Author Archive
May 28, 2008 by Jim Cain
Per my Shop.org blog post a few weeks ago, I see an eCommerce market that is still getting ready to fully embrace personalization, but still focusing more on the plans behind inputs (traffic) and outputs (customer retention/upsells).
Based on the recent analyst reports, thought leaders in the blogosphere, and topics covered at trade events however, web site optimization (of which personalization is a part) will be the number one focus of online business in the coming years.
With that in mind, I did a Google search this morning that was a total headscratcher. Click here to see the results.
The fourth link down is a CNET article from 5 years ago explaining the failure of web personalization.
” Companies trying to get personal with their Web site visitors in hopes of increasing sales are wasting more money than they’re earning.”
The third link down is a blog post from over 4 years ago discussing the steps required to build an actionable plan around web personalization. This blog was well written, insightful, and would work just as well today as it would have in 2004 (advances in technology notwithstanding). That said, it discussed the concept of personalization in terms that imply everyone is doing it, something almost all current practitioners will disagree on.
This is my favorite quote from the piece:
“It’s a tired old yarn, but if you hope to implement a web personalization strategy, the first and most important step is to develop and mature your business goals and requirements”
Wasting Money? Tired old yarn? How can the ‘next big thing’ in eCommerce have this kind of historical background?
I blame Google. (in a tongue-in-cheek kind of way)
If you look at the writing trends in 2004 about eMarketing, you see a fairly equal split between traffic generation and website optimization. With the increased competition that emerged online in recent years, traffic generation became so important that it completely dominated. The chart below is a Google Trends report for “PPC” vs “Personalization” to illustrate this point.
PPC is in blue, Personalization is in red
So…after a brief lapse, the hottest concept of 2003 is gearing up to be the hottest concept of 2009. Everything old is new again.
Now don’t get me wrong. There have been some major changes in the personalization space in the last 5 years. There are some great vendors out there to power different aspects of personalization (I know a great company called Sitebrand), and the analytics space has matured to the point where it is possible to reap great rewards from a properly executed personalization initiative.
This couldn’t be made any more evident by the following point: The CNET article referenced above cites a Jupiter Research article from 2003 called “Beyond the Personalization Myth”. Recent reports from Forrester and Aberdeen are of the opinion that this myth has become a profitable reality.
As a personalization analyst, I would like to welcome you back to the discussion about optimizing your website and growing your business. We missed you.
Cheers,
Jim
Related Posts
No Comments | Posted in Conversion, Customer Experience, Jim Cain, Optimization, Personalization, Search Marketing, Web Analytics, eCommerce
May 15, 2008 by Jim Cain
If you talk to ten different eCommerce sites that are using Analytics, the odds are good that at least five of them are using Google Analytics. If you talk to ten different Google Analytics (GA) users, probably only one of them is using their Goal tracking for anything other than cart conversions. Long story short, an awfully large percentage of retailers aren’t getting a lot of value out of Google analytics.
This is in part due to the fact that most companies using GA don’t tend to have a full time analyst asking specific questions of the website data, and also due to the fact that while GA is great free software, there is no vendor support in terms of best practices for tool usage. (If you want an analytics vendor with a top-notch customer support/analysis team, look no farther than our friends at Coremetrics)
Here are two alternate goals, and one new way to look at them using Google Analytics. They are easy to set up and monitor, they will give you a lot more visibility into website outcomes, and will help you start asking the right questions about what you can be doing to optimize your website for increased conversions.
For additional information about how to set up goals in Google Analytics, click here.
Goal 1: ‘About Us’ page visitor conversion
If a visitor cares enough to want to learn about your business, they are that much closer to converting. Set up a goal funnel with the first page being the About Us page URL, and the last page being the transaction completed page. You now have an report that shows you the conversion rates of people who visit your ‘about us’ page as part of a session. Once you have the results in, you can start applying changes to this page in an attempt to increase conversion outcomes.
