed pricing strategy.

That’s it for now and happy experimenting with Facebook Ads!

2009: An interesting year for idaconcpts

Hello idaconcpts readers,

Happy New Year! I apologize for the delay since my last post at idaconcpts.com but work has been quite busy (which is always a good thing!). I still cannot believe that it has been almost 6 months since I started this web analytics blog. The response has been quite positive, as the following chart shows:

Number of monthly readers at idaconcpts.com

Back in August 2008, this web analytics blog only had a bit over 200 readers but those numbers grew much larger because of the ongoing discussion about the key players of the photo sharing industry (Flickr, Photobucket, Snapfish, Shutterfly, and Slide). The use of great SEO and SEM tools such as StumbleUpon and Twitter, (my username is @idaconcpts and the username of the account of my company is @ilovephotos) combined with useful examples of web analytics tools such as Google Analytics, Google Trends, and Google Ad Planner, have brought a lot of people to idaconcpts to put ideas and concepts to work in web analytics.

An important update for this blog in 2009 is that it will migrate to another hosting company because I am getting quite irritated that I cannot implement a lot of useful Java script widgets while hosted at WordPress (e.g. here for a discussion about installing Google Analytics while hosted at WordPress.com). Also, I really want to install my Twitter badge here and, under the current hosting status, I can’t.

Regarding the number of Mac users worldwide, the question is still up in the air, but I had interest from Gartner, Inc. (@gartner_inc). I am still pushing the 80 million number for Mac users worlwide, what do you think? The Macworld 2009 keynote brought very little light into this issue and I am still looking for validation of this statistic.

On a personal note, what the Macworld 2009 keynote brought was a lot of heated competition for us at iLovePhotos. The new release of iPhoto09 caught us by surprise, particularly the “Faces” update. Some people say that a pictures speaks louder than a 1,000 words, so let’s give it a try.

iPhoto9 “Faces”:

overlay_organize_03_20090106 overlay_organize_01_20090106 iLovePhotos:

2840700284_b3e8210e64

tagging-faces

I know that the second shot is a bit overwhelming, but I am heavy user of iLovePhotos and I have a lot of tags already.

We live in a competitive market and the reality is that if it hadn’t be Apple, it would have been somebody else (e.g. et tu Picasa?). The important point is, in the own words of Seth Godin (page 108 of Tribes):

The only thing that makes people and organizations great is their willingness to be not great along the way. The desire to fail on the way to reaching a bigger goal is the untold secret to success.

2009 will surely be an interesting year and I am really looking forward to it because adversity always brings the best out of people.

Here at idaconcpts, I promise to be bring you the most interesting, up-to-date articles about web analytics and web marketing. I look forward for your comments this 2009.

Cheers,

Damian

How many pageviews do news-oriented web sites need?

Today I ran into an interesting article from my daily feed of Online Media Daily. If you are curious how it would look in your mailbox, it would be something like this:

Nicholas Carlson wrote an eye-catching article titled “NYTimes.com Needs 7X More Traffic To Survive (NYT)”, which summarizes the findings of an advertising study of contentNetxt. Basically contentNext states that news-oriented web sites operations can be sustainable at the 200 million pageviews mark.

Carlson explains further:

“Based on our research, the conversation [with advertisers] gets interesting at 200 million page views plus a month, but much more so around 800 million,” ContentNext’s Lauren Rich Fine writes in a report.

For big operations, like at Yahoo (YHOO), AOL (TWS) or the New York Times (NYT), that bar needs to be even higher. In order to survive as a Web-only news product, for example, Fine says the New York Times needs about 1.3 billion pageviews a month.

That’s about 1.1 billion more pageviews than the 173 million ComScore says NYTimes.com saw in October.

Here’s a bit of a problem that I have with these numbers. If you go to Google Ad Planner and take a look at the numbers for the New York Times, this is what you get: 490 million pageviews in a 30-day period.

Google Ad Planner shows web metrics of the New York Times.com. — Google Ad Planner shows web metrics of the NYTimes.com.

I might sound a bit picky but I wish there was more consistency in reporting a web metric such as pageviews. As a web analytics consultant, if I was to try to understand how to increase traffic at NYTimes.com, I would start by taking a look at the trends in daily unique visitors as to segment (or slice like a ninja, as Avinash Kaushik would say) the audience into customer experiences. Given that I am cheap, I try to work with free data as much as possible, so free tools such as Google Trends and Google Ad Planner are my best friends.

Going back to the cited article, contentNext mentions that for news-oriented online operations, the bar needs to be set really high at more than 1 billion pageviews. Yahoo! and AOL are cited as examples.

Let’s take a macro view of the audience demographics of Yahoo!, AOL and NYtimes.com.

My conclusion is that NYTimes.com should target more its female readers (notice how Yahoo! and AOL have a bigger percentage of female visitors) and its younger readers (notice how Yahoo! and AOL have a high number of visitors on the under 18 category, that is the first bar on the age graph).

This is just a general suggestion, but it’s a start. Besides, it’s free advice, unless of course they would like to hire a new web analytics expert. : )

What do you think?

Exploring Flickr Communities

For the last week, I have been quite busy exploring Flickr communities, in order to develop a better understanding of what drives people to organize and share their digital photos online.

