Analytics for Social & Viral Marketing

Social and viral marketing are all about knowing your audience, especially that most infectious segment of your target that you’re considering your seed vector, and the best way to learn about them is through analytics. I’ve been thinking about analytics for viral marketing for a few days and then I saw a post on Social Media Explorer about analytics for bloggers, which sparked me to write this post. There are 3 stages of usage for analytics in the viral marketing process: research, growth and evaluation. And one sort of theoretical one: testing.


When you’re planning your viral campaign, you should first focus on researching your target audience as I suggested in my checklist. This means not only understanding the demographics, preferences and behaviors of the entire group you’d like to reach but also identifying that segment of your target that shares content the most frequently and with the most people, this will be your seed vector. Typically these people are early adopters who are savvy social media users and you’ll be aiming to seed your content in a place where they’re very likely to see it.

Say you want to reach an older, more affluent female audience rather than the typically young-techy-male demographic that frequents sites like Digg. Head on over to Quantcast and look at the data on some sites like Kirtsy, Sugarloving, BlogHer, and StumbleUpon. You’ll see that of these sites, Kirtsy best matches your target audience. Quantcast also gives you some great content preference data indicating which category of sites Kirtsy users tend to like:

Jewlery/Luxury Goods
Politics & Commentary

To expand your list of target seed sites beyond Kirtsy, you can then jump into Google’s Ad Planner and find sites with audiences that match Kirtsy’s. Here you’ll find sites like CoolMomPicks, and The Pioneer Woman. You’ll notice that for some sites (including The Pioneer Woman), Quantcast even offers links to other sites that share the same audience which you can use to further expand your seeding site list.

Now you know where to put your content so that it will be seen by the most social-media-savvy members of your target demographic and you know which types of content they prefer viewing.


Lots of sites use Google Analytics, its a great tool and unbeatable for the price, but there is a reporting delay of a few hours so it doesn’t do much good in those critical hours right after you launch your campaign. To help manage and accelerate the seeding and growth phases of your campaign, you’ll need some sort of real-time-analytics.

I use and like Clicky’s spy feature (affiliate link). Using a real-time spy like this allows you to see where your traffic is coming from, right now, so if your campaign does begin to go viral and users start reposting it and sharing it across the web, you’ll be among the first to know. If your most recent blog post started getting a little popular on Digg and you noticed in the spy that it also began drawing traffic from Mixx, you could give the story a little exposure on Mixx too (perhaps by Tweeting about it or adding a Mixx button to your post).

Real-time-analytics and acceleration like this is an important paradigm to understand in social and viral marketing. You can’t just go and shove your content everywhere they’ll take it, real growth and contagiousness is organic. It is better to carefully and effectively seed your content in a few choice places and then let the savvy and prolific audiences exposed to it in those places spread it further. When they do begin spreading it, make sure you’re there and ready to pour gas on the fire.


Once you see that the growth of your campaign has started to plateau or slow down, it is time to leverage your analytics for evaluation. This is the stage where you compare your metrics to the goals and KPI you set before you began the campaign. Was the goal of this to get more links to your site? Check your inbound link count. Was it to generate blogger buzz? Do some blog searches for your campaign on sites like Technorati and Google Blogsearch. Were you trying to drive traffic to your site? Look at your Google Analytics reports. Were video views your goal? Click the Statistics & Data tab on Youtube and see how well you did.

Beyond simple traffic metrics, there are some more advanced systems available that allow you to measure the number of times your content was shared, some platforms (including emails and press releases) will tell you the number of times people used the send-to-a-friend functionality. I’ve also played around with a concept idea that can measure how many times your link was sent (via any online communication method, IM, email, Facebook, etc) and viewed as well as develop a viral family tree of sorts showing which seeds led to the greatest eventual growth. I may release some version of this in the future, and I know there are similar proprietary solutions available that do almost similar things.


The more advanced tracking and family tree tracing system I mentioned above leads into a possible use for analytics in viral marketing that I have not seen actually developed yet. It is theoretically possible to develop a split or multivariate testing suite that tests content not against conversions or even engagement metrics (time on site, page views per visit) but against actual content sharing. This way we could test which words, images and designs lead to the content being shared the most and actually begin to quantify ‘virality’. Has anyone seen anything like this working in the wild yet?

Read More

People Lie, Data Doesn’t

It’s a basic truth of the human condition that everybody lies. The only variable is about what.