Goal 2: Micro Conversion Points
A micro conversion point is a non shopping cart transaction. Examples include newsletter signup, catalog request or wish list signup. Better understanding of how many visitors choose these micro-conversions will give a better understanding of what a visitor really wants from your site. Also if any of these micro-conversion points has multiple steps, you can build a goal funnel and look at step abandonment, just like for your shopping cart.
Goal Tip: Use filters to segment your goal results
By filtering your Google results based on different traffic source segments, you can get a much better understanding of how visitors from different sources convert for different goals. For example, what does the cart abandonment funnel look like for direct type in visitors vs. paid search traffic? Setting up funnels is also fairly straightforward, and you can see a more detailed posting from the team at Lunametrics on how to accomplish this by clicking here.
A better understanding of site outcomes equals an ability to optimize them over time. Taking the steps above will add invaluable marketing insight to your analytics tool.
Cheers,
Jim
Related Posts
No Comments | Posted in Conversion, Customer Experience, Jim Cain, Optimization, Personalization, Web Analytics, eCommerce
May 6, 2008 by Jim Cain
Looking at online sales cycles is a Pandora’s box. Once you decide to look into your analytics to truly understand what steps a visitor needs to go through in order to convert, you always end up with more questions than answers. There are a number of ways you can look at what it takes for a visitor to become a customer. You can go the:
Engagement route: Understanding how and why a visitor moves through your site in a given session towards conversion.
Recency and frequency route: Understanding separate sessions and time between sessions as steps towards conversion.
Micro and macro conversions route: Looking at how pre-transactional conversions like newsletters, wishlists and downloads move a visitor towards a conversion.
I could go on for a while with other options, but the purpose of this post was to make things even more confusing by injecting a new term into the mix: Dayparting. (place maniacal cackle here).
Dayparting is a fairly established term in the offline marketing space, used for managing media buys in radio and television. An example of this is selling radio ads against the morning daypart so that you can have the largest audience (people in cars).
Dayparting is now making it’s way into online media buys, and there are some great articles and whitepapers on the web about how to optimize your search spend based on time of day. It makes pretty good sense. Look at your conversion rate based on the hour of the day (one click in Google Analytics by the way), or even the day of the week. Look at where conversion is higher. Plan keyword spend accordingly.
Using this concept for in-site marketing makes for a very compelling case. We have a few customers at Sitebrand who run personalization campaigns based on the day of the week, but imagine if you tweaked your website so that:
- in the mornings you ran your normal site messaging, as people are looking at you as they drink their morning coffee and aren’t buying
- during lunch hours and early afternoon your site pushes your wishlist instead of a sale, because people are looking for products on your site that they will buy later at home.
- From 5-10 pm you ran aggressive sales messages, knowing that people are on the home computer and much more likely to convert.
- From 10pm to 7am you run more discounts and promos, because you might get some ‘midnight special’ bumps to conversion.
Especially if run against a control group, this would make for a very interesting look at conversion from a dayparting perspective. That is, until another way to look at online sales cycles catches my eye….
Cheers,
Jim
PS. Note that I didn’t even TRY to bring time zones into this. Neo, there is no spoon.
Related Posts
1 Comment | Posted in Conversion, Jim Cain, Personalization, Search Marketing, Social Media, Web Analytics, eCommerce
April 22, 2008 by Jim Cain
The blogosphere right now is abuzz with discussions about the current economic downturn and what impact it will have on eCommerce sales, and the associated budgets that eMarketers will be given.
Forrester research sees eCommerce bucking the downturn trend, with online spending actually going up by a predicted 17% in 2008. That said there will still be a tightening of belts as far as spending is concerned, with most of the spending freezes taking place against in-site technologies. Search will continue to grow, as it is both vital to online sales and the most easily provable from a tactical ROI perspective.
According to our friends at Marketing Sherpa, while “38% of marketers are spending more on online tactics, such as paid search, natural search, email marketing and online events” this growth is happening at the expense of in-site technologies as “B-to-B and B-to-C marketers are conserving budget dollars by renegotiating contracts with vendors and agencies”
So let me get this straight. Online stores are spending more money on getting people to walk in the front door of the store, and less money on making sure they walk out of the store with a product in hand. Doesn’t make a lot of sense to me, especially when you listen to an online brand that has a keen grasp of the big picture.