Here’s my Flickr photostream: http://www.flickr.com/photos/i-love-photos/

Through the search of full text (and tags) such as face detection (face detection), face recognition (face recognition), and photo lover (photo lover); I am finding the most interesting people you can imagine.

Why face detection and face recognition? Because I believe that we can use face detection and face recognition to make organizing, sharing, and enjoying your photos better!

Why photo lover? Because that’s the term we use to describe the users of iLovePhotos.

I will keep on exploring Flickr communities the rest of this week and will write about my experiences later on next week.

A (not so) simple question: How many Mac users are there in the world?

How many Mac users are there in the world?

Pretty straigthforward, huh?

Not so much.

According to Apple in May 2006, there were about 15 million Mac OS X active users worldwide.

In March 2007, the number of Mac OS X users was estimated to be 22 million strong, according to analysts at Bank of America Securities. The Leopard release generated about 6 million more users.

Continue reading “A (not so) simple question: How many Mac users are there in the world?”

A brief history of web analytics

As part of a MBA assignment, I had to summarize the history of web analytics. I thought it would be nice to share it with you:

– – – – – –

The official definition of web analytics by the Web Analytics Association is the measurement, collection, analysis and reporting of Internet data for the purposes of understanding and optimizing Web usage. This standardized definition was not proposed until 2006, which reflects how young the field is. Only until the early 1990s did the use of log files become popular among nontechnical persons, particularly with the creation of Analog, one of the first log file analysis programs that was widely available on the Internet, by Dr. Stephen Turner in 1995 . Commercial web analytics started with the founding of WebTrends in Portland, Oregon in 1993, even if they didn’t start selling software until 1995 . Other important companies are NetGenesis (established in 1994 by MIT graduates), Accrue, Omniture, and WebSideStory (all founded in 1996). By the year 2000, web analytics vendors were struggling with web server logs as optimal sources of data and JavaScript tags emerged as a new standard for collecting data from websites. JavaScript log files are easier to maintain than web server log files and their use shifts the responsibility of collecting and processing data from internal company IT departments to web analytics vendors in most cases. Currently the three big vendors are Coremetrics, Omniture, WebTrends. Mid-market vendors are Unica, Yahoo! Web Analytics and ClickTracks. Consolidation is common in this industry, for example Omniture acquired the previously fourth big vendor, Visual Sciences (better known as WebSideStory), and Yahoo! Web Analytics was born out of Yahoo!’s acquisition of IndexTools. Finally there are several basic solutions such as StatCounter and Webalizer. Google reshaped the web analytics industry in 2005 when it purchased Urchin and, subsequently, released it as a free tool under the Google Analytics name. This made first-class web analytics tools available to anybody for free. The key to success in this industry is constant innovation such as the use of heat maps (cluster of clicks on a web page and their density using colors) from CrazyEgg . The latest trend in web analytics is moving away from prepackaged key performance indicators (e.g. number of pageviews) towards key actionable insights (e.g. visitor’s primary purpose of visit). The key trendsetter in the web analytics industry is Google with the Google Analytics tool and Analytics Evangelist, Avinash Kaushik.

Revisiting Flickr versus Snapfish versus Photobucket versus Slide versus Shutterfly

The response to the post:

Flickr versus Snapfish versus Photobucket versus Slide versus Shutterfly

has been great! Thank you very much for reading it. Here is a breakdown of the number of daily readers provided by WordPress BlogStats

I am really happy that my Spanish colleague, Gemma (who writes a great web analytics blog in Spanish called ¿Dónde está Avinash cuando se le necesita?, a great tribute to Avinash Kaushik) found some inspiration to use Google Trends (one of my personal favorite tools!).

Another great tool for broad research is Google Ad Planner (currently in Beta, so you need an invitation, mine took about a week to arrive).

Let’s take a look at the BIG 5 (Flickr, Photobucket, Snapfish, Shutterfly, and Slide) of online digital photo sharing using this tool:

Flickr versus Snapfish versus Photobucket versus Slide versus Shutterfly: using Google Ad Planner

Google Ad Planner provides much more in-depth information on a monthly basis (the figures for unique visitors and pageviews are on a 30-day basis). The reach is for the United States. As discussed on the previous post, Photobucket appears to continue to have the upper-hand over Flickr.

What I found really interesting is that if I was to theoretically put ads on these 5 sites trying to reach everybody in the United States, I would reach 31 million or have a country reach of 13% or have 660 million pageviews.

Do you think that is a lot? Wrong! Take a look what would happen if I just select Facebook and Myspace:

That’s 90 million unique visitors or a country reach of 39% or 20 billion pageviews!

Does that mean that advertising on Facebook and Myspace is better than advertising on the top 5 of online digital photo sharing?

If your objective is to monetize on printing services, I would dare to say no.

Why?

Take a look at the age and income distribution of the audience at Myspace using Google Ad Planner:

Compare it with the demographics of the audience at Shutterfly:

Obviously Shutterfly has a more mature audience with more spendable income, so they would be more prone to spend on printing services than a younger audience with a tigther budget.

Of course, you could argue against this hypothesis.

Let me hear your thoughts.