Conversion Rates Revisited – The One Percent Solution April 30, 2008
Posted by debbiepascoe in accessibility, high performance site analytics, privacy, quality, search engine optimization, usability, web analytics, web design, web standards.Tags: accessibility, cart abandonment, conversion, engagement, optimization, pivacy, quality, search, traffic, usability, web analytics
2 comments
Download the Conversion Rate Calculator
In my last post, I noted that Coremetrics has begun releasing benchmark data collected from their ~300 clients. A couple of the stats really caught my attention:
- The typical conversion rate is 3.29%
- Conversions Where Site Search Was Used – 14.84% of consumers used site search during their visits – conversion rate 5.60%
- The shopping cart abandonment rate is 68.42%.
Wow….think about it….
- out of 100 people, only 3 of them actually complete the desired action. AND
- out of 100 people that begin a shopping cart, only 32-33 of them complete the transaction and buy something (through the website – we don’t know if they walked in the store later and concluded the sale there); AND
- of the people that use internal search, they convert at a much higher rate than the ones that do not.
While some of my esteemed colleagues have decided that it’s not about conversions, it’s really about engagement (I’m not naming names, you know who you are 
), it really is about conversions. That’s the bottom line. For every small improvement a site owner makes to improve their conversion rate or reduce their abandonment rate, they get repaid many times over. The philosophical diversion into “engagement” is really code for “we know people are leaving the site and completing the sale offline, we just don’t have a way to tie all the data together”. When that day comes, conversion will come back into vogue in a huge way.
Take this rate that Coremetrics has given us as a benchmark. If this conversion rate of 3.29% can be improved by one percent, sales would increase by 30%. That translates to real money. Similarly, 68 of 100 people walking out of your virtual store when they have things in their basket that they abandon in the last aisle represents real money that didn’t make it into the till. Everything that can be done to chip away at that represents real money.
So, how do you identify those improvements? Strip it back to its basic elements; look at what you’ve created – how usable is it, how findable is it, how free of defects, does it respect the visitor’s privacy, is it accessible to all potential customers? Are the key pages that lead people to the conversion event optimized? Do they have the right stuff in the right places with the right call to action to propel people forward and keep them moving forward to completion? You’ll notice these questions don’t have anything to do with studying how people have reacted to your site; rather they have everything to do with understanding deeply what you’ve given people to interact with. Traffic is a measure after the fact – it’s forensics. Evaluating traffic is great for understanding what people did; it is not a predictor of what they would do if things were different.
To help you visualize the impact that small improvements in conversion rates and abandonment rates can have, I’ve created a “conversion rate calculator”. I am not an accountant or finance expert. This is not complex econometric modeling. This is just a simple way for you to plug in some numbers that are meaningful to you to see that the impact over time is real and measurable. Have fun, dream big, and see what it might mean
Here’s the link to the Coremetrics Benchmark page.
Big Week for Web Analytics News April 14, 2008
Posted by debbiepascoe in search engine optimization, web analytics, web design.Tags: abandonment rate, benchmarks, conversion rates, coremetrics, indextools, search, web analytics, yahoo
add a comment
Two big announcements in the past few days:
Coremetrics
Coremetrics just secured $60MM in new funding, then followed that up with an announcement that they will begin releasing benchmark data collected from their ~300 clients. Their announcement follows Google’s announcement about releasing benchmark data. Some interesting stats from Coremetrics:
- 22.41% of visitors left retail sites after viewing one page
- 51.65% got as far as a product page
- Conversions from Direct Load Traffic – 47.89% of traffic and 67.35% of sales came from visitors who typed in the retailer’s URL or clicked on a bookmark. The typical conversion rate was 3.29%
- Conversions Where Site Search Was Used – 14.84% of consumers used site search during their visits – conversion rate 5.60%
- The shopping cart abandonment rate was 68.42%.
Here’s the link to the Coremetrics Benchmark page.
Yahoo! and IndexTools
Second Yahoo! announced its purchase of IndexTools – start at Bob Page’s blog post to get the industry take on what this means. Plenty of pundits have weighed in on what this means, and most of them are linked from Bob’s post. The one post worth listing here is Eric Enge’s 10 Cool Things you can Do with IndexTools. If you’re not familiar with them, this post gives some info about the solution and its reporting capabilities.