I listened to a great speech at Shop.org by Patrick Byrne, COO of Overstock.com. It would take a series of blog posts to detail all the interesting things that have been done/are being done by the Overstock.com team, but two major points stood out:
- While they came late to the party regarding natural search optimization for Google, they have made search a number one priority over the last few years.
- They are in the process of rebuilding their marketing programs, customer service programs and website experience around 5 pre-defined visitor segments.
They recognize that while increased eyeballs on website equals increased revenue, the only way to break the vicious cycle of search reliance is though optimization.
I know that most of the people reading this don’t have Overstock.com budgets, but try a simple exercise.
- Pull up an analytics report on the conversion rate for one of the product categories that has received budget for keyword buys and search optimization.
- Look at the percentage of traffic and conversion rate for the traffic segment.
- Compare the conversion rate of this type of traffic to your site average and if it is lower, calculate your potential revenue opportunity.
There is a lot of money that gets left on the table by making traffic generation of any type your primary marketing mechanism. You should be budgeting time and money to optimize visits within your site as a logical component of your search marketing initiatives, especially if you want to take advantage of the 17% growth of the only growing retail market.
Cheers,
Jim
Related Posts
No Comments | Posted in Conversion, Customer Experience, Jim Cain, Optimization, Personalization, Search Marketing, Web Analytics, eCommerce
April 11, 2008 by Jim Cain
Just got back from the Shop.org regional in Scottsdale with two new insights. The first is that Arizona is MUCH nicer than Ontario as far as weather is concerned, and the second is the purpose of this blog post.
eCommerce marketing should not be treated the same way as traditional marketing.
There is no implied insult in this statement, rather a suggestion that there are much more tangible revenue numbers associated to eCommerce marketing, and in order to be truly relevant, this sales/conversion component must be directly addressed or risk being ignored.
At Shop.org, I sat in on great sessions and panels, had in-depth conversations with some really talented professionals in the eCommerce space, and watched really closely to see what statements and ideas were the most engaging of the overall audience. This was especially easy in the main ballroom sessions, where you could glance at the room and see when people were engaged and taking notes, or bored and playing on their blackberries.
Invariably, the pens came out when the topics were about selling, whether that topic was repeat sales to existing customers, or new customer acquisition through search, affiliate, or email.
Conversely, looking around the room when the topic was pure marketing/branding I saw a lot of people playing brick breaker on their phones. The same with my one on one discussions. Highest interest was always on growing sales and conversions.
If you are reading this as an eCommerce marketing professional, you are rolling your eyes right now. “OF COURSE it’s all about growing sales!” But think about the statement and then answer this question.
Outside of eCommerce, name an industry where people with pure-marketing jobs have defined sales quotas or to use industry parlance, conversion goals.
I can’t think of one, and I have been in sales and marketing for almost a decade. Every other industry has marketing as demand generation (funnel filling) with a separate sales mechanism in place for deal completion. These marketers still have goals and accountability, but they are not the owners of direct revenue targets.
Only in ecommerce do marketers have the job of filling the sales funnel, and then moving people through it. If you can not grow your conversion rate, you will not keep your job. (place Glengarry Glen Ross quote here….”AIDA!”)
In the attached image (Thanks to Darryl and Eric from the Sitebrand marketing team) you can clearly see the additional functions an eCommerce marketing team owns that would traditionally be the domain of a sales department. Click on the image to expand.
Having spent a lot of time in enterprise sales, I can definitely tell you that pure marketing is important and valuable but is secondary to hitting my quota. The same feeling was evident in watching the attendees at Shop.org.
