Photo By Taylor Sloan

But the truth is that I have a huge laundry list of things that need improvement. And the problem is that this list is quite large so I always have to prioritize my time especially since I have a day job and 2 kiddos to play with.

Also, keep in mind that my shop is live 24 hours a day, 7 days a week. Therefore, every single change that I make no matter how small needs to be tested thoroughly. Any bugs that I introduce to my shopping cart can have a drastic impact on conversions which is why I’m extra careful whenever I make any design changes.

So why is the title of this post about problems that “I finally got around to fixing”? As you know, I’ve been working on and testing my mobile site for quite some time now. And since I was already “under the hood” of my website, I finally decided to fix some really annoying issues with the store.

Honestly, some of these problems have either had a negative effect on conversions or have annoyed me for quite some time now and should have been fixed a long time ago. Please feel free to ridicule me in the comments sections if you must:).

The Dreaded Double Submission Problem

The first problem that I fixed is actually fairly common. Sometimes it can take quite a long time for a website to send a customer’s information to the credit card processor and receive a response whenever a customer tries to place an order. And during the 2-10 seconds that it takes to completely process an order, sometimes a customer gets impatient and hits the “submit” button again.

Now I don’t know why most shopping carts have this problem, but whenever a customer hits the submit button on my website more than once, a new and identical order is placed online. What this means is that a customer’s order can come through multiple times and their credit card can be charged multiple times. Needless to say, this is really bad.

We have lived with this issue for a quite a long time now simply because it doesn’t happen very often. Every month or so, we get maybe one or two double submits so I never made this issue a priority to fix. However, this past February was especially bad. In fact, we had one customer get so impatient with our website that they hit the “submit” button 6 times which resulted in 5 duplicate orders.

And when this customer saw 6 duplicate credit card charges on her statement, she was furious and sent us a nasty email. Granted, we always refund duplicate orders immediately after they are placed but sometimes it takes one full pay period for the refund to show up. In any case, this lady got so pissed off and thought that we were trying to cheat her that she cancelled her order and filed a complaint with her credit card company. After this incident, I knew I had to fix this issue.

Turns out that the fix was rather easy. Here are the lines of code I added to my website to prevent the dreaded double submissions problem. (Note: This fix requires JQuery)

<script type=”text/javascript”>
$(document).ready(function(){
$(“form”).submit(function(){
$(‘input[type=submit]‘).attr(“disabled”, “disabled”);
$(‘input[type=image]‘).attr(“disabled”, “disabled”);
});
});
</script>

What this code essentially does is it disables the submit button after it is clicked. Pretty simple eh? Should have fixed this long ago.

Confusing Login Screen

The second problem I fixed recently had to do with the first screen of my checkout process where a customer has to choose whether to create an account or checkout without an account. Here is a picture of my “login” page before my changes.



The problem was that some customers arrived at this screen and assumed that they needed a special login or account in order to complete their purchase. Now even though we clearly provide the option to checkout without an account, it’s not always readily obvious to people who don’t read the text carefully.

What usually happens at this point is that the customer either calls and places the order over the phone or abandons their shopping cart altogether. Now to my credit, I wasn’t really aware that this was happening until recently when we started receiving an unusual flurry of calls asking us how to place an order.

If you haven’t figured it out already, the problem with our login page is that the checkout buttons don’t stand out enough in the upper block of text. Since we lump the regular checkout button in the same paragraph as the express checkout button, it ends up looking like a block of text. So a customer who doesn’t read very carefully can be led to believe that they need to sign-in in order to make a purchase.

To solve this problem, I separated the 2 options and made the checkout buttons more obvious to the customer. I also rewrote the verbiage to make the “registering for an account” part of checkout less obvious.

