Google made two big announcements today around their Analytics service: a real-time chart which we covered earlier and a new premium analytics service. The premium analytics service is geared towards Google’s largest customers and includes the following items listed on their announcement posting:

• Extra processing power – increased data collection, more custom variables and downloadable, unsampled reports
• Advanced analysis – attribution modeling tools that allow you to test different models for assigning credit to conversions
• Service and support – experts to guide customized installation, and dedicated account management on call – all backed by 24/7 support
• Guarantees – service level agreements for data collection, processing and reporting

So basically for a fee, Google will provide a stronger server to run your custom reports, a dedicated account manager and a 24/7 support and service level agreement (SLA). The one thing Google didn’t note in their announcement post was the cost for the premium analytics offering.

I have received tips from several sources noting that the price for the premium analytics service is A WHOPPING $150,000/year.

If your company wants to use Google Analytics Premium, you can either signup directly with Google or with one of their resellers. Be ready to bring your checkbook.

For comparison, many report paying over $100K a year for Adobe’s Omniture SiteCatalyst offering. I can’t remember what we used to pay for Webtrends Enterprise at my old job but it was in the same ballpark.

Find more stories about: analytics, Google Analytics

Today Google announced the launch of real-time analytics within their Google Analytics service. I’ve been in the web analytics area since the first beta of Webtrends and am glad to see Google finally offering a real-time feature as part of their very popular metrics offering. Real-time stats are critical when you run ad campaigns or are now using social media tools to help spread your marketing messages. When you monitor real-time stats, you can make changes on-the-fly to your campaigns to optimize the heck out of your spend. I also hear from other bloggers that they use real-time metrics tools to monitor for traffic based on adjustments to story titles to see which provide a better response.

Naturally today’s Google Analytics announcement must mean that real-time analytics tools including Clicky, Chartbeat, SiteMeter, Woopra and ExtremeTracking are dead. Maybe these tools should consider a quick pivot to real-time daily deals (I kid!).

Blogger Holden Page immediately noted, “Google Analytics went real-time. In one fell swoop hit about four startups, one of which I use religiously (Clicky) with an iron fist.”

I like the blog post by the Clicky team – they always use pepper sarcasm into their posts. The title of their blog post reads, “SHUT. DOWN. EVERYTHING.”

They note, “Hmm… did anyone actually read the announcement that Google made today? This isn’t “real time Google Analytics”, this is a single report in GA that is real time. The rest of GA remains the same. This is more akin to Chartbeat, to be used as a real time compliment to a standard analytics package, rather than a full standalone real time service like Clicky is. But I guarantee you Chartbeat will be just fine, as will everyone else. We’ve all had, and continue to have, plenty of advantages over GA other than real time data.”

They then link to a number of tweets that signal death for Clicky, Woopra and Chartbeat. The quotes include:

• “Sites like Clicky will soon be out of business”
• “getclicky and chartbeat ought to run for the hills”
• “Is this the beginning of the end of Woopra?”
• “Google killing off Chartbeat”

I am still waiting to see proof where death of a service occurred because a large company put out an offering in that space. I guess the closest we could come would be all those tools that built on top of Twitter only to have Twitter build their own – but remember I warned against that many moons ago.

Note: we are an affiliate of Clicky but none of the links in this post carry the affiliate code.

Find more stories about: analytics, Chartbeat, Clicky, Google Analytics

The big news in the social media expert camp from last week was the notification that all links that are posted on Twitter, both on the website and using a third-party app, will now automatically be converted to use Twitter’s official t.co URL shortener. You can read the Twitter help post and the developer guide to learn more about t.co and how it works. Part of the conversation around the t.co rollout reminds me of some of the discussion when Digg launched the Diggbar.

Short tl:dr version: Just because a t.co url appears in your referral log, it does not mean that all the traffic to that url has come from Twitter.

