There’s theory, and then there’s practice. If you’re like most Web design professionals, you’re interested in the theory but focused on ideas that you can put into practice right now.
For Web design and user experience professionals seeking such ideas, Internet Retailer’s Web Design and Usability Conference in Orlando was the place to be last week. Sessions were full of hands-on, actionable information – so much that 97% of attendees say they’ll use the information gained in sessions such as site critiques with the e-tailing group. Below are some of the best ideas we heard during an amazing week.
Get them to the Payment button fast – but don’t lose sales along the way. According to Monetate’s Eric Miller and Envelopes.com’s Laura Santos, there’s a basic optimization problem here that’s often overlooked. Once a customer hits ‘Add to Cart’, a retailer must balance laying out a clear path to checkout with collecting accurate payment information and creating opportunities for additional sales. And like any optimization problem, the right way to solve it is to test rigorously and make data-based decisions.
Design for the new reality. Betsy Emery from TELLUS Web and Vijay Murali of Nutrisystem created a great deal of buzz with the idea of designing for mobile and tablets first. Many mobile initiatives start with making existing content available via mobile channels. That means that designers view the small screen and the touchscreen primarily in terms of constraints rather than opportunities. By beginning with a clean slate and designing for mobile first, a retailer or brand has the opportunity to take full advantage of those opportunities.
Make the most out of your interactive assets – including (of course) video. Office Depot’s Shannon Wu-Lebron spoke about video alongside Invodo CEO Craig Wax and made several key points. Being strategic about placing video calls to action in the right place yields a higher video view rate, getting more out of your video content. Video is often thought of as valuable for complex products, but ideo can also drive benefits for simpler products – even, as Wu-Lebron pointed out, printer paper. There’s a full writeup on the session here.
There were many more learnings as well. Chief among them – we’re already looking forward to next year!
Happy Monday! As a member of the Fun Committee and the Community Impact Team, I wanted to re-cap a great week at Invodo. In addition to working hard and celebrating new clients we brought on this past week, this past Thursday a handful of Invodians channeled their inner Dirty Harry (Clint Eastwood) and converged on Red’s Indoor Firing Range in South Austin, Texas where we target practiced with a variety of guns. We only shut down the range twice as we accidentally shot the pulley system that moves the targets forwards and backwards. A large man there muttered that we should be shooting smaller guns. While our aim was off at times, we had a blast!
Invodo is enjoying great success and we have amazing and talented people joining us on a regular basis. On Friday, we enjoyed a New Hire Happy Hour at Invodo South aka B.D. Riley’s on 6th Street to get to know our new Invodians. It’s a great opportunity to learn more about the fascinating people we work with—where they came from and what makes them unique.
On Saturday morning, ten Invodians braved northern breezes and temperatures in the mid-thirties (cold for Texas!) to volunteer with Habitat for Humanity. We installed soffit, framed windows and installed battens on a new home. It only took 2 hours for 2 Invodians to frame one of the windows perfectly. We worked alongside future Habitat homeowners who have to volunteer 300 hours before they are eligible to purchase a Habitat home.
With plans to add about 50 more Invodians in 2012, we’re working hard to stay connected to each other and our community. Check out our open positions and join in the Invodo fun!
Market Development Associate
Enterprise Sales Director
Software Development Engineer (UI)
Senior Software Engineer (Java)
Marketing Program Manager
Video Producer – (Dallas)
Have a great week!
Travis Kenney; Market Developer & Community Impact Team Leader at Invodo
We had great attendance on yesterday’s “Captivating Consumers through Cross-Channel Video” webinar.
Lauren Freedman of the e-tailing group debuted results from a survey into how over 1,000 consumers use video during the shopping process. Judging by the questions attendees asked, interest was high and so was the level of sophistication of the attendees.
We couldn’t get to all the questions during the session, so I’m answering a few more here.
Q: Is buying video production services ROI-positive for an etailer?
A: Yes.
(pause)
OK, sounds like you want more detail. Our clients consistently report realizing ROI based on the business goals they set.