Eric Peterson suggests that “this (acquisition) is potentially the permanent game changer”. So here we are – three giants – MSN, Yahoo! and Google:
- each providing search results,
- each selling ads,
- each with mass volumes of email account holders and other member-specific areas,
- all of which result in vast volumes of data about what people do online.
Before this acquision, two of the three had web analytics capabilities, and now they all have it.
On a related note, last week I ran across a blog that Avinash wrote over a year ago – Five Ecosystem Challenges for Web Practitioners – still as relevant today as when it was written. In it he talks about the fact that web analytics is not a silo – web visitor data is tightly related to the “upstream” (tv, magazine, newspaper, radio ads), and the “downstream” (phone sales, retail outlets, other sites).
We’re all watching to see what’s going to happen to the large pure-play companies – WebTrends, Omnture, Coremetrics – each with its own expansion strategy, and the multitude of smaller players and even new entrants into the space.
UPDATE – 4/15: Dennis Mortensen posted to his blog today that Yahoo! is making IndexTools free to all existing partners and client if they accept Yahoo!’s Standard Agreement.
Top Four Characteristics of the Optimal Web Team February 13, 2008
Posted by debbiepascoe in accessibility, content management, high performance site analytics, quality, search engine optimization, usability, web analytics, web design, web standards.Tags: accessibility, optimal web team, optimization, search, SEM, SEO, web analytics
1 comment so far
It doesn’t take too much reading and talking to people to recognize just how fragmented the various web management related disciplines are within organizations. This was recently reinforced by the results of the two surveys I did – Where in the Organization is the Web Analyst and What in the Organization does the Web Analyst Do. I also ran across a survey over the weekend that was done by the Internet Strategy Forum(PDF doc) that mirrored my findings in a lot of ways.
This is not a criticism – in fact quite the contrary. Even though the web has been around for what seems like a long time now – can you remember how you did certain things before the web, like locate addresses on a map, look up a phone number or make travel reservations – the “web” organizational structure has not yet reached equilibrium. It is still evolving, the moving pieces are still being identified, and people are working to get their heads around how those pieces fit together. With that as the backdrop, here is my list of the top four characteristics that the Optimal Web Team will have.
1. The Optimal Web Team will be multi-disciplinary. Currently, web analysts are gaining in numbers and growing as a discipline. SEO/SEM people may be in the same work group, but there are equal odds that they are in a different work group. Content contributors are scattered through the organization, and the people managing the content management system, if there is one, are in another organization altogether. The Optimal Web Team will be a multi-disciplinary team, where people with specific expertise will work closely and regularly to make decisions based on a 360 degree view of the complexities that impact data quality, site quality and compliance issues.
2. The Optimal Web Team will report to a senior level executive. Organizations have come a long way from “the web as IT…thing” to “the web as mission critical”. Increasingly processes are performed by employees, vendors, investors, customers, and prospects via web-enabled pages and forms. Even so, organizations have not yet fully recognized their web investment as equal to land, labor and capital. It is that important, and the management of it must be at a level in the organization that reflects that importance.
3. The Optimal Web Team will have a bigger seat and a louder voice. The current state of fragmentation results in people reporting into lower levels of the organization and despite investments in tools and training, the expertise that these people are developing is often not heard at a level within the organization that can effect changes as a result.
4. The Optimal Web Team will manage what it doesn’t directly control. Organizations everywhere are employing web analysts and people with SEO/SEM expertise. At the same time, other disciplines such as voice-of-the-customer (VOC) surveys, A/B and multi-variate testing are frequently best done by organizations that do it for a living. Content management will continue to loom large and touch many parts of the organization as the number of contributors continues to grow.
The Optimal Web Team will consist of people with specific expertise sharing knowledge across disciplinary lines, leveraging existing market expertise when it makes sense and coordinating requirements like content contribution and defect correction that are, of necessity, dispersed throughout the organization.
Judge Allows Accessibility Suit to Proceed October 4, 2007
Posted by debbiepascoe in accessibility, search, search engine optimization, web analytics.Tags: accessibility, lawsuit, search, search optimization, Target
add a comment
To update a previous story, a federal judge in California has certified the suit brought by National Federation of the Blind against Target, and rejected Target’s motion for summary judgement.