Sean at the analytics service Clicky posted about how the change to use t.co will make it appear that Twitter has zero influence because the referral logs will show t.co instead of twitter.com going forward. When I initially read the post, I thought there was something wrong but couldn’t put my finger on it at the time. Yesterday Zee at the Next Web (note: they are a Twitter default user) created a post titled, “Twitter Just Got the Respect it Deserves”. He continues along the same lines as Clicky and has several key points:

• Twitter is now influential in terms of traffic from an “eyes of the media” standpoint
• Facebook and Stumbleupon better watch out because now Twitter will appear really big as a social media traffic driver
• Brands and businesses will now take note of how influential Twitter is
• You can search to find out which tweet was the influential one because you can search on a specific t.co url

Unfortunately it looks like there are issues with both the Clicky post (and the very wrong change they made to their service) and Zee’s statements.

T.co and how it appears in the referral log

There are lots of times where a t.co URL gets copied OUTSIDE of Twitter. For example, yesterday Mashable reported about Gaddafi and the changes in Libya. When that story gets posted into the Mashable Twitter stream, it now takes on a t.co URL. But followers of the Mashable Twitter stream may copy the URL (which copies the t.co URL not the original URL) and place it into Facebook, Google+ or any other social network.

It is important to note that Twitter does appear to be changing the URL when a user copies a t.co url and posts it on Twitter – this will help with tracking back to a user assuming that the URL is never copied outside of Twitter. BUT, in my testing, if you retweet using the new retweet option, the URL stays the same – this will cause the same issue I discuss below.

When the URL is clicked, it seems to appear on most analytics tools that the traffic is coming via t.co and not from the source network (i.e. Facebook, etc.). If we use the theory that Zee posted, this will make t.co and Twitter itself appear more influential than it is because to his team, it now looks like Twitter sent all the traffic when it was another network. Additionally this would make the Twitter user who originally posted the link appear to have driven traffic that might have come from other sources.

And it’s not just an issue for social network postings – if I copy a t.co URL from another user and send it via email, why should that appear as influence from Twitter?

Additionally, if you use the search technique that Zee suggests, it may make a Twitter user appear more influential than they are because when the link is clicked 500 times on Facebook, now it will look like that user sent the traffic to your site when in fact it was shared by a popular user on Facebook.

Clicky’s URL change and why it needs to be reversed

I like Clicky and their team – I’ve used their service since the early days and they are one of the few services I will recommend when someone asks about analytics tools (note: there is an ad on the right bar but none of the links in this post carry any referral links). In their blog post, they note, “We just updated Clicky to convert t.co referrers into “twitter.com” automatically. Knowing the specific t.co link someone came from has no value, but knowing someone came from Twitter does.”

Update: Clicky has reversed the change!

Example of user copying t.co URL into Google+:

Zee and several other Clicky customers note in the comments that this change is a huge mistake – that it is important to be able to see the t.co referral to direct posts. I agree with the commenters here that changing any t.co referral to twitter.com is a huge mistake and one that needs to be reversed asap.

Oddly, in my Clicky testing today, it appears that clicking a t.co url from Facebook shows up as a referral from Facebook and clicking a t.co url from Google+ shows up as “plus.google.com/search”.

But when I copy the t.co URL and paste it into a Skype chat or into an email or Gtalk chat, the referral shows up as Twitter.com. These referrals should either show up as direct or as a t.co/xxxxxx referral.

Summary

Just because you see t.co in your referral log, don’t assume that all the traffic came from Twitter – it may have come from any number of social services and/or traditional sources (e.g. email, direct sharing, etc.). It looks like those of us who study analytics will need to sharpen our pencils as we try to figure out exactly where our traffic originates from and the path it took to get to us. And it looks like Twitter has some work to do on how and when they change URLs when a t.co URL is shared on their service.

Find more stories about: analytics, Clicky, Google Analytics, Twitter, url shortener

Angel investor and 500Startups founder Dave McClure took to the stage during the Lean Startup Day at SXSW — his presentation centered around startup metrics for lean startups. If you know Dave, you know that typically every third word is a curse word of some sort – my counter had him at about 3.2 words/curse word across his presentation. It was three years ago at SXSW that Dave McClure introduced the world to getting booty by using his startup metrics techniques for pirates.