Shopping Cart Abandonment On The Last Step

The final problem that I fixed had to do with customers abandoning their shopping cart on the final step of checkout after they’ve already entered in their credit card information. Under normal circumstances, this should be extremely rare. Once a customer enters in their personal information and their credit card numbers, it’s almost always a guaranteed conversion. That is…unless you screw up the confirmation page. Let’s see if you can see the problem in the screenshot below.



If you look at the page really quickly, you might think that you have already finished placing your order. In fact, we’ve had several customers call us to complain that they never received their order because they never actually completed the checkout process! One of the main reasons that customers think they are done with their purchase is because the “Confirm Order” button is so tiny and it’s displayed in the lower right corner in a color that blends in with the rest of the site.

By replacing the “confirm” button with a much larger button with a different colored gradient, we’ve since reduced the abandonment rate on this page.

The Changes Never End

These 3 changes have been live for a little over a month now and everything seems to be working smoothly. Our shopping cart abandonment rate on the confirmation page went down from about 6% down to 3%. Our abandonment rate on the “login” page has also seen similar reductions. And not having to deal with duplicate orders has been a major relief.

It just goes to show that conversion optimization never ends. The moral of the story is that no matter how well your online shop is designed, you will always find some customer who will get confused by your checkout process. If you keep things simple, you’ll have a better chance of converting the sale.

Final Note: The proper way to test your changes is to perform A/B testing on your pages. The reason I didn’t use A/B testing in the above cases was because I already knew that my pages had problems that needed to be fixed. In the future, I may test different versions of the login screen to see what works best.

Photo By Auswandem Malaysia

Let’s face it. Most of us aren’t web designers. Most of us aren’t very artistic and most of us are not well versed in the art of writing html or php.

The good news is that you don’t really have to be an expert in any of those things in order to put up a good looking online store.

The Beauty Of Open Source

One of the best aspects of using an open source shopping cart such as Open Cart is that it is a standard platform. As a result, many 3rd party developers actively create themes, templates and plugins for Open Cart and many of them are free or can be purchased at a very low cost.

For example, many of the templates for Open Cart are free or can be had for as little as $25. A template is a theme for your shopping cart that completely transforms the look and feel of your online store. While your products and your database remain the same, the aesthetics for your shop can be instantly changed to suit your tastes.

In addition, you can purchase a wide variety of 3rd party plugins that add additional functionality to your online store at little or no cost as well.

Installing A Shopping Cart Template

Today’s video will give you a brief overview about shopping cart templates and themes and provide you with a complete walkthrough on how to install your theme once you make a purchase.

Once again, there is no technical experience required to follow along. Enjoy the video!

Want To Learn More?

Have you enjoyed my videos so far? If you are itching to start your own online store, you definitely need to check out my full blown course on how to create a profitable online store.

I will teach you everything you need to know in order to run a successful online shop.

No Technical Experience Necessary

In an informal poll that I conducted with my newsletter readers a while back, most entrepreneurs want to start an online store but don’t want to be bogged down with technical details. They just want to start selling right away which is precisely why sites like Ebay and Etsy continue to prosper.

But what most people don’t know is that there are many open source options out there that are completely free. And the best part is that all of these carts support SSL and a plethora of credit card gateways out of the box.

What this means is that everything you need to run a fully functional shop that accepts credit cards and other forms of payment has already been created for you. You just need to know how to set things up and I’m going to show you exactly how to do it.

Video Tutorial

Today’s video is double the length of the last one at 28 minutes. However, I promise you that it is extremely detailed and shows you every single step that is required to start selling including…

• How to get a dedicated IP address for your website
• How to setup and purchase an SSL certificate
• How to enable SSL with Open Cart
• How to setup your merchant account and gateway
• How to setup Open Cart to accept credit cards

Enjoy the video!

Want To Learn More?

Have you enjoyed my videos so far? If so, there’s more where that came from if you sign up for my full blown course on how to create a profitable online store.

My course offers over 35+ hours of video and includes live office hours where you can ask me questions directly.

If you want to learn everything there is to know about ecommerce, be sure to check it out!

