Finding out what users REALLY do…

Leave a comment

We’ve had endless debates internally about what customers want to do with our site and what works or doesn’t work with the layout of the pages. We have tried things that work in traditional marketing such as focus groups, surveys, analytics, etc. Unfortunately, what people tell you they WANT to do doesn’t always translate into what users REALLY do when they are on a site.

One of our eminent advisory board members (thanks Wendy) spoke to us several months ago about the importance of tools such as heat maps to see what goes on with the site. We finally found a great tool that is beginning to expose real issues and opportunities in building a site that users will really Heat Mapuse.

CrazyEgg (www.crazyegg.com) is a great tool that allows web site managers to visualize what users do when they visit your site. Heat maps tell you by color where clicks happen. Here is an example – this is a landing page for Antiques. What you’ll notice is that there is a tremendous amount of clicks on the search box and go button on the right, on The Gallery and My Home on the left, and nothing on the orange Take the Quick Tour button in the prime location in the middle! Obviously this is a page that we are going to redesign!

Another great feature is called Confetti – which gives you detail by click about what users do. They show things like time to click, operating system, browser type, search term, and other important facts about user behavior and profiles.

The best thing about CrazyEgg is the cost and the ease in implementing the tool. CrazyEgg starts at FREE and is priced based on need up to $49/mo. We are using one of the lower pricing tiers at iTaggit and it’s working great for us. The best part is that we can ramp up our pages and tests as we need to on demand.

Implementing the tool is easy. It’s a line of code on the page that needs to be watched. It took us an hour to implement the tool and immediately started to see results.Confetti

I’m not in the habit of trying to push someone else’s product, but this one is a great tool for any web marketer’s arsenal. Let me know if you find this tool useful. I’d also like to know if tips on tools like this is worthwhile too.

Customer Acquisition Costs

1 Comment

Here’s a real challenge – what makes up the true acquisition cost for a customer in a user-generated content site?  And what constitutes a ‘customer’?  Since revenue on sites like ours is driven by advertising and affiliate product purchases a ‘customer’ can be either a registered user (who can contribute content) or a visitor (who just looks at the content).

Customer Acquisition cost is made up of 5 major contributors in a business like ours – Pay Per Click advertising (Google, Yahoo), Banner Advertising, SEO Optimization, Viral Marketing Campaigns, and community management (headcount driven).  But these costs can drive both registered users and visitors.   Here are the relevant metrics from my view:

1) Cost per Registered User – monthly aggregate marketing spend/# of registered users

2) Cost per Visitor – monthly aggregate marketing spend/# of Unique Visitors

3) Cost per Action for Pay Per Click advertising

4) Cost per Action for Banner and Viral Marketing

These are interesting metrics to track and trend, but they beg some real questions.  How do we define an ACTIVE registered user vs. just a registered user?  Does it matter?  How do you factor in the value of items added as compared to number of registered users?  Is one user that adds 200 unique/interesting items (which positively affects SEO and visits) more valuable than a user that adds 2 items (or none, but blogs)?   Is there a way to segment registered users in a way to spend more efficiently to get the high value registered users?  Is that a smarter spend than just driving activities to get visitors who may drive more purchases and ad views/actions?

If you think about this question, it has important ramifications for most web 2.0 companies.  I’d be interested in understanding how others approach this question.

Closing loops..

Leave a comment

One of the challenges that web 2.0 sites have (and iTaggit also has run into) is tracking and quantifying ‘mashup’ activity. If the value proposition includes linking to other sites and providing cross-site benefits then it’s probably important to be measuring the amount of activity in this area. For example, we have an Amazon link which allows users to download info from Amazon on specific items and includes a BUY FROM AMAZON button.screen-capture.png To understand how important and/or how well this works for users it would be important to be able to track how often the features are used. And, if the user abandons the activity, where and why. The problem is that with mash-ups and cross-site integration you can only use analytics to see what’s happening on YOUR site. What happens if there is something funky going on in the other site and a high percentage of users abandon the activity?

Additionally, as we add more features like the ‘What’s it Worth To You’ feature in partnership with WIW2U.com, we’ll need to start pruning off the mash-ups that are not effective or actively used. Once a user clicks on a feature we can count that, but does that mean the user completes the activity? Are they pleased with the results? We usually don’t know because at that point the user is ‘living’ on our partner’s site and infrastructure. They’ve gone through the teleporter and we won’t be able to see them again until they return to us. We’re spending time and energy trying to understand how to ‘close the loop’ on these cross-site activities. As we discover more things around this challenge I’ll share them here. If others have ideas or have been dealing with this issue I’d love to hear about it.