My favorite part of the presentation was when Dave asked the audience, “when do you stop adding features” and someone in the audience replied, “when you get acquired”. Everyone laughed and Dave quickly noted that was not the right answer – the correct answer is to stop adding features when your users are happy with what you have. I am not sure I agree – can you name one company that has actually stopped adding features?

Here are some of my notes:

• suggests you read the book Spent
• don’t bother pitching him or emailing him – if you want into his network, you better come via a referral (see slide 41)
• the important metrics aren’t uniques or cash – but visitor engagement metrics (see slide 11 below)
• disagrees with Chris Dixon and Fred Wilson around marketing and SEO (I agree with Dave)
• gathering customer data is good, email collection is great because you can “fucking spam your users” (see Dave’s comment below – just to be clear he was joking about spamming customers)
• if your customers pay for your product, you have MVP (minimum viable product)
• hate is more gooder than love (that was my Fake Grimlock impersonation) – basically what Dave means here is that when people hate your product they will tell you specifically what to fix – if they love your product, you will probably only get an “i love your product” response
• testing is cheap, coding is expensive – Dave notes that you are better off testing up the behind before you code because it’s easy to change a test, harder to change code
• Dave featured a slide about cartoons and noted that SouthPark is a really shitty product within a niche and that’s why they won

Dave also discussed pricing – in that you are better off starting with charging for your service and possibly moving free later on than the reverse. He said that way you will get customers initially who really, really want your product. I agree with his conclusion but for a different reason – it’s very, very hard to move to paid when you start free – just look at what the backlash when this is done – but if you go from paid to free, you look like a hero.

Below are Dave’s slides from the presentation:

Find more stories about: analytics, Conferences, metrics, startups, sxsw, SXSW 2011

I’ve been a fan of web analytics startup Clicky since the beginning. This week they announced a new partnership with Olark that will allow you to chat with your customers and users in real-time. Olark is a Y-Combinator startup that offers a widget which you can place on your site to allow for chat with your customers and/or users.

In all of my years dealing with web analytics, I can’t remember ever seeing an integration like this new Clicky/Olark functionality. Clicky offers a real-time analytics monitoring service called Spy which offers you the ability to watch your traffic in real-time. It can be very addicting, especially if your blog post or startup get some good links or press.

There are plenty of companies that offer popup customer chat windows – I know when I visit the Rackspace site I try to login very quickly so I don’t get hit with a popup chat box.

The new service makes it possible to chat with your customers or users by clicking on their link in the Spy.

A simple use case for this new service is watching a potential customer move around the site and then clicking to chat with them to close the sale. Another use case would be watching a user navigate through a number of pages in your support documentation — click to chat with them to directly answer their questions and be a hero.

To use the new service, you must have a premium account with Clicky (starts at $10/month) and you must have an account with Olark. Olark has a free account which is limited to 20 conversations a month and their paid subscriptions start at $15/month.

Note: We are an affiliate of Clicky but none of the links in this post carry the affiliate code.

Find more stories about: analytics, Clicky, Customer Service, realtime

Phil Gainley from FounderSpeak has put together a detailed, in-depth report regarding bootstrapped startups. As an aside, I hope one day we can get to industry accepted definitions for “bootstrapped” and “startup”. Two years ago I provided my definitions for both terms and many CN readers provided their own definitions. For example, can a company started in 2005 really still be considered a startup?

Phil’s analysis provides a lot of data points around bootstrapped startups and is well worth a read. Phil interviewed 107 startups to gather the data for the report. Phil notes, “included is a breakdown of trends in the tech space, what the lifetime and demographics of a startup looks like and where they fail and what their biggest problems are. Also it has my latest research around the tech startup space all put into a process that other tech startups can use that is around 10 pages long. This research will be part of my Aston Business School MBA.”

