ROB WEBB

E-commerce has undergone dramatic change in the last two years.  Zynga has cracked social gaming and virtual goods wide open, producing enormous transaction volume.  Groupon has crushed local service group buying and created a multi-billion dollar cash machine.  Gilt has pioneered invite-only branded high end flash sales.  All three of these companies have pioneered new e-commerce models in a matter of months, have been locked in the tech news spotlight, and have spawned hordes of fast-followers trying to get in on the action.

However outside of the consumer spotlight, there are several companies that have been revolutionizing web based e-commerce tools available for businesses.  Shopify, AvantLink, Shopatron, AdRoll, and FeeFighters are each dedicated to helping their customers sell more and sell better online.  They don’t receive the public fanfare of their consumer facing brethren like Zynga, Groupon and Gilt, but all have offerings that create unique value and some are growing at a similarly furious pace.

Shopify, based in Ottawa, is a dead-simple yet very robust hosted do-it–yourself e-commerce platform.  Largely bootstrapped, the Shopify application makes it incredibly easy for anyone to set up an e-commerce storefront.  Whether you are selling car mustaches or golf cart rims Shopify can get it it done. Users have a broad range of controls over the look, feel and function of the stores they create.  Inventory tracking is integrated and the Shopify application store has an ever increasing number of useful tools from 3rd party providers which can be integrated into any Shopify store.  There are currently over 60 applications, including QuickBooks integrations, user review platforms, email marketing integration, and to drop ship service integrations.  E-commerce platforms have been around for a long time but Shopify has made it easier than ever to get up and running.  A notable difference from past customizable e-commerce platforms is thatShopify is fully hosted so merchants don’t have to worry about servers or databases.  Shopify is also incredibly cheap.  The entry level package is $25/month with a 2% transaction fee.  Bigger packages scale down transaction fees and up on monthly fees.  Shopify prices out drastically cheaper than unhosted competitors like Magento ($250/month unhosted) and Miva Merchant ($60/month unhosted).  Centrally hosted, Shopify can move fast and update their platform at any time, allowing them to move forward more effectively than their installed competition.  Shopify definitely a company to keep an eye on over the next couple of years.  They are certainly very profitable today.

AvantLink, based in Park City UT, is a innovative affiliate marketing technology platform.  I have been working in and around affiliate marketing for some time and firmly believe that affiliate marketing will be massively disrupted within the next five years.  There is huge room for innovation in the space and AvantLink is leading the charge.  Online stores use affiliate marketing platforms to track people that buy on their sites and pay commissions to affiliates that send them good traffic.  Affiliate marketing platforms provide web stores with a platform to host their affiliate program and provide 3rd party cookie-based tracking technology that monitors site traffic and purchases.  Merchants then allow publisher sites to join their affiliate program and pay affiliates a set percentage for converting traffic.  As an example, you can see Liftopia’s affiliate program landing page on AvantLink here .  I have worked with AvantLink on the merchant side with Liftopia and could not be happier with their tools or their service.  I have also worked with AvantLink from the publisher side with GearBurger.  In both cases have I not only incredibly happy with the tools and service AvantLink provides, I have also been shocked at how antiquated the incumbent players’ tools are.  Simple tasks like creating links to merchants and getting inventory datafeeds are at times literally impossible with the larger players in the space like Commission Junction and Share-a-Sale.  These affiliate incumbents are firmly entrenched, with large pools of merchants and publishers. They aren’t innovating much and are very difficult to work with.  In the next five years, someone is going to come along and blow up the affiliate marketing status quo.  Google has been making moves in the space recently but their tools and services are still not up to snuff.  AvantLink spun out of Backcountry.com several years ago and has been moving fast ever since.