Photo By SheeShoo

However, this video got mixed reviews from my audience because it was too technical. Several readers tried to follow the instructions but got stuck trying to figure out how to use some of the free tools I recommended.

Another set of readers were turned off by the amount of typing of “random” commands into their terminals.

The reason I created that video the way I did was to show you at a fundamental level what was needed behind the scenes to set up an online store.

But after some thought, I realized that I went about things all wrong. Most people aren’t interested in the technical details behind a website. People just want to know the fastest and easiest way to set things up and start selling.

Starting An Online Store With No Technical Knowledge

So today, I’ve created a video tutorial on how to start your online store WITHOUT any technical knowledge required. Seriously folks, if you know how to use a mouse and basic knowledge on how to surf the web, you can start your own online store! And the best part is that you will have full control over your shop. No one will be able to change the terms and conditions. No one will be able to charge you a fee for selling online.

In addition, this time I’m going to take things one step further. In a subsequent video, I’m also going to show you how to set up SSL to make your online store secure and I’m also going to show you how to setup credit card processing as well.

In a nutshell, I’m going to teach you everything you need to know on how to set up a free open source shopping cart from scratch and start selling online without any technical experience whatsoever.

Enjoy The Video!

If anyone has any questions after watching the video, feel free to send me email.

Want To Learn More?

Did you enjoy this video? If so, there’s more where that came from if you sign up for my full blown course on how to create a profitable online store.

My course offers over 35+ hours of video and includes live office hours where you can ask me questions directly.

If you want to learn everything there is to know about ecommerce, be sure to check it out!

Right now while I’m developing my site it doesn’t really matter but how do I put my store under maintenance after I’ve launched without affecting my search engine rankings?

The great thing about this question is that it applies not only to ecommerce shop owners but to all webmasters in general.

If you improperly put your website under maintenance, you could potentially destroy your search engine rankings while driving away customers at the same time.

Especially if you plan on bringing down your website for an extended period, it’s essential that you do things the right way.

Common Mistakes

But before I discuss the proper way to put a website under maintenance and what I do with my online store, I thought that I would talk about some common mistakes that I’ve seen people make with their sites.

Probably the worst and most egregious mistake that I’ve seen is to simply bring down your entire site and leave a bunch of pages stranded with “Page Not Found” errors. Please do not do this! Unfortunately, you’d be surprised by how many times I’ve seen people edit their website this way.

Another common mistake I see is webmasters redirecting all pages on their site with a 301 redirect to a pre-made “under construction” page…and this is where the trouble starts.

On the surface, it might seem fine to bring down your site this away. After all if a human visits your site, they will see an attractive “maintenance” page and come back at a later date.

However in the eyes of the search engines, Google will think that your entire website has been moved and now points to your “maintenance” page. And given enough time, this could have disastrous results on your search engine rankings.

The problem is that it’s fairly easy to fall into this trap. For example Open Cart, a popular open source shopping cart that I recommend, offers a maintenance mode option which seamlessly redirects all customers to a custom “under construction” page.

However, the problem is that Open Cart uses “301 Redirects” which tells Google that the pages have been moved permanently. As a result, Open Cart’s maintenance mode is really only good for the beginning stages of web development when you don’t have any real customers yet.

A Better Solution

So what’s a better way to do this? A much better solution involves letting both humans and the search engines know that your website is only temporarily down and this can be accomplished with a special status message called a “503 status code”.

In a nutshell, a 503 code indicates that your website is temporarily unavailable. In addition to being unavailable, you can also tell the search engines when to come back again when you think your website will hopefully be ready.

Here’s how to issue a 503 status return code for your website. The first thing you need to do is create a special file called 503.php. Then in your 503.php file, you need to insert the following lines of code at the top of your file.

Note: The “Retry-After” parameter below indicates to the search engines when you want them to come back again. The time unit is in seconds so in the example below, we are telling Google to come back after 60 minutes.