So here is my big Arizona eureka moment, which I encourage people to think about (especially if you plan on speaking at events that eCommerce professionals are attending).
eCommerce is an emerging discipline, and as an industry we need to be sharing best practices and new ideas. Just don’t forget that if it doesn’t help grow sales and hit conversion quotas your ideas will end up at the bottom of an ever-expanding to-do list.
Related Posts
No Comments | Posted in Conversion, Customer Experience, Jim Cain, Optimization, Personalization, Web Analytics, eCommerce
March 28, 2008 by Jim Cain
The purpose of this weeks blog was going to be around the concept of the first time visitor segment and how it can be appropriately marketed to on a given web property. Once I hit the 2000 word mark, I decided to keep going and make it a whitepaper. More will follow on that in coming weeks.
Instead I would like to take a moment to throw in my two cents on a real challenge that I think most eCommerce properties face in regards to site optimization: horsepower vs proper planning.
The purpose of most of my blogs and of the upcoming whitepaper has to do with proper execution, building the right marketing plan to effectively achieve the right sales outcomes. While Sitebrand’s technology will assist in achieving better outcomes, it will not in itself give you a good plan.
This thought runs in parallel to the ’90-10’ rule regarding analytics software coined by Avinash Kaushik. If you spend 90% of your analytics budget on the right people who can make the right plans, and 10 percent on the technology to allow them to do their jobs, you will get the best possible results.
So what do I mean about horsepower vs proper planning? While eMarketers are starting to take more control of the web site, overall ownership is still firmly in the domain of the IT department and the CFO. This is for two reasons:
- Websites, despite all the talk of ‘commerce 2.0’, are still considered to be black boxes where traffic is plugged in one end, and revenue comes out the other.
- Most marketers have built their plans around this ‘drive traffic-manage customers’ model, and don’t have a sound strategy or plan for insite marketing.
If your site is viewed as a black box, or intelligent catalog, or online sales engine, the only way to optimize it is to bolt on ‘more horsepower’ in the forms of self learning software like merchandising management, site search tools, and multivariate testing engines.
With this thought in mind, re-read the websites for some of the vendors who have cold called you in the last few weeks. While the stated audience will be the marketer, the true stakeholders are finance and IT. The value propositions tend to revolve around the concept that the primary job of the marketer is to get good traffic to the website, and the primary job of technology is to manage the experience of the visitor through self learning databases and patented algorithms.
“More power!”
I could sell a LOT of software to Tim the Toolman. I can also guarantee you that while his website would get the basic value that any good software product will provide, the ‘sales engine’ would never be shifted out of second gear due to the lack of a solid plan.
There is no point in buying a Ferrari to be a grocery getter, and there isn’t much point in putting more horsepower into your website before you have a cohesive plan around how you are going to use it.
Cheers,
Jim
Related Posts
No Comments | Posted in Conversion, Customer Experience, Jim Cain, Optimization, Personalization, Search Marketing, Web Analytics, eCommerce
March 19, 2008 by Jim Cain
Clive “the Coupon Clipper” doesn’t just want value for money; he wants to know he did everything possible to get the best deal. He will come to your site because of online or email promotions, and every page needs to reinforce the concept of getting more and paying less. Create a series of documents around what makes Clive tick, and how that effects how he shops and surfs. Draw a big picture of what you think Clive looks like and put it in your office. Bring that picture to marketing meetings, and make sure Clive is represented on every page in your site.
Umm what?
In a typical eMarketing team, headcount is low, and each one of those heads is wearing five hats. Email, search (paid and natural), merchandising, promotions, copy, design……all these things are owned by, and need to be executed properly by – 3 people in many cases, 1 person in many more.
Note that I didn’t put analyze data in the job description above. This is for two reasons:
Reason one: Most eMarketing groups barely scratch the surface of their respective analytics package.
Reason two: eMarketers don’t use analytics because they are super busy and analytics has no direct perceived value.
In my last blog I ranted a little about how there are too many “What” and “Why” bloggers in eCommerce and very few who focus on “How”. Here are a few thoughts on how to take the concept of personas and make it worth your while to squeeze them into your already full day.