ROI is based on deciding which needle you want to move and why, and on how much you spend to move that needle. For Bizchair.com, increasing sales with video was a key goal. They found that the 13% of their site visitors who viewed video accounted for 33% of their sales. Stacks and Stacks reported increased sales with video as well. Living Direct had increased engagement as a key goal, and realized that with a 9% increase in time on page. And, as we learned in the webinar, consumers consistently say that quality matters. That’s why our clients choose to invest in well-produced, but cost-effective,video production.
Q: What has been the reaction from online retailers to the idea of category videos versus product-specific videos?
A: There’s certainly value in both. It’s attractive to leverage your video investment across a category at times. That said, I think the place to start is with your business goals. Analyze your site traffic and customer base, then do the math to determine where you can have the most impact. Product-specific videos are attractive in that they should pay off in clearly incremental sales, based on conversion tracking and A/B testing. That makes the business case easy to build. However, if you’re converting well on the product page but not doing as well in moving consumers down the funnel, you might want to start at the category level.
And – final note – a video gallery can be a great way to leverage your investment in product-specific videos. One video, two use cases, two locations. And if you do it right, you can index for category-level search terms in the gallery while indexing for product-specific search terms on the product page. That brings the right searchers to the right pages, making the most of your video SEO.
Q: Do you have data showing consumer preference between third party production versus company generated video content?
A: Yes. Stacks and Stacks found that Invodo-produced, retailer-branded video content drove higher conversion than manufacturer-supplied content. We also see that in data from Video Ratings and Comments, where consumers rate Invodo-produced, company-branded video as more helpful than video produced by other sources. And, as we discussed on the Webinar, companies that have traditionally used highly informal videos are recognizing the benefits of emphasizing good quality sound and lighting.
Q: What do you think about YouTube as a search engine as opposed to just a video hosting platform?
A: YouTube’s the second largest search engine – bigger than Yahoo! or Bing. It’s a great place to be found. Your video strategy should include leveraging your video assets across channels, including YouTube. I recommend using socially optimized edits for YouTube and Facebook. Reason: on a product page, your goal is to drive conversion – to get the consumer to click the “add to cart” button. But in social distribution, your goal is to bring that consumer to your site. That may require a shorter edit and an explicit call to action, such as “learn more at www.invodo.com.“
Q: What’s a good source of video statistics and information?
A: As Lauren said on the Webinar, she searched quite a bit and didn’t find anything close to a definitive resource. We publish a lot of information here on the Invodo blog and in our newsletter (kindly subscribe, if you haven’t, and thank you!). I frequently see journalists and bloggers linking to our video statistics page. We keep it updated with both Invodo and third-party information.
Did I miss a question? If so, email me at russ
invodo
com
and I’ll be happy to respond.
We’ve recently received coverage in Mediapost, Retail Customer Experience and Website Magazine around a study we completed with the e-tailing group. Some of the findings seem to have struck a nerve in the industry – especially around what is sometimes called “the 30-second myth.”
OVPs (Online Video Platforms, the technology-only side of the business) are fond of publishing viewing time metrics showing how long viewers watch before clicking away. They then characterize that as insight into consumer behavior and urge retailers and brands to create shorter videos – regardless of the message to be conveyed or the quality of the content.
The problem, of course, is that their approach assumes that all content quality is the same, all product categories have the same level of complexity to communicate, and all audiences are equally engaged. Pretty big assumptions. That’s why we commissioned the research as a consumer-facing study. Rather than guessing about consumers based on clicks, we thought we’d directly engage them to find out what makes video content valuable to them. By combining that research with data mined from our network of close to 100 major brand and retail sites, and tempering that with insight from researcher Lauren Freedman, we hope to form a more complete picture.
Here’s the cool thing: we haven’t even formally released the study, and it’s already generated quite a bit of industry dialog. The findings on video length are only a small part of what we learned. We’ll have a better industry discussion when we can put that data in context. We’ll do that, with Lauren Freedman’s help, on Wednesday, February 8th during a Webinar, “Captivating Consumers through Cross-Channel Video.” We hope you can join us!