Shopatron, based in San Luis Obispo CA, helps manufactured product brands balance branded e-commerce sites and brick-and-mortar distribution.  I have written about Shopatron in the past.  Shopatron solves a unique problem for brands that want to sell their products to consumers directly on branded company sites while at the same time selling through brick-and-mortar distribution partners.  A company like Smith Optics wants to sell sun glasses at Smith.com but also sell in sporting goods stores around the globe.  Shopatron links distribution partners with the brand e-commerce site and gives resellers the opportunity to fulfill nearby orders placed on the brand site.  This is not a problem that seems obvious to an outside observer but is truly a large one.  Brands want to sell as much product as they can, capturing online traffic and maintain strong relationships with brick-and-mortar distribution.  This is where Shopatron helps.  Shopatron is growing like a weed, with a historical focus on the outdoor industry but recently branching into many different manufactured product verticals.  Interestingly, Shopatron recently announced a partnership with AvantLink that will provide Shopatron powered brands with an AvantLink powered affiliate marketing platform.

AdRoll, based in San Francisco CA, helps merchants advertise to customers that visit their site but leave before making a purchase.  This type of advertising, known as “retargeting” or “remarketing” is somewhat controversial and has privacy advocates up in a tizzy.  Sites that retarget place cookies in user browsers upon arrival to their site.  These cookies allow for ads to then be targeted at that user elsewhere on the web after they leave.  If you have noticed that you consistently see banner ads from a couple of companies, you probably visited their site in the past and your are probably being retargeted with ads.  I find this to actually be an enjoyable experience because these ads are from companies that I’m actually interested in and I’m happy to have those companies vying for my business.  Given that only 2% of traffic to an average e-commerce makes a purchase, retargeting allows merchants to follow up with the 98% that leave without buying.  Retargeting has been around for nearly 10 years but only recently did large ad networks like Google start allow for cookie based targeting within their display networks.  This change swiftly brought retargeting mainstream and companies like AdRoll have been running with the ball ever since.  AdRoll is unique in that it allows merchants to easily cookie all the site visitors and then target them in the most efficient manner possible.  This is accomplished in two primary ways.  First, merchants can use AdRoll to reach their customers across many different display ad networks like Google, OpenX, AdBrite and Yahoo simultaneously.  This equates to more available pageviews and lower costs to the merchant.  Second, AdRoll places cookies on merchant sites using a “smart pixel” which dynamically reads the URL of the page a user is visiting and places cookies accordingly.  This is a must have for a merchant trying to grow multiple retargeting lists on many different pages.  The alternative to AdRoll is hard coding multiple pixels into the bodies of specific pages: a hugely manual, error prone process that is difficult to modify, particularly for a bigger site.  Disclaimer: I used to work at AdRoll.

FeeFighters, based in Chicago IL, helps merchants reduce the transaction costs associated with doing business online.  FeeFighters is the youngest company of this group, launched only a few months ago, but they are tackling a huge problem.  Negotiating the waters of transaction processing is a huge headache.  I have written about this issue in the past.  There are so many layers with so many different types of fees it is nearly impossible to determine the best course of action for your business.  The FeeFighters platform allows merchants looking for transaction processing services to put their business up for bid, forcing transaction processors to fight for the business.  I hope that FeeFighters will find a way to keep transaction processors honest and force them to disclose all elements of a bid as part of the process.  The headaches I have gotten in the past when dealing with transaction processors is that there is not “one” rate that you will pay to process a transaction.  Depending on the customer, merchant and card type, some transactions are “qualified”, some are “non-qualified”, some are high risk, etc.  The fee that is often pitched is not the fee that you will end up paying.  For FeeFighters to be effective they will need to find a way to force transaction processors into structured disclosures that are easy to compare against others.  Transaction processors are very comfortable with paying on a CPA basis and pay fat fees for signups so a transaction processing comparison shopping site like FeeFighters could be a great way to capture these referral fees.  I hope FeeFighters can pull this off but if they can’t I sure hope someone else can because it would eliminate a lot of headaches.

Who did I miss?  Who else is shaking things up with innovative web commerce tools?  Let me know in a comment!

