Post #100 -- Looking at what works.

Well it seems like I finally made post #100, a milestone I thought would never happen. I now have access to the coveted large PM box, which I'm very much looking forward to (snicker).

As seems to be the tradition, I'd like to give a bit back to the community. My company provides services to large companies and we sell their products by managing their websites -- we are provided a portion of the sales, but we deal with each client multiple times a day, are a part of their management team and provide everything from web hosting to advertising consulting, competition analysis and much more. Because of this, we focus more on conversions and ecommerce than we do on SEO, but SEO is still a HUGE portion of our business. Since my business is very different than most, I thought this forum would be best suited for my 100th post.

So what works for us? I'll tell you...

1: Content. Anyone who argues content is NOT king is full of it. Because we deal with many different products, we know the value of getting someone to our various websites who may not be looking for our products, but might be down the road. If we continuously are helping them, they take note of this and perhaps we get their business the next time around. I can tell you this works -- for us, we have increased the business of every single client by at least 40% their first year with us. We have our clients featured in large newspapers bragging about their increased business, while their competitors were singing the blues. If you are not giving enough information about your product, you are not doing your best at converting visitors into buyers. Tell them the history of your products, the trending, what's new and hot, etc, etc. Our largest website has over 50k pages indexed in Google and not a single one of them is spam. Each one provides very usefull information for the visitor. We almost want visitors to come, read that one page, and leave. We know that the next search they do, we'll come up on top again, and they'll be back. Eventually, they'll dig deeper.

2: Conversions. Is your website built for conversions? Are you *sure* you are doing everything to convert your visitors into buyers? If not, follow the guide below:

• Analyse *everything* -- don't look just at the bottom line, look at which search engines *and* which phrases, are converting the best. Build more content around that. We go so far in this that we know exactly how much a visitor is worth to us the minute they show up on the site based off the search engine they came from, the phrase they used to find our site, and the page they landed on. The devil is in the details.
  
• Bring up your landing pages, one by one, and have a note pad handy. Spend a minimum of one hour looking over *everything* you can do to make that page better. Remember that sometimes less is more, and the number one rule in B&M selling is "location, location, location." If your users can't figure out how to convert, they won't. Try different colors, positioning things differently, larger and/or different fonts, etc. Each page should be very professional and should provide a clear understanding what you intend for a visitor to do on that page. Now when you're finished with this, you should have a list of things to test that's as long as the page you're viewing. Start A/B testing every one of them.
  
• Bring up your sales pages and repeat everyting from 2b. Are you providing enough information? Are there enough photos? Do you list similiar items? Do you provide your user a clear path to purchase this item? Do you highlight, near the purchase button, your free shipping offer? If not, you've got a lot of work to do.
  
• Bring up your shopping cart. Is it too long? Does it need to be broken out into different pages? Perhaps you could consolodate it into *less* pages (we changed ours from 2 pages to 1 and shopping cart abandonment dropped by 60%). Is it user friendly? Are the buttons in the right place? Is it easy to re-visit a product in the shopping cart? Can they return to their search results? Do you have non-secure elements on the page that may cause some users to see a popup (alot of webmasters have this option disabled, but don't count on your users having it disabled). Do you show similiar products or complimentary products they may want to add? (Like in a check-out at a grocery store, these accessories can mean big bucks!)
  
• Do not be afraid of trying new things. The Discovery store website added a 10 second video to their front page and increased sales by 500%. They were worried that it would cause a lot of people to actually leave, but it ended up turning in more sales. Run those A/B tests and see what works!

3: Advertising. We advertise on over 1,000 different websites for our various clients. We know which ones perform and which ones just plain stink. No matter how big you grow, you have to look at every advertising dollar as if it were your last one. If they aren't performing, cut them loose! Don't EVER believe the stats someone tells you when they call you up to renew. Look at your own stats and determine how right they are. In our experience, they're normally padded by 20-30% or more. Don't fall for the line "your competition uses us, if you don't, you'll do poorly." It's BS and it's a line used by companies that don't perform. If you hear that, there's no point in giving them your business to begin with.