-Dr House

House often didn’t see his patients so they wouldn’t have a chance to lie to him. People lie, data doesn’t.

The same is true with analytics. Designers, developers and owners lie, statistics don’t. People have ulterior motivations and egos, numbers don’t.

Read More

Don’t Leave Conversion and Usability for Last

It used to be an easy target to warn against only thinking about search engine friendliness after a site was built, every few weeks another “seo expert” would come out and tell stories of entirely built sites that had to be re-engineered to allow spiders the best possible access to its content. And while, at least in my little corner of the web world, that lesson has been learned and is starting to sound redundant and obvious, those who refuse to learn from past mistakes will repeat them.

These days I’m finding it very common that things like conversion and usability based on real user testing and analytics data are only being thought of after a site is built. Often highly arbitrary “best practices” guidelines are followed (or not) during an entirely aesthetically-driven design process, that leads to a pretty looking site that has a huge bounce rate and horrible conversions. Then 6 months or a year later an analyst is called in and clicktracks is expected to save the day.

Some common reasons justifying this I’ve heard are “user testing is too expensive” (sure, but its less expensive than doing a free redesign for the client because their site looks nifty, but doesn’t make them any money), or that only after a site is live do you have “real” data. Thats true only in the narrowest analytics sense of the word, qualitative user testing is certainly possible with 5 to 8 subjects before a site is live, and if even that is too much to ask, someone with knowledge of usability and conversion enhancement should be involved in early in the design process, not just graphic designers.

Jakob Nielson’s most recent alertbox columm touches on this a little:

Having a good designer doesn’t eliminate the need for a systematic usability process. Risk reduction and quality improvement both require user testing and other usability methods.

Designers are just that, designers, not user experience or interaction specialist, and certainly not target users.

And on a side note, I’m curious, what would you guys call an “average” homepage bounce rate?

Read More

Social Technographics and Cross-Segment Appeal

Forrester Research recently released a report, that I’ve been lucky enough to read, entitled Social Technographics that profiles adult US web users and their level of engagement with social media:

“Site features can also influence participation profiles. Not all Social Computing/Web 2.0 sites are created the same — the profiles for and YouTube differ significantly, given the activities available on those sites.”

The report describes 6 levels of web users who engage in social media activities, they’re described as a ladder of increasing involvement, starting with spectators and ending with creators. According to the report, myspace is more popular with the joiner segment and while youtube is more popular with spectators.
This means that sites that wish to engage social media users must have features that speak directly to each segment. In this way, those users who are just testing the web 2.0 waters and are still on the lower rungs will have a low barrier to entry and those users at the top will be able to interact with your site on a level of sophistication of their choosing.
For the creators, you should make your story or message easy to repost on their blog or remix or re-create however they want. This means a Creative Commons license and actively asking readers to show you what they can do with your content.
Critics like to review and comment, so make sure your site has at least someplace for these users to post comments. A ratings system will also help engage these users.
Collectors use RSS and tags, so make sure your site has an RSS feed for updated content and the reader-specific chicklets are in place. You may also want to think about including visible tags links to places like technorati or flickr, to suggest possible tags for your content when these collectors repost it. You can also setup special tags so that viewers can track posts to various social sites that refer to your site.
To attract and engage the joiner profile, message boards or other social networking features can be invaluable. This type of functionality requires a large investment in setup and maintenance so they may not be feasible for every creative. Instead you may want to try to add widgets or other portable content that can be integrated into your user’s existing social networking profiles.
The lowest level of social media users in this report are the spectators, those users that read blogs, watch videos and listen to podcasts. These people love content and media, so give them your message in every possible media, video, audio, and written.

Read More

Scientific Web Site Conversion Enhancement

I heard a client say recently that trying to make changes to an established site to increase its conversion rate was just haphazardly guessing, and they were corrected by someone who said that the right way to do it would be to guess and then test with multivariate tests. I disagree.
The scientific method says we should study the subject first, then make a hypothesis and test it. Study the site visitor’s current behaviors first through analytics or user testing/surveys. Then you can begin to make assertions about what problems said site visitors are having on the site and you can make the necessary conversion enhancements and test them.
Email me if you want to talk about boosting your conversion rates.
PS this site’s 1 year birthday was 2 days ago.