I’ve been helping a friend write a business school case on his company. I haven’t done too much but it’s been fun to see the process. Recently I helped with an appendix covering the ROI calculation when dealing with online advertising campaigns. I thought I would include it here in case it might be useful for other folks.

Calculating Online Advertising Return on Investment

The end goal of any advertising campaign, online or off, is for the campaign to generate more profit than the campaign costs to run. When a campaign reaches the elusive state of generating more money than it costs, it has become “ROI Positive” and is generating a positive return on the investment made to run it. In the online advertising realm, it is possible to track ROI from a nearly infinite number of campaigns and channels concurrently. This sort of tracking is far more difficult or impossible within offline channels and campaigns like billboards, TV, print and radio advertising. A key aspect in calculating true ROI from an online campaign is attributing visitor conversions to specific campaigns and assigning a value to each conversion. In the e-commerce space, a conversion is usually defined as a purchase, and the conversion value is generally the profit generated from the purchase. The definition of conversion and the value assigned to them will differ for different types of sites. For example, a social network might define conversion as a signup that does not deactivate within 1 week. Without conversion tracking, campaign analytics could be limited to how many people see an ad, click a link or land on a site. In these situations, the ROI calculation must use assumed conversion rates, which is very risky practice because traffic from different campaigns will likely have drastically different conversion rates (e.g. SEM traffic will convert at a much higher rate than traffic from a rented email list). Finally, true ROI calculations must include post-conversion factors that impact the value of the conversion like exchanges, refunds, churn, etc.

(Value from Campaign) – (Cost of Campaign) = Return from Campaign

A heavily underutilized tool in the UI development and testing toolset is the paper prototype user test.  Paper prototype tests have literally saved me and the companies I have worked for hundreds of hours and tens of thousands of dollars.  The test itself is really simple.  Like all other forms of user testing, the most difficult step is getting the testers to show up and do the test.

A paper prototype test is just like an on-screen user test, except instead of using a computer, mouse and keyboard, the user looks at mockups, uses their finger as a mouse pointer, and uses their mouth to type.  The test administrator plays the role of “computer” and presents the appropriate mockups in front of the user, depending on where they “click” with their finger.  These tests are phenomenally effective at testing navigation and design on real users.  In my experience paper prototype tests are just as good as on-screen tests as catching areas in need of improvement with the added benefit of being able to improve on and test new designs throughout test day.  The process allows you to quickly and cheaply iterate on designs and test them without ever touching code or expending development resources.

The book Paper Prototyping: The Fast and Easy Way to Design and Refine User Interfaces is the definitive resource on paper prototyping but it is several hundred pages and honestly it is overkill. If you have read Don’t Make Me Think (and if you haven’t, read it today) and you are comfortable conducting on-screen users tests, you can absolutely conduct a paper prototype user test.

This past summer we worked through a massive consumer facing redesign of the Liftopia website and paper prototypes played a significant role.  After the internal product team iterated on the designs and got them as good as we thought we could, we put up signs around our building calling for testers, printed out our mockups (complete with dropdown menus), taped out a “monitor” area on our conference table and went to town.

I was able to jump into Fireworks halfway through the test at lunch, make several modifications that after the first few testers was glaringly necessary, then reprint and test the modifications on the afternoon testers.  By the end of the day we had our mockups fully tested and could confidently move them on to implementation.

If you have any experiences with paper prototype user testing I would love to hear them.  Fire away in the comments.

Over the last year I have been inspired for a number of reasons to bone up on my skills related to improving employee performance.  This great post by Ben Horowitz led me to High Output Management by Andy Grove, which is a phenomenal reference for any manager. I have found Chapters 11 through 16 to be particularly useful.  Grove’s core message is that more often than not, an underperforming employee is likely being mismanaged.  Specifically, it is likely that the employee does not know what management expects the employee to do.  When an employee underperforms, Grove advises managers to double back and make sure it is abundantly clear what exactly is expected of the employee, what metrics will be used to evaluate the employee’s performance, and when the employee’s next performance review will occur.