The survey provides data and charts on the following areas:

• Number of founders
• How do you provide monetary support for your startup
• Monthly hardware costs
• Misc. spending costs per month – marketing, networking, travel, etc.
• What phase is your startup currently in
• How long have you been working on your startup
• When do you expect to reach breakeven
• Why haven’t you reached breakeven yet
• Type of product/service

I think it would be great to be able to refine the data – for example, breakeven data for startups with 2 founders that are in business more than 18 months, etc. I believe most of the startup data in the report comes from startups based in London/UK and note that the monetary figures are in pounds not dollars.

Find more stories about: analytics, bootstrapped, London, reporting, startups

Marketing service Hubspot has put together 100 marketing charts and graphs which you can use in your upcoming presentations. I’ve embedded the presentation of the charts and graphs below. The charts and graphs come from Hubspot’s original customer research. Data categories include: lead generation, blogging and social media, marketing budgets, and naturally Twitter and Facebook. It appears the data was posted at various times throughout the year but this presentation brings it all together in a neat and tidy slideshare presentation.

Find more stories about: analytics, Graphing Social Patterns, Hubspot, marketing

I’ve been a fan of web analytics startup Clicky since the beginning. Last month the company announced their new video analytics functionality.  This week the company announced a beta test of a new interface to their web analytics product. I guess Twitter wasn’t the only one who launched a new Web interface this week. The new interface is available at beta.getclicky.com.

The Clicky blog explains the changes to the analytics interface which include:

• Ajax page loads which make the pages load faster and smoother
• navigation that changes based on the context of the page
• important stats for each report remain visible as you scroll

My favorite part of Clicky is the real-time spy. The spy lets you watch traffic to, and on, your website in real-time.

The past couple of months I have also been using Piwik which is a free, open source analytics tool. It’s not as polished as Clicky but I am totally in love with the Piwik real-time monitor as well. Piwik must be installed locally while Clicky runs as a web service.

It’s amazing how far web analytics tools have come from the earliest days of Webtrends in the mid-90s.

Note: We are an affiliate of Clicky although none of the links in this post are affiliate links.

Find more stories about: analytics, Clicky, Piwik

What would be your reply to the following question, “what’s your most prized possession?” UK-based online insurance company Protect Your Bubble asked 500 people on Facebook and the responses are just as expected. Unfortunately no demographic data was provided on the respondents which makes it hard to take the results seriously. Naturally if you ask wealthy youngsters what the most prized possessions are, you will get a very different set of results than if you ask a group of senior citizens.

The Apple iPod comes out on top with 12% of respondents noting that the device is their most prized possession.  All of the Apple products on the list add up to 35% of all of the responses. Interestingly dogs beat out cats which typically online only cats get any love. And sorry Apple fanboys but the Blackberry beat out the iPhone with 11.5% vs. 10% respectively.

Here are the full survey results:

• iPod 12%
• Blackberry 11.5%
• iPhone 10%
• Laptop 9%
• Dog 9%
• Games console 8%
• Macbook 7%
• iPad 6%
• Cat 5%
• Camera 5%
• Television 4.5%
• Car 4%
• Video camera 4%
• Other jewellery 4%
• Wedding ring 1%

I’ve been trying to think about what is my prized possession since I first read this press release yesterday. I guess it would have to be my worldwide transit poster and map collection including a large-scale 1960s NYC transit map. My most prized possession is certainly not my very old, refurbished iPod Nano.

Find more stories about: analytics, Apple, iPod, social media, Social Networking

Web analytics service Clicky has announced the launch of their video analytics service which is part of the overall Clicky web analytics platform. Clicky has integrated directly with video hosting company Viddler and provides the ability to track videos from YouTube and Vimeo as well.

The Viddler integration will go live sometime in August as Clicky notes that the Viddler team has some additional work on their end before the connection is made live. The Viddler integration will not require you do anything different to make sure the video stats reach Clicky.

To use the YouTube connection with Clicky, there is a bit of work involved. You can read about the changes to your blog or website that must be added on the Clicky help site. It addition, the way you embed videos is different than just copying and pasting the code directly from YouTube into your blog. The same goes for Vimeo…you would need to add some code to your templates and then change the way you insert videos into your blog or website.

Here’s an example of the stats page:

Note – we are an affiliate for Clicky but none of the links in this post are affiliate links.