Step One: Find a traffic segment
Don’t worry about building Lucy “the Listmaker” or Stevie “Gen. ‘Y’” level personas. Open up your analytics, blow the dust off, and play around in the reports. Take the last six months or year of data, and start looking for segments of your traffic that are:
· Larger than 10% of your total traffic
· Converting lower than your site average for the time period
A great example is the “First Time Visitor” segment of your traffic. Every analytics product can give you the conversion rate of your first time visitors, and they tend to represent 60-80% of your traffic, and convert anywhere from .5%-1.5% lower than your site average.
Step Two: Use some persona marketing best practices on your segment
Use some of the persona building techniques for this basic type of visitor (for more info click here. You can even give the segment a cool name “Fernando First Timer”. You are quickly going to see a lot of things that you can add or change on your site (messaging, design and layout) that would have a positive impact on this segment.
Now if I was a member of an overworked 3 person marketing team, I would see a ton of value in altering my site to speak to a large and underperforming segment… the analytics package has already showed that money is being left on the table. Building a strategy to go and get it, providing it doesn’t become a full time job, is nothing but goodness.
More best practices to follow in a subsequent blog on first time visitors. In the meantime, feel free to post comments on Persona building in general, or segments you have identified and optimized in the past.
Cheers,
Jim
PS. Happy St. Patricks day!
Related Posts
No Comments | Posted in Conversion, Customer Experience, Jim Cain, Optimization, Personalization, Web Analytics, eCommerce
February 27, 2008 by Jim Cain
A little background on today’s post. There are many smart people in our space who are starting to answer the “What” and “Why” questions of eCommerce. I therefore spend a lot of time wrapping my head around questions that start with the word “How”. Three specific instances in the last two weeks have had me asking a lot of “How” questions about the concept of Personas.
- Jim Sterne gave a fantastic presentation with the WAA in
Ottawa two weeks ago and spent some time talking about the concept of personas. In order to have a truly relevant website, he feels that you need to understand who the primary personas of your target market are, and ensure that each page of your site has something for each of them. I asked a question to a subsequent speaker regarding the difference between personas and segments. For more details check out this great Bryan Eisenberg article called “Getting the most out of your Personas”
- I spent all of Monday evening skipping back and forth between the NHL trade deadline and an insightful article by Theo Papadakis on Occam’s Razor about engagement profiling and metrics. He states that while visitor engagement is important for marketers, web analytics can only provide relative degrees of engagement (and even those can’t be proven to be positive or negative). This post created a ton of great comments, and blog crossover to a post by Joseph Carrabis on Web Analytics Demystified about Engagement….I could barely concentrate on my hockey following that thread around…
- I am working with a Sitebrand customer right now to build programs to optimize the overall conversion of high level segments. We are going to be using our product to create an ongoing dialogue with certain types of site visitors through every step in the conversion funnel. So I have spent a few weeks talking about nothing but segments and personas and engagement…
A lot to chew on huh? To break all the above into one sentence: The age of actionable analytics is upon us. Clicks and pageviews and basic reports are now on the shelf beside my betamax player and Commodore 64. All of the thought leaders in the analytics space are focused on using data to support marketing initiatives like persona and engagement optimization, as opposed to doing marketing to fix/resolve issues found in data.
So here are my list of “How” questions, which my customers and I are starting to answer together with the Sitebrand tool (shameless plug):
- How does a typical eMarketing team cost justify the bandwidth required to build personas?
- How can engagement be used (and optimized) as a tactical element of every emarketing initiative?
- How can a properly executed segmentation initiative feed into (and out of) persona building and engagement mapping?
Generic questions for sure. But once we can start mapping actionable and tactical best practices to these “how” questions, we can earn the right with our internal budget holders to ask more questions down the line. As always feedback is appreciated, specifically on the ways you have asked and dealt with “how” questions in your organization.