At Liftopia, I have found this method to be extremely effective in managing the performance of both contractors and full time employees.  New contractors and employees now receive an expectation document tailored to their team and role which outlines what will be expected of them in regards to communication, project ownership and quality, and self awareness and honesty.  Depending on the team the employee is joining, the expectation document will include specifics around usage of communication and tracking tools that are relevant to the role, such as Salesforce, Trac, Basecamp, etc.  New employees and contractors also receive the form that will be used to evaluate their performance and are told when their first written performance review will occur.  Together, these two documents make it abundantly clear what is expected of the employee and how and when they will be reviewed against these expectations.  I have found this system to be extremely effective.  Of course there will always be employees that don’t work out, but if you have a system like this in place you can be sure that the mistmatch was not a miscommunication around expectations.

In case anyone might find them useful, I have attached a generic Development Team Agreement and Development Team Performance Review documents.  Both are generic enough to be adapted to fit any division within a company.  If you have any resources, tips or thoughts on managing expectations or evaluating performance, I would love to hear them.  Fire away in the comments.

Gear Burger has been turbocharged.  You can read about the improvements here.

A much needed update!  I joined Liftopia back in November as VP of Product.  My primary role is to lead the charge on the technology development side of things but as is the case with all startups, everyone does a bit of everything.  I joined the team right as we were heading into our peak season and have spent the last six months trying to pump out features that would improve our season results and keep the wheels on the bus.  We still have a long way to go but we had a great season, and when we have a great season that means we were able to help ski resorts have great seasons as well which is very cool.  We’re now heading into a period of lower transactional volume which means we will be able to crank out a ton of development work, as we don’t have to worry as much about bug fixes, transaction processing optimization and customer support issues.  I’ll post updates here as we release features and design changes.  I’m also going to start blogging again (although Twitter is still sucking me away).

For the last several months, we’ve been undergoing a large redesign of a critical and complex user flow at Adroll. At this point, several hundred hours of work have gone into the project. Our UX intern started the project with a company-wide customer identification exercise that got everyone focused on who we’re targeting with our product and who we should be designing for. We then conducted extensive user testing with target customers throughout development which resulted in radical shifts from what we thought would be the “right” solutions out of the gates. The result is not only a better and easier to use product, but also a unified team. User testing not only focused our team on our target users’ behavior but also virtually eliminated internal debate and the resulting tension regarding who thought what would be best for our users.  In the long run it’s very possible that the side benefit of team unity will match the value of the much improved product.

My buddy Chad and I launched Gear Burger a while back as a little side project. When you sign up for the service, you select your preferred gear type and gear brand preferences (e.g. men’s hiking and skiing gear made by Patagonia, Salomon and Volkl).   Gear Burger then monitors one-deal-at-a-time sites like Steep and Cheap and Tramdock and sends you email alerts when an item that matches your preferences goes on sale.  The sales we see are generally 10 – 20% cheaper than the cheapest sale prices on the web.  You can also browse past deals within the catalog where you can also find individual products that you would like to follow and get alerts when they go on sale.   The live deals section contains all of the deals we are seeing that are currently active.  To date our system has seen 21,740 deals and tens of thousands of alerts.  We find it super useful for snagging great gear deals and we hope you do too.

I recently met a scrappy entrepreneur who is working on some very interesting technology that has potential to help large enterprises. He described his product as “ten times better and ten times cheaper than the competition.” It’s easy for entrepreneurs to slip into the mindset that their product needs to beat the competition on all fronts: price, service, technology, etc. The truth is that if you build something that really is 10x better than the competition, there is no reason to massively cut the price as well.

It’s been a busy couple of months.

1. I started as Director of Product Management at Adroll
2. KnowledgeBid is cranking in a nicely automated manner
3. I’ve been posting a lot on twitter
4. Semi-regular posting will resume/continue…