After these lines, you can then use the rest of this file to create a custom “under maintenance” page. Finally, you should edit your .htaccess file to redirect all visitors to your 503.php page by adding the following lines.

In a nutshell, the above lines in your .htaccess file will redirect all visitors to your 503.php file which will tell both humans and search engines that your site is under maintenance. Make sure that you put your own IP address where you see the numbers 111.111.111.111 above so that you can still access your site from your own computer.

The Best Solution

So creating a 503 maintenance page is all fine and good but obviously the best solution to your maintenance problem is to never bring down your site at all which is what I do with my online store. Instead of taking down my site whenever I need to work on it, I have 2 identical versions of my website installed on my server. The second copy of my site is called a staging site.

If I need to make changes to my online store, I can leave my existing website live while I work on an identical copy in the background. Then, once I’m done making and testing my new changes, I can immediately copy them over to the live site. This way, I get a chance to thoroughly test and update my changes without worrying about losing sales.

There are two ways to create a staging site and it doesn’t matter which way you choose. One option is to create a completely separate subdomain for your site and create an identical installation. For example, you could use the URL staging.yoursite.com for your staging site.

Alternatively, you can simply create a brand new subdirectory under your public_html directory and perform an identical installation there. Either way, you’ll end up with 2 versions of your live site to work with so you don’t miss out on any revenue.

So the next time you feel like you need to bring down your website for maintenance, you might want to consider implementing a staging site for your online store. After all, if your store is not open, then you aren’t making money.

Photo By Old Shoe Lady

First of all when I go shopping on my iPad, the last thing I want to do is to have to download an app in order to shop efficiently. The iPad screen is large enough such that browsing full blown sites is comfortable for most online stores.

The only thing that would be a detriment to online shopping on an iPad would be if the site used flash, which most sites already do not.

In any case, reading this article made me wonder how many customers shopped with iPads and other mobile devices at my online store during the Black Friday weekend and how many customers actually made a purchase. After all, I hadn’t done this analysis for several months now. Was I wrong in disagreeing with this post? Do I now need a mobile version of my store? What metrics should I use to make my decision?

Last Year’s Results

Last year, only about 6% of our online visitors arrived through a mobile device using data gathered starting from Black Friday through the first week of December. Note that this period of time reflects our busiest shopping days of the year. Here was the breakdown among devices.



It’s important to note that during this period last year, we had 0 customers actually make a purchase from our store. If you recall, last year I was considering whether to create a mobile version of my website but was still hesitant to pull the trigger.

This Year’s Results During The Same Time Period

This year, the number and percentage of mobile visits increased dramatically by over 3X. In fact, this was the first year ever where the percentage of mobile visits hit double figures at over 14%. Here was the breakdown among devices.



In terms of sales, mobile visitors accounted for about 6.6% of revenues with about 90% of mobile sales coming from iPad users. Not too shabby. Clearly, the number of people who shop on their mobile devices has been growing exponentially!

Time To Pull The Trigger On A Mobile Optimized Site?

So here’s my situation. When I first designed my online store, I had dial-up visitors in mind. I chose a screen resolution of 800×600 and optimized the site to render the pages in under 10 seconds on a 56K modem!

These days, the newest Android phones pack screen sizes that are 1280×720-pixels in resolution with download speeds that are even faster than my cable modem connection at home. It’s pretty clear in my mind that mobile devices are getting better and better at browsing full blown websites even on a phone. Am I losing out on potential customers by not having a cell phone miniaturized website?

Personally, I try to avoid most mobile optimized shopping sites when browsing from my phone because they are usually poorly done. Having another site design would mean that I would have to do double testing on both full and mobile browsers with my shop.

Real Data From My Online Store

Whenever I have to make a design decision, I always look to my stats so here are some web browsing statistics for mobile users compared to the desktop/laptop users on my website.



In addition to analyzing the data above and drawing some conclusions, I also did some research about the inherent differences between mobile versus desktop traffic. Here’s what I found.

• Mobile users are more affected by proximity. For example, if someone types in “bike shop” on their desktop computer, they are probably looking to shop online. But if they type “bike shop” into their phone, they might be looking to shop at the bike store closest to them. What does this mean? Mobile searchers might be expecting to find a place within walking distance or within a short drive. If these visitors land on my shop which is only offered online, they might leave immediately.
• The search listings are usually abbreviated on a mobile device. The fact that the title tags and meta descriptions are often abridged on a mobile device means that a user has less information when they actually click on a link. As a result, bounce rates will tend to be higher in general
• People don’t like scrolling to the right. If your site is too large to fit horizontally on the screen, chances are that your customer will leave.
• People who shop on their mobile devices have more money. I didn’t know this before but it turns out that 42% of mobile internet users earn over $80,000 a year, and tend to spend more on the run.
• Mobile users are more impatient than desktop users 60% of mobile users only wait 3 seconds or less for a page to load before leaving. 74% of mobile users will only wait 5 seconds. That’s not very long.

Analyzing The Data

Ok, so back to the million dollar question. Do I need to design a mobile enabled version of my store? If you look at the metrics above, the statistics that compare desktop users vs mobile users is pretty frightening. All across the board, the bounce rate is mostly higher, the average time on site is lower and the pages per visit is lower as well.

But if you take a closer look at the data, I chose to break things down by traffic type because I wanted to specifically separate out cost per click traffic. Why? It’s because my CPC traffic is my most predictable and consistent form of traffic. If I want to do an apples to apples comparison of my data, then the CPC data is the way to go. After all, every keyword that a visitor uses to land on my site from a PPC ad is extremely targeted and brings them to a very focused landing page.

So if I look at CPC data alone, it looks like my bounce rate isn’t really an issue. For this limited data set, the bounce rate for mobile users was actually better than desktop/laptop users. However, what worries me is that both the time on site and the pages per visit metrics were drastically lower.

Given the differences between mobile and desktop traffic, I’m not quite sure that I can rationalize a 27.76% drop in pages per visit on a mobile device. In addition, a statistic missing from the table above is that my mobile conversion rates are almost 50% lower than my desktop/laptop conversion rates. Naturally I would expect mobile conversion rates to be lower but definitely not as bad as 50%!

What this tells me is that I need to take mobile a little more seriously. As of today, none of my competitors for my main targeted keyword phrases in the top 10 of search have a mobile enabled version of their site. If I start now, I can get a jump on the competition. After all, mobile traffic seems to be growing exponentially with no end in sight.

The big question in my mind is whether the browsing experience on mobile will improve to the point where it will be equivalent to a desktop computer. Don’t get me wrong. I’m not worried about iPad or tablet traffic. I’m more concerned about people who shop using their itty bitty phone screens.

At this point, it would be a mistake to sit back and hope. I’m going to start experimenting with a mobile site to see if it makes a difference with my numbers. What do you think? Do any of you have a mobile enabled version of your online store? Are your numbers better than mine? Please go to my online store at Bumblebee Linens on your cell phone and let me know what you think.

Which One Is Cheaper?

The first and foremost criteria that most people use to select a credit card processor is price. But when it comes to credit card processing, the different fees are not always straightforward to calculate.

For example, some credit card processors charge you a registration fee, a monthly gateway fee, a statement fee as well as a percentage of your sales. Some merchants allow you to apply your monthly gateway fee towards your discount rate fees. Some merchants provide a tiered discount rate depending on how much you sell. There are many different ways to get charged depending on your sales volume and all of the numbers can get confusing depending on the scenario.

The reason I like Merchant Plus is because it is cheap and the fees are very straightforward to understand. For example, there are no complicated rules on how you get charged.