Read More

Reductive Analytics and Testing

Let’s say you’ve got a website. Consumer e-commerce. You get lots of visitors and you have lots of pages. Most of your traffic is from search engines, and your keyword range is wide, with a natural head-tail power law curve to it. Some of these people buy things, most do not and like any other business owner, you want to figure out how to make more people buy stuff from you. Nothing on the site is broken, or screams for help, like broken search or poor navigation and the site does sell a few things, just not a lot.
For me, the key has been segmenting visitors into behavioral cross sections. Like I’ll first look at the keywords the visitors came from and see if there are any keyphrases or root-keyword groups with abnormally low time-on-site averages or conversion rate. If one or many do then I would stop paying per click on those words if I was and I would exclude them from further research. I do this to rule out keyword-intent as a possible source of low conversion rate.
Then I’ll look at entry pages the same way. If I find an under performing landing page, I’d also expect to see a high bounce rate. If I didn’t have any landing pages relevant to the traffic the bad one was receiving, it would be come a candidate for testing.
The same analysis would then be applied to key pages in the shopping and checkout experience and if I found any un-replaceable pages I’d move them to the testing phase.
The pages selected for testing would be broken down into component pieces and variants of each chunk would be run through tests. Each variation would have large differences and I’d only test 2 or 3 of each, perhaps even using a “bad” one. Testing a low number of variations will allow me to run the test quickly and look for which page sections produced the biggest range in performance among the variations. These are the important parts.
Then tap the skills of experts in design or communication for interative testing and fine-tuning of the important page elements. When your improvements plateau, return to an earlier step and repeat the process.

Read More

Frank Luntz’s Words that Work and Youtube Dial Sessions

I just got finished audio-booking pollster Frank Luntz’s new book Words that Work. While I’m very different from Luntz politically (he’s the guy who renamed the estate tax to the death tax, and is anti-“illegal immigration”) I’m absolutely fascinated by his work with language, specifically his testing methods. The book is a great read/listen especially for an online marketing professional who relies on words and images entirely to sell a product or a viewpoint, lots of great stuff in there.
But what really piques my interest in Luntz, and has since I first became aware of him when I watched the documentary “The Persuaders”, is his testing methods and how they overlap with the split and multivariate testing stuff I’ve been doing. Obviously testing wording variations with something like Google’s Website Optimizer is easy enough, but he uses another method that I think is missing from the internet marketing toolbox: dial sessions.
A dial session is shown in the movie and he refers to them in the book, but the idea is that a bunch of people watch a video and turn a knob at the same time. If they agree with the video they turn the knob one way and if they disagree the other way. I’m going to look into implimenting an ajax slider control in conjunction with youtube videos to create an online dial session system. I think the sheer breadth of content and test subjects available makes this a goldmine of useful data.
Obviously there are some challanges to address, like recording second by second data, correlating video content to slider position and interfacing with youtube’s flash widget, but I don’t reckon it’ll be too hard.
Stay tuned, I do plan to share my tool when developed.

Read More

Google Website Optimizer Review

After surviving the webmaster deathmatch at pubcon Las Vegas, we were able to secure beta access to Google’s sweet new multivaritate testing system, Google website optimizer. I’ve been using it for a few weeks now and I’m liking it so far. For a free product, its awesome, but it does seem very beta still. Here’s a couple of points of contention I have with it:

Once you start an experiment, there’s no way to edit variations or any of the other experiment settings, and if you stop an experiment you can’t restart it, you have to create a new one from scratch.

In creating an experiment the only way to specify page sections to test is by putting section code in the test page and using the javascript to name it. For instance if you want to test 3 sections of the page you have to put three occurances of the code on the page and name the sections in the code. I wish there was an interface in the Website Optimizer system where I could create and name the page sections to test and add the code seperately, but this isn’t a huge deal.

When you’re creating variations for a page section to test, you have to hit save before you can preview the page, again not a huge problem, but a little confusing.

If you’ve created an experiment and not finished building it (like if you messed up) there’s no way to delete it from your list of experiments.

The navigation of the Website Optimizer seems inconsistant, sometimes I can’t find my way back to a screen I was previously on.

Sometimes I log into other google accounts (like my personal one, rather than our adwords account) and when I log back into adwords, I can access everything but the optimizer. If I click on it’s tab I’m sent momentarily back to the log in screen then back to the adwords account homepage by a redirection. I’m forced to restart my browser (this is happening on firefox).

The only success metric it seems I can measure is a direct conversion, sometimes I’d like to just check out time on site, or pageviews. Perhaps this is doable via integration with Google Analytics, but I’m not sure.

But again, these are mostly little convienance issues, and overall I love the app, especially at the price (free).

Read More