Web 2.0 challenges traditional marketing approaches

1 Comment

The traditional four P’s of marketing championed by Phil Kotler of Kellogg have been a cornerstone for many years. Product, Pricing, Placement, Promotion. At least TWO and maybe three of the P’s are much more difficult to manage given the changes driven by the internet and specifically Search Engine Optimization and Search Engine Marketing.

Pricing is not as straightforward and direct as it is for traditional bricks and mortar (and even web 1.0) businesses. There are multiple levels to get to the paying customer. The revenue is much more indirect than in any other business. For example, if a web 2.0 site is dependent on advertising revenue how do you control the ‘P – Price’? Yes, once you get to a certain size you can demand higher CPA or CPC rates, but you are still dependent on generating traffic to show the ads and even then your ability to drive the conversion rates necessary to recognize the revenue is semi-passive at best. And this is AFTER you have reached a traffic level where you can even set a price at all. Is this truly ‘Pricing’ in the traditional sense? Instead of focusing on the traditional levers of COGS (cost of goods sold), Gross Margin targets, and variable sales and marketing costs the web 2.0 marketeer has to focus on traffic, eCPC (effective CPC) rates, Click through rates, and ad priorities. And where the site actually generates direct revenue (such as subscriptions or premium features) there are additional metrics and drivers that the marketeer needs to focus on that are far outside the realm of traditional marketing.

Placement also has a very different meaning in the web 2.0 world. There is no shelf space (relatively) with the exception of search engine rankings and banner advertising locations. Banner ad locations also present a challenging difference. Does Banner Ad location represent Placement or Promotion? A/B testing is so critical in web 2.0 marketing! Unlike traditional advertising A/B testing where the marketing professional is testing both creative and media in the web 2.0 world the marketeer is testing location (placement) on page, media, specific site results, and creative. All of this with very questionable and inconsistent metrics between different sites and tools.

So, what does this mean? In my mind we’re beginning to see a new paradigm develop. Marketing will be much more dependent on analysis of metrics such as click through rates, exit rates, and time on site. Rather than being able to run focus groups to find out what customers ‘want’ the web 2.0 marketeer is going to have to be able to do in-depth statistical analysis to understand how the customer ‘want’ is translated into actions and web development.

Deepak Jain, the Dean of Kellogg’s School of Management, has a Masters in Mathematics and Statistics and is a top professor of marketing. I had a class with him early on and he spoke about the increasing integration of math and statistics into the marketing process. I didn’t get it at the time because I was in a traditional business. Now I do. He’s right and it’s moving faster than even he suggested.

Are others finding similar things? I’d be interested in hearing your thoughts. We’re beginning to look for more quant-friendly marketing interns here. Are you? Mail me at daltounian@itaggit.com.

Recognizing Actual Net Worth

Leave a comment

Many people ask me about the genesis of iTaggit and why I would create a site for ‘things’.  The question usually is positioned based on the perspective of the person that is asking the question and it veers from one extreme to the other.  For example, a serious collector of antiques asks why we’d create this site when Antique collectors are historically not web friendly.  Others, who have a lot of clutter, ask why it’s relevant to them.  Both sides miss a critical point which is the germ of the iTaggit idea – that’s that each of their holdings represents a significant asset which is either likely undervalued or not valued at all.

The reality is that people’s net worth is made up of three major baskets (sorry for everyone who is sick of hearing this from me):

1) cash and marketable securities  – easy to value, easy to track, everyone is interested in them, etc.
2) large value items  such as homes, cars, boats – a little harder to value, need intermediaries to sell, etc.
3) everything else you own – very difficult to value, hard to track, really hard to connect to others with interest.