Find more stories about: analytics, Clicky, online video, Viddler, YouTube

Video hosting service Viddler has announced the availability of their new analytics tool today. The reporting engine is provided through a partnership between Viddler and TubeMogul.

From the announcement, “Comprehensive statistics about your videos are now only a click away. Of course, Viddler Analytics answers even more questions than those above. How much storage has my account used this month? How much video bandwidth have we used so far? Am I close to my account level threshold yet? These questions, and others, are very important to our Business level accounts.” The partner analytics provides more “social” stats around comments, total video views and individual video stats.

To use the Viddler analytics tool, you must be a business or partner customer. Business accounts start at $100/month. When I attempt to load the analytics tool I am redirected back to the Viddler home page. They should probably change that to redirect to a page describing the analytics option and how to get access to it.

I love seeing more stats and analytics around the content I create so I welcome a new set of video metrics. Now I guess I need to create more videos…anyone see the BloggerKing lately?

Here’s an overview video of how the analytics tool works:

Find more stories about: analytics, social video, Viddler, Video

One of the most practical talks at FOWA London came from the CEO of Freshbooks, an online invoicing app for individuals and small businesses. Mike McDerment reminded the audience of designers, developers and wannabe web 2.0 rock-stars that the excel number-crunchers of the corporate world can teach us a lot about how to measure business performance. When you’re trying to create the next widgetized social aggregation platform which is powered by supercharged virality it might not seem very sexy to fire up excel and start facing those cold white boxes. However, Mike urged everyone there to cook a monitoring system directly into their web-apps so that they know where their visitors are coming from, how many of them are signing up, and, crucially, how many of them are actually sticking around. Importantly, he urged everyone there not to assume that Google Analytics is the answer to all your measuring needs. Measuring is a business philosophy and Mike practices what he preaches. Freshbooks ads are all over the web and the ones that aren’t delivering ROI quickly are cut with the ruthlessness of a ninja swordsman.

This kind of obsession over the numbers is common among successfully bootstrapped companies. When you don’t have an enormous pile of green bills to cover in petrol and set alight, you don’t – simple as that. After spending the first two years in his parents’ basement, Mike has built Freshbooks into an extremely successful business with dozens of employees, thousands of paying users and, most importantly, extremely happy customers. Viral growth is a rare and elusive feat achieved by very few companies. Predictable growth is driven by regular measurement and informed improvement. In any case, if your app is growing like a weed, wouldn’t be nice to measure that?

@HelloApp – Almost awesome
I’ve been to a fair few conferences in the last couple of years and met hundreds of people. I don’t usually go with any kind of objective in mind other than to chat to as many interesting people as possible. However, if I had the ability to sift through the thousands of attendees to see if there was a core group of people that should definitely speak to that would be awesome. That’s where I’d hoped that Hello App would come in for FOWA 2009 (http://hello.carsonified.com). When I first read about Hello App on Think Vitamin I thought it was a brilliant concept. HelloApp is essentially Twitter-powered web-app that lets people “check in” to the conference, tag themselves, and start meeting one another. I thought it would be like conference networking on crack. However, I think that HelloApp tried to do too much. To demonstrate, here’s the list of instructions on how to interact with the app.

Checkin
@helloapp sat XXX
Categorise and tag yourself
@helloapp hello ![des/dev/biz] #tag #tag #tag
Met someone
@helloapp hello @their-username
Post message
@helloapp message your-message
Give High Five
@helloapp hi5 @their-username
Claim a token
@helloapp claim XXX

Personally I think that all it needed were the first two features. The other ones turned it into a game where people tried to “meet” as many people as possible or find tokens around the conference. It went from being really useful to being a touch gimmicky. All I really wanted to do on HelloApp was meet people who were rails coders. Although I could do this with HelloApp I think that a lot of other people didn’t bother because they saw HelloApp as more of a game than a serious networking tool. What a shame.

