For my last round, I was supposed to share the results from that 300% sales explosion on the leather dog collar with nameplate page, but the folks at Google decided otherwise with the Panda update targeting content farms. Panda bit a few ecommerce sites pretty hard, so I’ve been helping out some new client-folk sort out the mess. So, today we’re getting the Retail Smarts column back on track.
In my first installment, I told you about how I spent less than thirty minutes optimizing a single product page, which resulted in a 300% explosion in sales on that SKU.
That column pressed the right buttons for some folks. I got emails from vendors wanting me to optimize their product page next. The collar’s manufacturer emailed to thank me for cranking sales. Got some Twitter love. I even got an email from a direct competitor thanking me for some new ideas. Doh!
To recap, back in November of 2010. I cloned a popular product page, and named it what normal people would call it instead of using the manufacturer’s brand/model/name just to see what would happen.
After doing some lightning fast keyword research, I chose the generic product keyword: “Leather dog collar with name plate” as my primary keyword phrase which would also be the name of the product for this test.
To optimize the page for this phrase, I did the following:
- Made the URL leather-dog-collar-with-name-plate.html
- Wrote a decent Title tag and Meta Description
- Made the Headline “#1 Best-seller: Leather Dog Collar with Name Plate”
- Wrote 500 words of keyword-rich content
- Linked to it on both the subcategory page and category pages
- Linked to it from a couple of external domains
This was originally intended to be an SEO project, but it turned into an attempted conversion improvement project as well. Sometimes it’s hard to draw the line between what is search engine optimization and what is conversion rate optimization.
What content am I adding for SEO benefits, and what content is to educate and compel my prospects to open up their wallets and buy? When there’s a conflict between the two, these days, I tend to do what’s best for conversions!
Today, we’re going to look at the results. In this column, we’ll cover:
- How much SEO traffic drives sales
- What SEO keywords generate revenue
- How additional content in product descriptions helped
- What tools I used to measure results
- How you can apply all of the above
I’m most excited about what this experiment taught me about subcategory pages. At the end of this column, I’ll disclose what’s really responsible for at least half of the success of this increase. This revelation has changed the way I tweak pages on all my sites using Yahoo! Web Analytics.
Traffic Is Up 600+%, Sales Are Up ~300%
Five months before the experiment, the old SKU generated $6000, and in the following five months, the NEW SKU + OLD SKU generated $17,500. Sweet! The ~300% increase is still holding up after 5 months.
In the past 6 months, the NEW SKU and the OLD SKU combined got more than 10,000 unique product views. During the previous 6 months, the OLD SKU alone got less than 1,500 unique product views, so traffic to these pages is up over 600%.
I’ll Admit It.. Some Mistakes Were Made
One thing I promised was to show “how much better the new page converts than the original page”. Well… uh, it doesn’t.
You know how the more organic traffic you get, the more bad, non-converting traffic you get as well? The NEW SKU and the OLD SKU now both convert less than just the OLD SKU by itself, which makes sense, if only because the same product is featured with two different names.
We also had a much higher bounce rate on the Leather Collars subcategory page, which also got a boost in traffic after we changed the content. To be candid, our site is a hunting dog store, and some dog owners don’t like sites associated with hunting.
A third problem is that we keep selling out of stock! You can see by the graphs that there are gaps in sales. I’d rather lose online sales than tick people off by taking their money then putting them on back-order. We remove sold-out sizes on the product page as soon as inventory is gone, which is killing me!
Where Is All This Traffic Coming From?
The boost in traffic is completely organic — a combination of Google with a little Bing & Yahoo! thrown in.
Some of the sales come from Google Adwords. I didn’t specifically buy an ad for this NEW SKU, but some more generic keywords drive traffic to pages where both the NEW SKU and the OLD SKU are featured. We also got a smidgen of sales from Google products feed, but not enough to comment on.
To my surprise, we also got a lot of internal traffic from other pages on the site. More on this in a bit…
Overall SEO Efforts Of Old SKU vs New SKU
Since November 2010, 113 of my NEW SKU orders tie directly to SEO efforts compared to 18 SEO sales for OLD SKU for same time period before the experiment. Don’t get too excited! All that means is that I was doing a horrible job as an SEO on this product page.
In the past 21 months (since I was running the new version of Yahoo! Web Analytics) the OLD SKU page only had 1079 organic clicks from 83 different keyword phrases, of which seven phrases which converted.
In the first 5 months — 1372 organic clicks from 200 unique keyword phrases, 25 of which converted.
SEO efforts targeting one phrase worked pretty fast, too. Before we launched NEW SKU, we accidentally ranked #1 on Google with another really crappy page, so it was no surprise when the NEW SKU page featuring “leather dog collar with name plate” replaced it within the first 10 days.
Google Webmaster Tools is great for showing you the value of various position(s) for each search query. For example, for the specific query: “leather dog collar with name plate” we have two pages listed on page one of the search results.
As I write this, the NEW SKU page is in the #1 position, with 170 impressions, 60 clicks, and a 35% click-through rate. My collars.html category page is in the #2 position with 170 impressions and 16 clicks (9% CTR).
What’s really cool is that I can see that having two pages on page one as clustered results nets me an additional 30% in traffic, as well as pushing my distinguished competition down the page. Sorry, fellas!
Google Webmaster Tools data is very useful, but it frustrates me that the data rolls off after 35 days or so. If you don’t live in there, you might miss something! C’mon, Google! Data storage is SO cheap! Charge me for it if you must, but let me see more than a month!
More Keywords In Content Matters For Traffic & Revenue
The OLD SKU has 50 words of unique text in the product description. The NEW SKU has 500 words of unique text.
Wordle.net is a great tool for visualising keyword frequency in a set of keywords with keyword clouds.
Rankings and traffic are great, but I’m after conversions! The phrase “leather dog collar with name plate” was the best performing converting keyword phrase for NEW SKU, but a #1 ranking on one phrase isn’t everything. We actually made more money on secondary phrases (combined) and on long tail phrases (combined) than on our primary phrase.
Primary Phraseleather dog collar with name plate — 10 orders
Secondary Phrasesdog collars with name plates — 9 orders
dog collars with name plate — 8 orders
dog collar with name plate — 7 orders
leather dog collars with name plate — 4 orders
leather dog collars with name plates — 3 orders
leather collar with name plate — 2 orders
leather collars with name plates — 2 orders
dog collar nameplate — 2 orders
Long tail phrases (1 order each)leather collars with name plates
leather dog collar with brass name plate
dog collar name plates
leather dog collars brass name plates
leather nameplate collar
nameplate dog collar
hunting dog collar with name plate
hunting dog collars with name plate
name plate leather dog collar
leather dog collar with brass id plate
leather o ring hunting dog collar with id riveted
leather dog collars with brass tag
dog collers with nameplate
hunting dog brass dog name plate
mendota leather dog collars
It’s better to have a wide keyword portfolio of converting phrases than it is to live and die on a few phrases and pages. Your main takeaway here is that the more content you have, the more long tail keywords you’re going to rank for. Pimp out that content!
Ask Customers For Product Reviews For Content With New Keywords
So I was on deadline for this column and I noticed that I had no product reviews for the NEW SKU. No reviews?!! I panicked and sent an email to everyone who had ever bought an OLD SKU or NEW SKU in the past 5 years asking for a favor, and within 24 hours we had 121 reviews consisting of 4400 words of unique text. Holy cow!
One cool thing about these reviews is that the keywords used are different from the keywords I wrote in the product description, so I’m sure I’ll see a growth in converting keywords once this content gets indexed.
Also, my buddy, Bryan Eisenberg gave me some good advice last month about leveraging content in customer product reviews to increase conversions. He said to take what your customers say in reviews — what they like about product features and benefits — and use the exact language that they do to rewrite and improve your product description, headline, and calls to action. Great tip!
What Really Moved The Needle Saleswise?
Here’s the secret: It’s impossible for browsers to buy unless they actually make it to a product page. The biggest contribution of the NEW SKU was to get more folks to click the NEW SKU thumbnail on the category or SUBCAT page and make it to the product page.
At least half of our sales increase came from existing visitors already on the site looking at something else, but buying a “leather dog collar with name plate.”
Every visitor to gundogsupply.com probably has a dog, so at some point they’ll need a collar. The collars page always had a lot of drive-bys but no purchases, but I wasn’t quite sure why.
I’ve always started my optimization and the bottom of the funnel in the shopping cart and checkout pages because that’s the place where you get the biggest bang for your buck. A 10% improvement in checkout is 10% more revenue, but sometimes there is low hanging fruit higher up in the funnel.
Confession: I totally stole this from Khalid Saleh at Invesp, my conversion optimization guru and sometimes fellow panelist.
Sometimes you can do more good on category and subcategory and send more traffic down to the product pages, and it’s a lot easier to get real results on an already optimized cart.
We drove more internal traffic to the NEW SKU page via a combination of things:
- using very customer-friendly link text: Leather Dog Collar with Name Plate
- including the free nameplate offer in the thumbnail image
- positioning best possible position on leather collars page
- adding 5-star rating under the thumbnail
- listing NEW SKU thumbnail on the master category page (collars.html)
Positioning the NEW SKU thumbnail first row, top left doubled traffic.
Just like in the SERPs, the position of a thumbnail on a subcategory page affects the traffic to that page. I’ve believed this for years, but Michael Whitaker at Monitus proved it to me with some A/B testing he did last year on Yahoo! Store category pages. Thumbnails above the fold (top two rows) get more clicks, and the upper left thumbnail is the best position.
Last year, on the SUBCAT page, the OLD SKU was first thumbnail on second row (6% clicks). This year, the OLD SKU is 4th thumbnail, top row (4.4% clicks).
Now, the NEW SKU is in the best position: 1st thumbnail on top row (11% next clicks). Last year, a different product in same position got 9% of clicks, so this NEW SKU is a better thumbnail and at twice the price, drives much more revenue.
Position alone is not enough to drive clicks and sales when the products featured are not what folks are looking for. For example, the smaller, puppy-sized 3/4″ collars get almost no clicks, no matter how well positioned they are. I need to move these to a separate page ASAP.
When making decisions where to send potential shoppers, I like to segment by customer types and look at where folks who spent money went on the site. One thing I noticed was that both shoppers and browsers seem to click the same thumbnails based on position.
Shift your best sellers to the top of category pages and put your very best performers in the very best positions.
The Bottom Line
By adding this Keyword Clone of the product page this year, with NEW SKU and OLD SKU, we’ll generate more than $20K in additional revenue (which accounts for any cannibalization of OLD SKU sales).
$20,000.00?!! Not bad for one page with about 30 minutes worth of work!
Report on your results. I gave out an assignment or two in the first installment, so a lot of you guys have some homework to show me. Hopefully, you got similar results. Let me know in the comments below or via @robsnell on Twitter. And show your work!
P.S. – For the record, it takes more time to write about SEO than it does to do the actual work. I spent around 20 hours writing and researching these two columns, so now I figure I owe my baby brother $400K in additional sales! I gotta get to work!
Opinions expressed in the article are those of the guest author and not necessarily Search Engine Land.