“Plans are worthless, but planning is everything. There is a very great distinction because when you are planning for an emergency you must start with this one thing: the very definition of “emergency” is that it is unexpected, therefore it is not going to happen the way you are planning.” — Dwight D. Eisenhower
When I started at Invodo about 1 year ago, the primary hosting for applications was on physical servers in a datacenter here in Austin, TX. There were plans to move to an infrastructure that allowed better failover and geographic diversity, but work had not begun for the actual transition. All components required to provide service to end-users had been pushed out to the content delivery network, so the risk of the application infrastructure impacting clients and end users was very low. That made the relative urgency of accomplishing the move lower than other improvements to our technology at first, but beginning in 2011 we had a strategic goal of improving our system uptime/availability to 99.999% before the holiday season. After I joined, priorities aligned and we decided the time was right for the transition. Two of us then set out to make the move happen.
Reasons Behind The Move
While end-user service is all pushed to the content delivery network, several components of our infrastructure are necessary for managing that service and providing the full service offering to our customers. What this means is that a complete failure of our infrastructure should have no impact on end users, but could be seen as a degradation in service for our customers. Things like data analytics for the period of the outage and access to the tools we provide (such as to push new content or retrieve reports) would be unavailable to our customers.
One thing I have learned over the years is that no system is perfect. Any system that is supposedly incapable of failure will have a problem at some point and, since it is impervious to failure, it will take a significant amount of time to recover the system. Instead of trying to build the perfect system, we planned for failure. The primary goals of the new architecture were:
1. Focus on mean time to recovery (MTTR) for all systems so any outage is reduced to the minimum time possible.
2. Customers should not be impacted in any way unless it is a catastrophic failure of multiple systems, and even then, it should be the minimum impact possible.
3. Any given system can fail at any given time. Any individual server or system going down should be expected and ANY recovery necessary should be quick and straightforward, independent of what caused it to fail.
4. The architecture should easily survive the next 5 years, meaning how and when systems were to scale with increased usage had to be decided out of the gate.
There’s more than one provider of cloud-based hosting, each with their own focus and strengths. After reviewing providers in the context of our goals, we found that Amazon Web Services was the clear choice for how we wanted to accomplish them.
First Few Days On Amazon Web Services
Our first few days on Amazon Web Services were exceedingly scary. We had tested all of the components, we had verified our monitoring systems, but there were still lots of unknowns. Would the new infrastructure perform the same was as the previous one? Was our security model sufficient to provide access to all services? Were there any components that needed to communicate that we overlooked? The end result was a very smooth transition, even done in a way so that our customers wouldn’t notice a change. Before we were fully transitioned, Amazon Web Services had a major outage in one of the components. Numerous large scale companies ended up having issues because of the outage. However, thanks to our new infrastructure design, there was no noticeable impact for our customers. This helped provide validation that we were going in the appropriate direction for our architecture.
Extended Period on Amazon Web Services
We have now been on Amazon Web Services for quite a while. There have been a few more outages of one type or another on Amazon’s infrastructure and we have had multiple components fail and need to be recovered, all without noticeable customer impact. That is all great for normal operations, but what about the peak retail usage that starts just prior to Thanksgiving and ends just after the new year? In one day, we tripled the capacity of each individual server and doubled the number of servers that would be involved in handling the additional load. And the entire holiday season was reasonably uneventful (from a technical perspective). Looking back, had we been on our previous infrastructure, it might have required a lot more work to keep up with demand. We did not have the ability to rapidly scale up our infrastructure, we did not have the same detailed monitoring on our infrastructure and our ability to recover from faults in the infrastructure were significantly lower than they are now. In the final analysis, the uptime goal of 99.999% of critical services during the holiday season was achieved, in large part do to the migration to AWS.
Our previous infrastructure was not poorly designed. In fact, it was the previous design that allowed the move that we did in the period of time we did it. Everything about the move did not go according to plan, but through hard work and ingenuity we managed the transition without negative customer impact. Looking back, the take away is that what worked for you yesterday will not work for you tomorrow and that planning needs to happen today.
Matt Brace, Invodo Lead Envisioneer
invodo
com | What We Do | Products & Services | Resources | About Invodo | Contact Us | Social    |
Thanks for your interest in Invodo! Please fill out the short form and we'll contact you. If you're in a hurry, you can call us at 1-800-280-4122.
We hate spam as much as you do. We won't rent, sell, give away, or otherwise abuse your email address and other information