Cut back and make it awesome
When it comes to software, Dave McClure seems to hate unnecessary features with a passion. Instead of adding more functionality, Dave stressed that you must hack away at your app like the grim reaper until it is a simple and functional as it can be. His talk was dubbed, “Startup Metrics for Pirates”. Almost every techie I’ve ever chatted to for any length of time about an app has said: “Wouldn’t it be cool if …”. By the sounds of it Dave would like to remove this phrase from all techies’ vocabulary. His line of reasoning is similar to the philosophy adopted by 37signals where they argue that you should “underdo”, not outdo the competition in terms of features so that you can deliver a more elegant, usable solution to your customers. Although the idea of setting out to offer less than your competition might initially feel uncomfortable, it actually allows you to focus and create better, not bigger software. As McClure puts it: “Cut a feature and if your customers don’t scream, let it die”.

Find more stories about: analytics, fowa, fowalondon07, FreshBooks, future of web apps, Richard Burton, startups

Web metrics reporting service comScore has announced a new partnership today with the newly-Adobe-acquired Omniture. The new strategic partnership is being announced in conjunction with New York’s Advertising Week. Financial terms of the partnership were not disclosed.

From the announcement, “the offering will combine the power of Omniture’s Web analytics with comScore’s new Media Metrix 360 hybrid audience measurement to help provide publishers and advertisers with a unified and comprehensive view of online audiences.”

My guess is that Omniture customers will now be able to view their own data next to industry and competitor analytics provided by comScore. It’s a smart move for both companies and should help Omniture customers save time and generate better analysis.

Could this bring comScore closer to the “quantified” service offered by Quantcast? From the announcement, “The strategic partner relationship is intended to allow joint Omniture and comScore customers to use Omniture tags to collect and share information with Media Metrix 360 using Omniture Genesis integration technology, quickly bypassing the normal implementation process for Media Metrix 360. The relationship also opens up the possibility of joint product initiatives that will leverage the granularity of the Omniture site-specific data with the Web-wide view of Internet user behavior provided by comScore.”

Find more stories about: Adobe, analytics, comScore, Omniture

We were the first to report last month on Twitter tracking your links. The click functionality adds a “click event” to add links clicked on the site. So if you click a link to CN, click a friend’s profile or click a link over to one of the third party apps, Twitter is tracking it.

The links appear normal to the naked eye but when clicked they morph to something like this:

http://twitter.com/link_click_count?url

From my previous post, “It does not look like third party apps (HootSuite, Tweetdeck, Sobees, etc.) are affected by this link tracking change. Frankly tracking links without tracking the third party apps seems silly as apparently the hardcore Twitter users use some application with the service.”

My guess is that they are tracking the links for both internal and external purposes and perhaps this is the first step to some sort of ranking matrix. Please leave your thoughts on the reasoning behind tracking every link in the comments below.

Update: A CN reader tells me that TweetMeme appears to not be counting correctly and he wonders if this new click tracking has anything to do with it. I don’t use TweetMeme but will contact the service for more details.

Find more stories about: analytics, reporting, Twitter

If you read CN regularly, you know that I am an analytics nut. I’ve been interested in analytics since I was a baby playing with my Fisher-Price cash register. In the online space I like Clicky because they offer a real-time spy where you can sit and watch where your users and how they are interacting with your website.

Today analytics service ClickTale has upped the ante on real-time analytical monitoring with their own “Real-Time Monitor”. The ClickTale service provides a similar set of features to the Clicky real-time Spy product but adds additional functionality that monitor user’s mouse moves, clicks, scrolling and keystrokes.

Each user session is recorded and can be viewed as a video where you can watch exactly how the user interacted with your site or service. You can watch how users scroll, click, etc. This is pretty powerful stuff as it enables the ability to modify a site not just based on content but also with interactivity and location. It’s a great complement for any A/B testing plan.

The new Monitor service is free for all ClickTale plans including their free offering. ClickTale notes, “the Real-Time Monitor automatically refreshes itself every 10 seconds for our paid plan customers, and every 60 seconds for our free plan customers.”

Checkout all of our ClickTale coverage including why I think Omniture should acquire the analytics service.

Find more stories about: analytics, ClickTale, reporting