Creating a set of tools and an eco-system of services is what I view as our mission.  Helping users convert their items to true assets is our overriding goal and mission.   It’s not only about organizing or selling items (which are both important aspects of converting items to assets) but also about protecting the value of your assets.  Here is a blurb from a recent US NEWS AND WORLD REPORT on managing ‘clutter’:

“Everyone should inventory his or her possessions in case disaster strikes. Right. “it’s the unusually prepared person who has done an inventory,” concedes Michael Barry of the Insurance Information Institute. But Hurricane Katrina and the California wildfires are spurring people to action. Getting ahead of disaster is as easy as walking around the house with a camera. Then send the pics to a safe place outside the home. That bare minimum can help reconstruct belongings after a fire, flood or burglary.”

My mission is to help users optimize their net worth by helping them recognize their true net worth, whether they want to convert items to cash (sell) or just organize and value what they have.  As the economy and the equity markets come under more and more pressure people are going to need to be able to harness the value of everything that they own.  I think it’s an important cause and continue to try to drive iTaggit towards providing a valuable service for our users.

Location Dependence (independence?) for Emerging Tech

Leave a comment

There has been a recurrent question in discussions that I’ve had recently with potential investors, other web executives, and partners. That question is, ‘does an emerging tech company need to be in the Bay Area to have a chance of success?’?

There are two sides to this issue; the first is that there are several advantages to being in Northern California. There is an abundance of funding sources out there and they all seem to prefer to invest in the area. There are also a great number of tech companies already established out there so the opportunity for partnerships may be easier due to location. Finally, because there is a rich business base in tech there is an incredible amount of services and resources that are competing for mind and dollars of the tech community (note SEO/SEM consulting firms – there are only a few in Texas so they are in high demand. In the Bay Area there are so many that they are fighting for clients).

On the other hand there is also a tremendous amount of competition BY the tech firms for technical and marketing employees so wages are high and it appears that turnover may be a challenge there. The area also acts as a fairly ‘small town’ so secrets are not so easy to keep. For example, last week I was at a popular Woodside breakfast place and it was very telling to see who was meeting with whom.

So… what does this mean for us at iTaggit? Well, over the next few months we are going to attempt to build a stronger corporate presence in the bay area while maintaining our base of operations in Austin. We have a fantastic, dynamic workforce and we need to develop the tech infrastructure for Web 2.0, 3.0, 25.0, etc. right here. We believe that we can combine the best of both environments to deliver a better solution for our users and a lower cost of operations for our investors. We’ll see how it goes and I’ll keep you updated as best that I can on my blog.

Quickly adding items to iTaggit using the Amazon Add Item Feature

Leave a comment


One of the most daunting tasks for new users of any website that uses user-generated content is how to get stuff in. In the case of iTaggit the question is ‘how do I quickly add items to my account’? The ADD ITEM USING AMAZON wizard comes to the rescue!

When you go to add an item to iTaggit you are presented with the screen above. You can add items individually or by loading in a group of pictures, but the quickest way is to choose the IMPORT FROM AMAZON.COM option.

Most people think that Amazon only has books, movies, and music but they are WRONG! Amazon has one of the most comprehensive databases of items available. You can find almost anything! In the search below you can see that I was able to import information on a Tailgate for a Ford F250 pickup!

Once you have added the item you then can put it into either a collection or your wishlist. The really cool thing about this feature is that it adds a Buy from Amazon link into the record so if someone sees this on your wishlist (or in your collection) they can click on the link to buy a similar item.

Try out the feature, it’s a great way to QUICKLY build your wishlist or personal collections.

Using iTaggit to post items for Sale or Trade

1 Comment

iTaggit has a great feature in that items that have been posted for sale or trade show up in a public collection titled ‘items for sale or trade’. Any item posted for sale on eBay through iTaggit or put up for trade using the ‘Trade This Item’ item action automatically get posted to the public collection. This collection is available to search engines so items posted in their also may get picked up by the search engines. This increases the visibility opportunities for items that you are trying to sell or trade.

This feature is free to users and expands the number of people that are likely to see the item. Try it – it’s easy – and FREE!

Using iTaggit to add depth to your Craigslist listing

Leave a comment


There is neat utility in iTaggit called ‘Publish’ that allows users to generate HTML code about an item stored in iTaggit and to allow embedding in other web apps.
This feature is used today to add items to myspace.com and other cool sites. I just listed my Jeep on Craigslist and was able to use it to do the same thing.

This means that you can essentially add multiple pictures, video, links, and documents about your items to your listing by using the html generator. What you do is use the Publish menu option, select and copy the html code, go into your craigslist listing, and paste it in the description field. Below is what you will get: