feelbest.com

For online stores like feelbest.com using Sitebrand’s Segment&Serve Web personalization technology, creating more relevant experiences is comparable to having a virtual sales assistant. The concept behind Web personalization is very much based on what consumers have come to expect from the traditional brick and mortar shopping experience. Consider the traditional brick and mortar store where a sales person
easily observes and responds to variousshopping behaviors and body language. If someone is a repeat customer, they get to know their buying habits. If someone is a new customer, they work to build a level
of trust by offering helpful suggestions, sharing information about a hassle-free return policy, showing the layout of the store and so on. If someone is looking at high end fashion apparel, they look to cross-sell high end accessories. If someone is looking at sale items, they show them all the sale areas and so on.

Understanding visitor intent

But move online and suddenly the visual cues are gone. However, thanks to Web analytics, there are other cues the online marketer can follow. From the moment a visitor arrives, there’s intent to do something – to research, to buy, to register etc. And every move is monitored through Web analytics. The bottom line is that every move a visitor makes tells a story – from how they arrive to what they click and how long they stay on any given page. If someone arrives using the keyword “sunscreen”, they’re looking for sunscreen. If a Web site responds appropriately by showing a selection sunscreen and the visitor clicks to learn more or buy, they are that much closer to buying. They are being guided through the sales funnel. But if they bounce out of the site before converting – either pre-checkout or during check-out – something went wrong. And it likely relates to lack of guidance and direction from the Web site.

Responding to classic e-commerce challenges

Feelbest.com is Canada’s largest online health and beauty aid store and it faces many of the classic ecommerce challenges, including low conversion rates. As such, it is always looking for innovative ways to convert a higher percentage of Web traffic into buyers. The company also wants to recognize and respond to visitors’ geolocations; especially in terms of seasonality trends associated with many of its product categories, such as sunscreen. And it wants to create a superior, personal online experience that makes feelbest.com the online retailer customers turn to when they can’t find what they need in a regular store. The company actively encourages customers to tell it what they are looking for, no matter how obscure it may seem. The retailer specializes in finding and offering hard to find health care and beauty aid products. “If there’s demand – even from just five or ten customers – and the product is available somewhere in the world, we’ll go directly to the manufacturer and make it available to our customers,” says Darrin Pickard, feelbest.com’s sales and marketing manager.

The need to convert more traffic into buyers

The retailer’s approach is quaintly reminiscent of the corner-store owner who would get to know his/her customers one person at a time and stock accordingly. Although this personal approach to serving customers is like those of days gone by, the scale of the operation is surely different. Products are shipped to clients around the world with roughly 65% heading to the US. That’s also the nub of the challenge: motivating feelbest.com to explore an alternative strategy for further customizing and personalizing the Web experience of all visitors. The explicit goal was to find a solution that would convert a higher percentage of new and repeat traffic into sales. “Back in the late 1990s, it was easier to stake your claim as a top online retailer. But today, it’s much more competitive and you can never be complacent,” says Pickard.

Changing each visitor’s experience in real-time

Feelbest.com chose Sitebrand with its promise of superior traffic conversion to literally change the experience of every visitor in real-time while on a Website. What the retailer particularly appreciated
about the Sitebrand solution was the extensive support provided to help get up and running with a customized solution quickly. “Sitebrand is like a natural extension to our marketing team,” says Pickard. “It’s not a one-strategy-fits-all approach. What Sitebrand does is analyze your traffic and provide you with customized solutions based on your business goals and what your customers are looking for.” In the case of feelbest.com, the Sitebrand solution resulted in recommendations for the type, placement and frequency of marketing campaigns to target specific customers and boost sales in specific product categories. In Pickard’s words, “I think any online retailer worth their salt knows you can’t mass market on the Web and expect to achieve success. As best you can, you must try to speak to each site visitor as an individual. When you show people you are interested in getting to know them, they’ll show interest back. It is “Customer Service 101” and we’ve seen this with the personalization campaigns we’ve built with Sitebrand.”