4: Expenses. A real estate agent wears nice clothes and drives a nice car. There are expenses to every business, and if your business is on the computer, you need to spend money to make it. Do you have the most up to date software? The best chair to sit in? The best monitor? If not, you're loosing money in lost productivity and time needed to complete a project. If you're blowing money on your Porsche, but sitting in a $50 chair, you're not in things for the long haul. Spend thiftly, but also spend wisely for productivity. My personal weakness is silk Tommy Bahama shirts. I bet I've got 250 of them by now. But I buy them at the outlet and spend the savings on a $6k computer and $20k of productivity software. My girlfriend has been telling me to get a new car for years, but my 98 Ford Explorer Eddie Bauer edition is just fine for me -- it's been paid off forever, and my clients could care less what I drive.

5: Personal time. Being devoted and being crazy are two different things. I spend 16 hours a day working, but I also spend an hour each day working out or swimming, and I try to take an afternoon off every other week. I just got back from being on the road for a month visiting clients, and I can't tell you how much that helped.

6: Risk. If you aren't taking them, you're in the wrong line of work. Nobody ever achieved a dream without taking risk. You can only play safe for so long -- eventually you have to be willing to bet on yourself in order to make a huge jump forward. Doing this caused our business to go from nothing to having 3 of the top 5 clients in our industry.

7: Focus. In the beginning, you'll want to explore every new idea, esp. if you think it "fits" in your business. Be careful -- sometimes these are great, but often times they are time wasters. Learn to approach a problem with an open eye and a true understanding of if it will genuinely help your business. If you haven't got everything on your "to do" list done, spending time spreading yourself out isn't going to help -- it's going to hurt.

8: Learn. If you stop learning, you die. Read everything pertaining to your industry, the products you push, and the SEO world. Spend an hour each day, minimum. Don't be afraid to spend big bucks for the better subscriptions. I spend a ton on "The Economist," but one issue easily pays for the year's subscription.

9: Listen. You're not an expert at everything, nor will you ever be. Be willing to accept you're wrong and use the opporutinity to... LEARN! Don't be afraid to ask questions.

10: Share. We teach our clients as much as possible. Others, like Randfish, teach the SEO community. The mistake many people make is "if I share what I know, I'll be out of a job." While "knowledge is power," sharing it can be also. We know the value we bring to our clients and we educate them on this all the time. They know that without us, they would spend more on personnel, research, tracking, etc. By teaching them as much as possible, we don't make them experts, but we do help foster a relationship that continues. And you'd be surprised at what other experts share with you when you share yourself.

This is by no means a conclusive list, and what's posted here may not help everyone. I didn't go into as many details as I would normally, but my fingers are tired.

I hope this helps!

S

Wow, congratulations lovethecoast

Excellent post, ltc. I've printed this out. The points are very well made.

congratulations

Had to bump this up...because its such a great post.

Had a conversion today...almost right after I saw this...and I thought of the post. We won the transaction on detail and content. Competition couldn't touch us on content.

We had researched points that come up all the time. We responded to every detail point due to long research, experience, testing, adjusting.

Love your 3rd point also. I have yet to have an advertiser quote the exact same traffic that we measure. Isn't that strange...not one of them has ever quoted traffic that is less than what we show.

Keep up the good work.

(BTW...I still think your buyers/decision makers get worse as the buying season progresses!!!!)

Dave

We're starting to tackle that on one of our websites pretty well. I'm hoping we can take what works there and use it on the others. We're also working on a new promotional that's new to the industry. It's cheaper than discounting, but provides more "ummf" to the consumer.

What's really funny is we've been able to show our clients that whenever they do a huge promotional, they end up giving business to their competitors (we track competitor sales based off online inventory and we always see *them* spike after we do a huge offline promotional). This is because our clients don't have enough people answering the phones -- the person is now interested in the product, so they go to a competitor instead. Because of that, we're looking at ways of getting more people to buy online (more money for us, but more sales for our clients also). That should prove pretty interesting come Q1 of 2006.

