TapeFailure Archive

We’ve reviewed a number of web usage analysis services including ClickTale, RobotReplay and CrazyEgg. We’ve also written about an analysis service named TapeFailure which was closed over a year ago. This week TapeFailure founder Joshua Gross has launched his new web analysis service named Vistrac which is a complete rewrite of TapeFailure plus more functionality and analysis.

Joshua notes, “This service (Vistrac) will analyze everything from where users click, to where they pay attention when scrolling. It takes the concept of sites like UserFly one step further and breaks down the data for you.”

The features page is worth a look for detailed info on what Vistrac provides. There are reports on browser type, OS, screen resolution and browser size. I’ve never seen browser size on a report before but it’s pretty interesting – this is where a user is using their browser at a size other than fullscreen. Other reports include link tracking, heat maps, scroll metrics and form metrics. For a startup, form analysis can mean more signups and more paying customers.

There’s also a Superfunnel report  which provides a map for each user and how they navigate through your website or application.

Vistrac pricing ranges from free to $149/month for a corporate plan. Most of the web analysis services offer a free trial. My suggestion is to give each of them a try on the same website, compare the results and move forward with the services you find beneficial.

Day 2 of the Ten Days of Holiday Gifts features TapeFailure. We covered TapeFailure when they launched and the basic idea is that TapeFailure captures visitor’s Web sessions so you can find and fix the failures. It’s like a focus group without the expense. I’ve used it several times to watch how ya’all use CN. It records every mouse movement. Check out my video review from their launch. If you are launching a startup or typical Web site, you should test it using TapeFailure as it will help you find the flaws and help get your customers to the most optimum user experience.

Here is a simple overview of the difference between TapeFailure and analytics packages: "Tapefailure does not collect basic data that other analytics software (such as Google Analytics) gathers. Instead, it focuses on what the user does, not who the user is or what they use."

Update: Please DO NOT signup for TapeFailure – the site will be closing at the end of this month :(

I’d like to thank each one of you for your loyalty and patronage over the past year. I’ve got a full recap coming but it’s simply amazing where CN has grown and of course it’s thanks to each of you. Like the Wine Man says, "It’s a lot of you and a little of me and together we are changing the tech industry." (I tweaked it a bit).

With that said, I’ve been working hard to line up a variety of exclusive Holiday Gifts for you. Over the next ten business days, there will be one gift each day. From service discounts, to premium beta access to free conference passes, we’ve got something you will like!

In the graphic below, you will note one slot with ??? in it. That’s because I have left one slot open. I’d like your feedback on who should fill that slot and I will go after them! Leave your ideas in the comments. And if you have a service to offer as a treat, contact us. It can’t be a free service is the only rule.

Here’s who’s participating — pretty impressive lineup!:

• OurStage – 12 free song downloads – get holiday gift
• Compete – free credits towards the search analytics tool – get holiday gift
• Clicky – two free pro accounts – get holiday gift
• Freshbooks – 30% off new subscriptions – get holiday gift
• TapeFailure – 40% off new subscriptions – get holiday gift
• Drop.io – free premium upgrade codes – get holiday gift
• CrazyEgg – Some pro accounts – get holiday gift
• Future of Web Apps in Miami – 5 free conference passes – get holiday gift
• CenterNetworks – 50% off job board posts – get holiday gift

Update: Please DO NOT signup for TapeFailure – the site will be closing at the end of this month :(

You won’t know which Holiday Gift will come which day nor what time – so be on the lookout! Happy Holidays and Happy New Year to All!

Has one of your stories hit the front page on Digg, Reddit, StumbleUpon, etc? What I find is that the traffic from these social sites is great (some sites live off this traffic) but the conversion is horrible. Many speak about the way these users spend less than one second on a site and do nothing on the site itself. No RSS pickup, no ad clicks, nothing, a pack of sheep moving from site to site. This is frustrating for me as a marketer — all of these visitors and no real way to capitalize on them. Even if they would just take my RSS feed for a day, I would consider that a win. A click on an advertisement would be gold – not because I might pick up a few cents but because my sponsors would see a higher rate of return and my visitors would learn about these services (which I handpick).