Leveraging best industry practices to create smart content

All Sitebrand’s recommendations are based on industry best practices, backed by hundreds of successful implementations in similar industry sectors. This enabled feelbest.com to quickly develop “smart content” for specific customers and product categories. With the Sitebrand solution, specific areas of a Web page are allocated for the strategic placement of ads or campaign messages. Campaigns are developed around various criteria, including geo location, keyword searches, seasonal promotions, product categories, and many others.

Salvaging underperforming segments and sales

Campaigns result in a highly customized and personalized Web experience for all visitors from the moment they land on the Web site. In doing so, online retailers like feelbest.com report immediate and measurable increases in sales lift, superior click through, and higher conversion of existing traffic. “We see an immediate increase in the number of existing visitors converting in the checkout process,” says Pickard. “Seeing success around existing customers in currently established product categories makes us want to leverage Sitebrand to build campaigns around new, underperforming product categories. These new product awareness campaigns will be designed for existing and new customers,” he adds. In this way, feelbest.com will be able to test different offers with respect to new product categories. This ability to raise the profile of lower performing product categories will be designed to help increase average cart spends and total sales.

Close-up of first-time visitor promo

Recognizing a first time visitor has its rewardsTapping into underperforming segments, like first time visitors, has also proved highly successful. The first time visitor needs a different experience than the repeat visitor. They want a feeling of trust. They want to feel reassured they’re on a credible site.

When this first time visitor segment is served personalized messaging that reinforces credibility and trust versus the control group segment that receives no reinforcement, the personalized messaging always sees higher revenue per impression.

In the case of feelbest.com, the revenue per impression lift for personalized first time visitor campaigns is 207% higher than the default control group campaigns with no personalization.

Increased ROI from the feelbest.com e-newsletter

Sitebrand also provided feelbest.com with specific recommendations for print ads and its e-newsletter to create a more holistic and integrated marketing program. The company sends monthly emails to roughly 25,000 opt-in subscribers to promote the e-newsletter. Sales generated from the newsletter had begun to drop. Once the Sitebrand solution was integrated with the email program, feelbest.com saw an increase of 34% in the number of orders received within five days of the newsletter being broadcast. “Whenever we send emails, there’s an instant spike in Website traffic,” says Pickard. “It’s going to get even more interesting when we start adding more automation into the mix.” For an online health and beauty aid store, it will be a highly beneficial to trigger purchase reminder emails, i.e. “Your 90-day supply of vitamins is almost gone. Don’t be disappointed. Buy more now…”

The power of personalization for unique market segments

“The way I see it, not having a personalized approach to online marketing is like calling every one of your customers ‘Bill’. Worse yet, it’s like expecting them all to take advantage of a deal on mint toothpaste. But in reality, your customer’s name might be ‘Susan’ and she wears dentures…” says Pickard. “Website personalization allows you to find these unique market segments so you can serve up relevant offers that will convert visitors to buyers.”

NOTE: This post is also a featured article I contributed to the October 2008 issue of ”Direct Marketing” – a Canadian publication about interactive marketing and sales. Since it’s print-only (odd for an interactive pub, and apparently a website is in the works…but hey!) All that said, I felt compelled to share it online. I hope you enjoyed the read and I welcome your comments!

At Sitebrand we’re always publishing new case studies around classic e-commerce challenges associated with low conversion rates, high shopping cart abandonment rates, low average order sizes, high bounce rates, low return visitor rates, market segmentation and so on.

In our most recent case study, we dug into life online at BBCrafts.com – a leading retail and wholesale company for craft, wedding and party supplies. Sounds fun already, doesn’t it! Even though we’re talking about a very niche website with a decent amount of activity, there were concerns about cart abandonment and low order sizes.  

After approaching many potential personalization and conversion vendors, BBCrafts.com selected Sitebrand because in addition to having a proven personalization platform (Segment&Serve), BBCrafts.com was and is impressed by our consultative approach, aka our Blueprint Success Services. We have a team of pros who help customers optimize their every website move with respect to personalization opportunities.

While high shipping costs are a big reason consumers abandon shopping carts, it’s not the only reason… so BBCrafts decided to do some testing. With respect to reducing shopping cart abandonment, one of the most powerful lessons they learned is that you don’t always have to “give away the farm” using hard incentives and discounts. In some cases all it takes is a bit of reassuring and relevant information – compare this to that friendly in-store assistant at the cash. Presenting this type of reassurance and building consumer trust at the right point in the shopping process (aka the sales funnel) has proven to have significant impact. Call it that little nudge or handshake to go ahead and complete the check-out.