The progress of this case will be closely watched. If NFB prevails, the issue of accessibility will extend past federal and state agencies and become much more of a concern to corporates than it has been to date.
Corporates shouldn’t wait until the outcome of the case, though, to put accessibility on their radar. A more accessible site is also more searchable, more easily indexed by major search engines, more likely to be found and visited. The more visited it is, the more “traffic” it has, and the more opportunities to generate a lead or a sale, allow customers, employees and investors to self-serve, thereby reducing calls and the costs associated with them. It’s not a simple one-for-one equation – an organization that takes the approach to accessibility as a point problem with a point solution is leaving money on the table.
As with so many things in the web space, accessibility is not an isolated, unrelated-to-other-site-characteristics issue. Some people would have you believe that the only reason to pay attention to accessibility is because it impacts one discrete group of stakeholders and the only reason organizations have done it to date is as a defensive move – to avoid lawsuits. I reject this view, and prefer to advocate a view that doing “the right thing” has broader benefits to the organization.
The essence of synergy is producing a whole that is more valuable than the sum of it’s parts. Accessibility is a “part”. Search optimization is a “part”. Transitioning from call centers to online self-serve is a “part”. Utilizing techniques and best practices that address these parts will result in a larger benefit to the organization as a whole.
When all you have is a hammer…… September 5, 2007
Posted by debbiepascoe in ads and ad spending, mobile web, search engine optimization, usability, web design.Tags: javscript, mobile device, search, web access, web analytics
add a comment
An article got my attention today that reminded me….when all you have is a hammer, everything looks like a nail. Marketers seem to be approaching through-a-mobile-device-access and through-the-computer-web-access as if they are the same, when in fact they quite different.
The article – A Thorny Issue: Detecting Mobile Search Click-Fraud is a very interesting piece that lays out several challenges marketers have in their desire to push adverts out to people on their mobile devices, and focuses the discussion around the issue of preventing click-fraud.
The author identifies numerous ways in which the mobile internet is nothing like the web, and still looks optimistically toward a solution that would treat them as if they are the same. Among the differences, as pointed out by the author:
1. Trackability – mobile devices have a limit on the number of cookies they can set – the article mentions 4
2. Trackability 2/Usability – mobile devices don’t process javascript – oops -this is the cornerstone of modern web analytics
3. Trackability 3 – mobile devices aren’t tied to IP addresses the way computers are
And a couple of my own:
4. Usability and web design – there is simply not very much screen “real estate” on mobile devices – making stuff smaller is not a good option
5. Search – given the way mobile phone plans work, the more precise and onpoint the search result is the more likely it will be seen – how many pages will people view if they know they are incurring data transmission charges?
The real question is whether pay-per-click or pay-for-performance are even appropriate for the mobile world. What are the options? One that comes to mind, in thinking about this from the consumer’s perspective, is this: give me the ability to find the closest restaurant, ATM machine, gas station, hotel, post office, UPS or Fedex/Kinko’s store, police station, library, airport, then give me the directions to get there, and while you’re at it give me their hours of operation, and a phone number to call in case I get lost. If a marketer can sponsor all or part of an entire search vertical and the bits I need in real time to act on it, I might be very inclined to look favorably on them and their products. If marketers sponsor set volumes of “serves” in a defined vertical, they’d worry less about click fraud and more about building market share by giving people something they actually need and will use.
Of course, this option is back to the “one-to-many” approach, so it seems to fly in the face of the one-to-one holy grail being pursued with such vigor on the web.
How marketers would track ROI with this approach is anybody’s guess. The point is that mobile access is a different problem requiring a different set of tools. Throw out everything you thought you knew, and start from what you actually do know. The “thorny issue” article has some great observations about mobile internet access, and why it – and the challenges of marketing into it – are so different.
And to take the vertical/local search topic a step further, I’m waiting for the day when maps are beamed into vehicles in place of existing GPS systems that charge $200 a pop to update the maps. BMW, in collaboration with Google, has taken the lead position on this – can other auto manufacturers be far behind? Marketers will have a lot more screen real estate to work with than on the mobile phone or PDA.
Bottom, line – I pay for my cell phone and everything that comes to or emanates from it. Give me what I want, not what you want me to have, and I’ll be your friend for life.