I decided to change the game for some of these social visitors. I spent a good bit of time analyzing exactly what these users do when they visit CN. And the results were sad. Visit the page, leave. Over and over I saw this pattern thousands of times. In my new test group, users clicked the ads an average of 7% more and picked up the CN feed 12% more than in the non-modified group. I consider that a huge gain. The key is a simple thing I learned early in my career as a salesman at The Wiz. Welcoming someone by name will increase the likelyhood of a purchase. A person’s name is the most beautiful sound they can hear. In this case, I used the referring social media site as their name.

The test was relatively simple. I used "seconds" to define whether the visitor would be part of the "modified" group. Odd, yes – even, no. Both groups saw the exact same article and same Web site. The only difference was the inclusion of a "name" box for the modified group which is shown below (it’s a mockup example). What this name box did was call out to the user "Hello!" and drew attention to it.

The box included the following elements:

• Welcome from x site
• Message about RSS with orange RSS icon
• Targeted ad

As for the specific ad types, gadget ads appear to work best for Digg visitors, while content ads worked better for Stumblers.

The great thing about what I setup is that only visitors from those sites saw the box. Many sites will run a "Welcome Digg’rs" type message when their story hits Digg but everyone will see that message. So not only are you not welcomed by name, you are welcomed by someone else’s name. I found that this tactic didn’t work as I had hoped over the past year.

My test has concluded and I am very pleased with the results. The next step is to refine the Welcome box and determine what it might look like going forward. Should every Digg user see it every time? Should a user be able to close it permanently? How far can I push it before I annoy the new visitor. It’s important to keep the brand and trust high.

As you can see a simple change returned dramatic results. And it has me wondering, who else can I welcome?

(tools used include: GetClicky, CrazyEgg, TapeFailure and ClickTale)

The team at ClickTale have released a report from data they gathered from users of the ClickTale service. It would be great if other services such as TapeFailure and CrazyEgg did the same thing. As long as the data is not identified down to the individual service user, this type of data can do two important things: teach us about trends and sell their service! The report provided by ClickTale is very detailed and is a great selling point for why you should be using ClickTale.

The report released today discusses amount of scrolling a site visitor is willing to process. What would be great is to know what type of sites make up the report. For example, people might be more willing to scroll further on a blog than on a product site or a corporate brochure, etc. Most blogs seem to have a good bit of scrolling, especially on the home page.

A few of the interesting conclusions from the Scrolling Report:

• Almost identical percentages of page views (15%-20%) reach the page bottom regardless of page height.
• It appears that regardless of page height, scrolling reach is very similar on the relative scale with some resemblance to a linear model (or a very flat exponential): between 64% to 68% of the page views are likely to reach the 1K pixel line and 15% to 20% will reach the bottom of the page.
• It appears that visitors scroll in a relative way – about the same percentage of page views will reach the middle of a web page regardless of the actual page height in pixels.

Do you believe that blogs have changed the way we scroll on a Web page? From the old days where the belief was to fit as much as possible on the first screen, I would say they have. Check out our previous ClickTale coverage.

We had an exclusive private beta look at TapeFailure back in March. The invites we had to give away went in minutes. Everyone who gives TapeFailure a try seems to have positive feedback about the app. As we look to the future of CenterNetworks, I have used it to take a look at how you use the site. Is it a testing app? Is it an analytics app? To find out the answer and learn more about the founder and the app, I chatted with Joshua Gross who founded TapeFailure. Check out our discussion below.

Allen: Can you provide a brief background about yourself?

Joshua: I'm an 18 year old student just trying to get an early start on business. I've been programming for about 9 years, working with PHP for the past 4, and Javascript for the past 3. Independent freelance work has been my main focus for the past few years until recently when I developed the open-source web IM, ajax im, and now Tapefailure. I still do freelance work, but it is balanced amongst my other projects.

Allen: What is Tapefailure and where did the idea come from?

Joshua: The idea for Tapefailure started February of 2006, when I was experimenting with some Javascript – I realized it would be very easy to capture mouse movements and clicks, and then record them to a database. I developed this idea into a service that allowed people to create short "videos" of browsing and then eventually created an (what I called it at the time) "API" that allowed users to put this type of service on their site. It wasn't very successful or very well built, so I had let it fade away. I took another look at it early this year and decided to redevelop it into what it is today.

I realize the name is strange, and that sprouts from the fact that I actually registered the domain before the project, and the name is partially where the original idea came from.

Allen: How does the service work?

Joshua: The service uses Javascript to record mouse movements, clicks, scrolling activity, and keystrokes which are then passed to a script on the server which processes the data and puts it in the database. We then analyze each tape automatically, as well as allow the user to play their tapes back whenever they like.