Using an a/b testing type approach, BBCrafts.com experimented with two different types of personalization campaigns at the check-out stage. One focused on information security and the fact that BBCrafts.com does not share or store buyer’s credit card information. In this day and age, this is a huge deal for consumers.

A second campaign simply reinforces that BBCrafts.com usually ships orders fast – often within 48 hours.  Of course a default group would see nothing.

The results? Like the headline of this post says – shopping cart abandonment dropped from 85% to 45%…

For more on BBCrafts.com success with website personalization – including how they achieved a 15% increase in average order sizes and a 160% (+) increase in return visitors, read the full case study!

As an analyst for a vendor, I get a question asked all the time that I am sure all vendors and practitioners work with when dealing with their respective HiPPOs (Highest Paid Persons Opinion – An acronym I use fondly).

“How can I prove your campaigns worked?  Show me a nice clean ROI report in my analytics.”

A good question, and a valid one too.  I wish that analytics technology had been built in such a way that it answered it easily and properly…

As a product that personalizes a web visit, Sitebrand has a profound impact on the conversion rate of a targeted traffic segment.  But what if a given client sends out a huge email blast that touches the segment we are optimizing?  What if they redo the SEO on a number of key pages?  What if they do multivariate testing on a shopping cart page?  Who gets the credit for increased conversions?  

The reporting system in our Segment & Serve product is very strong, and has a series of control groups built in to ensure the highest level of data accuracy.  However, our reports don’t take into account any other work that is being done by the customer outside our product.  And this isn’t a Sitebrand issue: with few exceptions, every other vendor in online marketing software is in the same boat.

In their most recent Web Analytics Buyers Guide, Jupiter Research states that “A resounding 86 percent of analytics clients said attribution measurement capability would be the most beneficial feature for their respective businesses.”

Most of the larger paid analytics vendors have some way of dealing with the attribution issue (Google Analytics does not), but these are still evolving, more abacus – less calculator.  So if I click on two different paid keywords for company XYZ.com over my first three visits, and then click on an internal marketing message en route to a purchase, the marketer might get some visibility into those three things in regards to my conversion.  But which was the most important?  Tough stuff.

I wanted to bring up the attribution issue for two reasons.  The first reason is that it drives me nuts, and I felt like sharing.  The second is because the concept of attribution speaks directly to one of the core themes of my blogs.  Technology will not save a marketer with a weak plan.  If you know your web numbers and have confidence in them, you can build a plan to effect significant and transparent change.

Cheers,

Jim

PS. For some nice insight into how attribution works,  check out these great videos by Avinash Kaushik and John Marshall from earlier in the year.

http://www.kaushik.net/avinash/2008/03/standard-metrics-revisited-5-conversion-roi-attribution.html

There has been a bit of concern recently in the analytics community about the new version of Internet Explorer and its potential ability to significantly increase the number of Internet users who delete or block cookies from their browser.

I read this and had a moment of panic.

For those of you who don’t know, it is just about impossible for any web analytics or optimization product to collect historical data without the use of a cookie. Without the cookie, you lose recency, frequency, and historical behavior and conversion information. For any web tool that generates reports, this is a significant loss of data and customer value. For more info on what cookies are, click here.

Why would Microsoft help people to delete information that a multi-billion dollar industry (eCommerce) needs to survive? The answer is a bit too long for this blog, but has to do visitor (and legislator) concern over abuse of personal information, as well as unfounded fears about viruses and spyware.

About a year and a half ago we had a similar fear about IE7, which would set off security alerts for every website which runs third party cookies (advertising cookies). We had to completely change the way we deploy our product to make sure that our cookie would be served as a first party one by each customer. Suffice it to say, it was not a fun transition.

So I started emailing the CTO at Sitebrand, talking to colleagues and reading the WA forum to see what other people think about the potential impact of massive increases in cookie blocking and deletion. Most of the feedback was resoundingly grim…..