I still say we need a "meeting of the minds" for like-minded people on here... Doesn't seem to be much interest in conversions though -- everyone just wants more traffic. Oh well.

We run strong from the beginning of the year to the end of summer. Business generally tails off starting in September and gets real weak in November and December. Late spring is our busiest time of year

We promote hard starting in August...pointing out why this is a great time to take our program.

Yr after year it hardly works. At the end of the year we do heavy price discounting.

In our case...I think potential customers are saving for Christmas.

We'd do anything to reverse that trend.

Dave

Lovethecoast, that was information-packed. Thanks for taking the time to write it.

I have a question about your A/B testing you mentioned (I've been reading through your posts on different threads and have seen this term a few times).

How exactly do you carry it out? Make a change and track it for 7 days and compare the results to the old way/layout? Or make two distinct pages of the same product/info and ship visitors to one or the other alternately and see who converts?

This is probably better suited for a different post, but I'll give you the short and skinny.

The length of time an A/B test is run is dependant upon how much traffic you have, and how many conversions you do. Generally speaking, we run an A/B test at a minimum of 10k visitors for prelim results, then if the results aren't too bad, we allow it to extend up to 50-100k visitors. We also don't limit things to just A/B -- we might have A/B/C/D/E/F/G/etc (note: that generally only works well with many many visitors).

Also, many will claim you should do only one A/B test at a time, but we've found running multiple tests is ok as long as they don't overlap each other. For instance, you would only want to change one thing at a time on a product details page, but if you have multiple landing pages, testing changes to each of them probably won't hurt things all that much. When a user enters our check-out process, we can have up to 10 A/B tests going at one time and we store those values in their purchase record.

We also worked on making sure our check-out process was as painless as possible. This set of testing is actually still in the works, but for at least one of our sites, changing the check-out process from three to just one page decreased abandonment by over 30% (for the first couple of weeks, it was down over 40%).

Another thing to keep in mind is to not give up on the testing. What works today may not work tomorrow. What doesn't work now may work great tomorrow.

Also note that if you're asking this in reference to AdSense placement, I have no clue. We spend a ton on AdWords, and focus on getting conversions from our advertising dollars. We do not focus on AdSense, as we generally don't want to have outside ads on our websites. We do, on an extremely limited basis, show AdSense, but that's more because we simply have so much traffic on a few of our sites, we might as well try to recoup some of our advertising dollars. That only brings in a couple grand a month, but like I said -- it is *very* limited.

BTW -- Dave should be sharing some of his secrets on here. The man really knows his stuff and has given me more than a few ideas for the future! Everytime we end up talking, I walk away with 2-3 ideas! Where are they Dave?!

S

You got me thinking about this in the thread I started about conversions, and this thread just prompted me to post. Not thinking about Adsense necessarily, but now that you mention it I should think about it for the sites that run it. I was thinking more in terms of conversions (whatever type they are), and how to tweak them. I've made some site changes to our main site, but there was little testing done - more like "something is better than nothing, so let's do something. Later we can worry about doing something else."

Is there specific tracking software you use to be able to hone in on what triggers are changing your conversion rates?

Everything we do is 100% custom. It's easier to develop inhouse rather than try to use a third party solution.

I'm very enterested in conversion, and the issue of abandoned carts/baskets. If the visitors are interested enough to put something in the cart, we have had a good strong 'nibble' and have to wonder why they didn't finish.
How do you get the detailed data that you quoted?

Very informative post. The one thing you mentioned in your post that I've found to be the crux of ecommerce success is learning. Like most people I came into a venture gung ho with preconceived notions about what it would take to get "over the top". Like most things in life, however, once you get your hands dirty allot of these notions "fly out the window". Sometimes the key is not to steadfastly hold onto your contrived plan, but to adapt to your market place and, well ... market accordingly.

Everything we do is custom programmed, so for us, we just looked at the number of people that landed on the check-out page and compared that to the number of people that hit the "thank you" page.

A stats package could provide this information easily as well. If you have a multi-page check-out process, then you'll probably want to program something custom.