Related Posts
1 Comment | Posted in Jim Cain, Optimization, Personalization, Web Analytics
February 15, 2008 by Jim Cain
It has been a while since my last blog post…and the reason is pretty good. For the brief time I have had this forum available to talk about eCommerce, it has always been (at least to me) a chance to vent about what a salesperson sees in the market, and provide a unique perspective as well as some education on the ‘whys’ and ‘hows’ of conversion optimization.
I forgot people internally would read the posts…
Flash forward to January, where I was approached by management to take charge of growing out and continuing to productize our business support program, the Blueprint to Success.
I was more than interested, but having been been in sales for 9 years, it still felt like the scene in Lethal Weapon where I was asked to give up my gun and badge. “Cain, you’re being suspended from sales. Leave your phone and forecast on the desk!”
Then I realized the truth of what this job really entails. I am being given the chance to perform the ultimate solution sale: getting an entire industry to “sign on the dotted line” on changing the way they market within their respective websites.
Now that sounds like fun.
In my new role as Senior Analyst, I have three goals:
1)Take all the great knowledge from within Sitebrand and our customers, and continue to build out a truly productized go to market program so all of our new customers can quickly get into the process (and the results) of conversion optimization
2) Work for ‘Eureka’ moments with all of our customers: There tend to be specific moments in working with Sitebrand where a customer says “Wait, you mean I can do X? Without help from a developer? And I can prove it worked?” A big part of this job will be helping people have, and then replicate, insightful and profitable
Eureka moments.
3) Create a closer dialog with the market at large about the concepts of Conversion Optimization, Personalization and Actionable Analysis.
The moral of this story is that while I may have given up a quarterly revenue quota, I took on a new one in regards to selling success into our customers, and selling the industry on Sitebrand’s take on Actionable Analytics.
Looking forward to working with you.
Cheers,
Jim
Related Posts
No Comments | Posted in Jim Cain, Optimization, Personalization, Web Analytics
January 14, 2008 by Jim Cain
As the focus of many eMarketers starts to move from raw traffic generation activities to in-site marketing optimization (and it’s about time), it is a perfect time for us to have a look at the ‘vendor sandbox’ around web personalization. By Sandbox, I mean the category of technology and service providers that can be commonly grouped around a particular discipline. ‘Analytics’ is a good example. “Do you have an analytics vendor?” ……is a reasonable question to ask. “Do you have a ‘Personalization’ vendor?”…… is not.
It has only been within the last six months that we have taken most of the commonly used terms in the analytics space to the point where there is an agreed upon set of definitions (Web Analytics Association Standards Committee – http://www.webanalyticsassociation.org/en/cmt/?5).
Some of the site specific terms we use as eCommerce marketers are fairly straightforward. Product Recommendation software for example, has a fairly self-evident role on a website (people who bought that also liked this!). Analytics, need I say more.
Certain terms are less clearly defined, but still pretty obvious to wrap your head around: Word of Mouth (WoM) marketing as an example. Customer driven content in the form of reviews, uploaded video, pictures etc.
Personalization? At least as it relates to internet marketing, it’s a term about as nebulous as they come….·
Is it based on explicit user action? (Think customizing your Yahoo or Google homepage)·
Is it based on implicit customer need? (Customer buys radio, you show them batteries)·
Is it based on the expectation of implicit need based on profile? (Customer behavior indicates they are lost mid-visit, so content is predictively displayed with customer service/sales incentive options)
(My eyes crossed just writing that - this post is getting very “Zen and the Art of Website Maintenance”) Here’s my two cents on the issue. Personalization is not an industry sandbox, it is an objective. It can only be measured in qualitative terms. Want to know if your website is personable? Start interviewing your customers. Looking for a personalization vendor? Make sure you look at the ways in which you would like to be personable and then select a vendor accordingly.
My next few blog posts will describe in more detail the three ways of making your site more personable, with pros, cons and better descriptors so that we can advance beyond the term ‘personalization’. Feel free to post any thoughts to the comments about either your experience with, or interest in, making a site more personable.
Related Posts
2 Comments | Posted in Jim Cain, Personalization, Web Analytics, eCommerce