Enter Jim Sterne, one of the founders of the WAA. I am going to print his thoughts on the WA forum in their entirely below. I can’t say it any better than he can, and it made me have a ‘eureka moment’: There is nothing creepy about cookies, as long as you earn the right to use them. If your visitors are deleting or blocking your cookies, you have bigger things to worry about than data accuracy.

So worst case scenario, the new version of Internet Explorer increases cookie block and deletion rates from 3% to 30% for your average website. Your cookie deletion rate becomes less of a data accuracy issue and more of a KPI on site value to visitors. It can be monitored, marketed and managed like your bounce rates.

Thinking of cookies this way, as a indicator of visitor engagement rather than a pure technical component of web analytics, is a major departure from current thinking.  It is also the right way to view cookies as we migrate towards analytics/web 2.0

Thanks to Jim for the concise (and funny) reminder that website visitors SHOULD have control, and that we as online marketers create a forum for the best application of that control, i.e. to engage MY website and business instead of my competitors.

Cheers,

Jim

(Jim Sterne Post to WA Forum, Thursday August 21st)

Giving control to customers (visitors) is always the right move. It then places the responsibility on the marketer to offer sufficiently significant value that the customer is willing to exchange personal data.

Level 0 – Cloaking device engaged

View all of our marketing materials

Level 1 – Cloaking turned off – cookies enabled – javascript tags accepted

Configure products

Stock-on-hand viewable

Use of shopping cart

Access to blog

Latest white papers available

Level 2 – Email address

Download screen savers

White paper archive

Ability to comment on blogs

Newsletter

Notification of special deals

Webinars

RSS feeds

Level 3 – Postal address & preferences

Product discounts

Special event invitations

Access to local call center

Member-only webinars

Level 4 – Answer surveys, participate in Advisory Council

Negotiated pricing

Client conferences in Aruba

Level 5 – Reveal most intimate personal details and predilections

Marriage proposal

Level 6 – Vulcan mind-meld

Resistance is futile

A little over a year ago Clare (my better half) was getting ready to buy a new car. I went to a few of the dealerships with her, and at the first place we went to salespeople assumed that because we were a couple and I was the man, I was the one making the decision. Suffice it to say they did not get a deal, and this is an example of bad segmentation that I will tackle another time…

The second place we went to, we got a good salesperson. He lead by asking who was making the decision, asked about budget, and sat down to work his deal. Clare met all the criteria of someone he can sell to. She was in the market for a new car, she had the money to make a purchase, and she was sitting in front of him talking about the Honda Fit (a wicked car btw).

Now I don’t think it’s a stretch to say that the sales director at Ottawa Honda doesn’t sit down once a month and calculate his closed sales metrics based on:
• The population of Ottawa
• Visitors to every Honda dealership in North America
• People who have gone past the dealership on bikes

He looks at his sales metrics based on:
• People who came into the store
• People for whom a car purchase was financially feasible
• People who engaged a Salesperson.

Seems pretty obvious right? Sales 101.

So why do the vast majority of websites calculate their conversion rates against all visitors to their site? Why isn’t conversion a function of the close ratio of people who could be closed at all?

Example: Many eCommerce websites in the United States don’t ship outside the continental United States, and the ones who do tend to make it so complicated it amounts to the same thing. Why do they not calculate their conversion rate against the traffic that originates in either the 50 US States, or the lower 48 that they build their business around?

I spend a lot of time in other people’s data, and I put this thought to the test through a number of analytics accounts in the last week. The changes in conversion rates are profound when you look at the percentage of prospective customers who convert instead of the percentage of all visitors who convert.

Separating the wheat from the chaff will not only provide clarity into your actual close rate, but will allow you to see opportunity. After you have segmented out the US, you might see that the UK generates 20,000 visits a month but has very low conversion. Perhaps it is time to build a cost-effective and easy UK shipping plan.

It’s not your fault as a marketer if you don’t close any visitors from Ulaanbataar, Mongolia. You don’t ship there. Clean up your conversion reports, and you won’t only make your numbers better (and more transparent), you might find your next major market opportunity.

If enough Ulanbataarians show up, maybe “Spend over $200 and get Free Shipping to Mongolia!” isn’t such a bad idea.

Cheers,

Jim