Allen: Any interesting tape stories you can share?

Joshua: During pre-beta testing, I'd point my friend to pages with the code and let him know he was being recorded…I would go and take a look at the tape and the mouse would be drawing smiley faces or boxes using mouse clicks! Go figure.

But on a more serious note, I have to say that during testing, I found an issue on my own freelancing site. As people browsed my site, I noticed they would constantly "rehover" over my menu because the item they chose would become deselected — that was when I knew the service would be useful to others.

Allen: What's the team like at Tapefailure?

Joshua: Right now, it is pretty much me (handling programming and support) and a friend, Dave, who helps (and has helped significantly) with database and PHP optimization.

Allen: Who are your competitors?

Joshua: Mainly ClickTale and RobotReplay.

Allen: Do you have a monetization plan? If so, can you share some details?

Joshua: Right now we're just doing a simple subscription based on the number of tapes recorded, but we are definitely looking into what other options and possibilities are available. Of course, all of these ideas are still preliminary because we want to make the best choices possible.

Allen: What's coming in the next 3-6 months for Tapefailure?

Joshua: Over the next few months we will be working on adding a ton of new features from "element hover tracking" (like click tracking, but with the mouse hovering over elements) to "text selection recording" (record selected text). We are also looking into different methods of generalizing all the tapes into "sets" of information that detail a visitor's whole browsing process from start to finish, allowing the user to gain a faster and more accurate idea of what their visitors are doing.

Allen: Do you see Tapefailure as an analytics app or a testing app?

Joshua: I see it as an analytics app that can be applied to testing as well; it really depends on how you use it. It may be used to test out how visitors use a new feature or tool on your site, or just simply analyze how they navigate through your pages and design.

Allen: Do you find it harder to get the startup buzz in NY vs San Francisco/Silicon Valley?

Joshua: Definitely a bit more difficult. It is difficult to go to these Web 2.0 conferences to get your product out there more, especially as a student. I'm generally limited to web-based and some word-of-mouth advertising.

Allen: What's been your biggest lesson learned since you started Tapefailure?

Joshua: I think the biggest lesson I've learned is that you shouldn't let the competition scare you. Sure they may be bigger, or have more experience, but that doesn't mean much if you put your full effort into what you do.

Allen: What advice do you have for those thinking about starting a web application or web?

Joshua: Don't let yourself get frustrated or down. I know this is common knowledge for many, but if you feel that you have hit a wall with your project, or part of your project, put it down and come back to it later. Think about something else, go out with friends, or watch some TV; just don't let yourself get frustrated! That will put you in a downward spiral where you feel you just can't get anywhere, and it is necessary to be positive when working on things like these.

Allen: Which RSS feeds are you reading these days?

Joshua: Strangely enough, I tend to browse my news "old school" style and actually visit the sites, so I don't use feeds. Most often I read things from Digg, Slashdot, Google News for regular news, and more recently, this data visualization blog called Infosthetics which is great for inspiration when I'm looking for new ways to visualize Tapefailure data.

This evening Tapefailure has moved out of beta and into full production. We completed a video review of TapeFailure in March and the invites we had went very quickly. It received very positive feedback.

I am very impressed with Joshua Gross, Tapefailure founder as well. We have exchanged emails and IMs and he is very open to comments and suggestions and has a very open attitude.

Joshua sent over the following details regarding the pricing for Tapefailure:

Pricing starts at $8/month for the "Basic" plan. This plan has a 5,000 visitor limit, and 5 "tracker identifiers" (which are used to label your tapes for different sites or parts of one's website), and includes all the standard features (tapes, statistics, visual maps).

The other plans are as follows:
Intermediate: $18; 25,000 visitor limit; 10 tracker identifiers
Super: $48; 100,000 visitor limit; 15 tracker identifiers
Professional: $98; 250,000 visitor limit; 20 tracker identifiers

Two of the new features are: Visual mouse maps and Most Average Tape. I love analytics apps and this really is a great one. Frankly I am guessing someone will acquire Tapefailure by the end of 2007.

Here is a quote from their blog post about the launch. Really makes you realize how close we are to the creators of the products we use and enjoy.

We're quite nervous over here; we've already chewed all the way through a couple boxes of pencils! No, not really, but we are nervous, and definitely want to hear feedback from everyone.

Please report back with your feedback after giving Tapefailure a try!