• There’s a monthly fee of $19.95
• There’s a 25 cent transaction fee per charge
• The discount rate is 2.15% for Visa and Mastercard

After researching many different credit card merchant account and gateways, the rates for Merchant Plus are very reasonable, the service is very dependable and they provide good support as well. Now let’s take a look at Paypal Website Payments Pro.

The fees for Paypal Website Payments Pro are fairly straightforward as well but the rates are more expensive and tiered depending on your sales volume.

• There’s a $30 monthly fee
• For a monthly sales volume of $0.00 USD – $3,000.00 USD, you get charged 2.9% + $0.30 USD
• For a monthly sales volume of $3,000.01 USD – $10,000.00 USD, you get charged 2.5% + $0.30 USD
• For a monthly sales volume >$10,000.01 USD, you get charged 2.2% + $0.30 USD

Overall, Merchant Plus is always going to be cheaper than Paypal Website Payments Pro.

Which One Is More Convenient?

Money is important but it’s not everything. When you sign up for Merchant Plus, you are more or less signing up for a vanilla merchant account and gateway. In other words, you’ll be able to process credit cards but that’s about it. What’s nice about Paypal is that they offer additional features that are very convenient.

For example, your online store is probably going to accept regular Paypal in addition to credit cards. Paypal Website Payments Pro allows you to accept both Paypal standard payments and credit cards all through the same account. This is really nice simply because you’ll have one less account to deal with. In other words, that’s one less account where you’ll have to transfer money out of and into your business checking every single month.

The other really nice thing about Paypal is that they offer an extremely convenient and easy to use one click shipping platform. Whenever someone makes a purchase from your shop, the address information can be sent directly to Paypal. Then, using their multi-order shipping interface, you can quickly and easily ship all of your packages and print shipping labels from Paypal’s website. The best part is that all of the shipping fees are automatically deducted from your Paypal account.

If you run a small online store, that you means you don’t have to sign up for a 3rd party shipping platform like Stamps.com. You don’t have to deal with downloading and uploading customer addresses to a separate shipping service. You can do everything without leaving Paypal’s website!

The Intangibles

Sounds pretty good right? But the one caveat with Paypal is that Paypal is not a true bank. With Authorize.net and your own merchant account and gateway, any money that you receive is automatically yours once it reaches your bank account and no one can easily take it away from you.

With Paypal however, they get to call the shots with your money and can place holds on your account whenever they feel like it. For example, let’s say that Paypal detects some abnormal charging patterns with your account. Without any warning, they can freeze your assets and prevent you from accessing your funds. A full on fraud investigation can last for several months and during this period you will not be able to withdraw your money from your account.

While this has never happened to me (knock on wood), there are many websites out there such as www.paypalsucks.com that are filled with stories and complaints from Paypal users who have had their money frozen for long periods of time.

Which Should I Choose?

If you don’t care about the integrated shipping that Paypal offers and you don’t mind having yet another bank account, going with Merchant Plus is a no brainer. In addition, readers of MyWifeQuitHerJob.com can take another 20% off of the monthly fee by using coupon code: MYWIFEQUIT upon signup.

However, if you hate maintaining a bunch of different accounts and want a nice, integrated way of shipping your orders, you should consider signing up for Paypal Website Payments Pro. While it starts out more expensive, once you reach a monthly sales volume of greater than $10000, the difference in discount rate is not that big of a deal.

The other major thing you have to consider is whether you are comfortable with Paypal having full control over your funds. If you are even remotely worried about having your assets frozen, then avoid Paypal altogether. Once again, this has never happened to me before but just be aware that it’s a possibility. Good luck with your decision!

What Do I Need To Begin?

In order to install an open source shopping cart and follow my tutorial, you’ll need to do the following.

• Register a domain name – You can grab a domain name from Go Daddy for a dollar a month.
• Sign up for a web host – I personally recommend Blue Host because I ran my online store for 2 years there and had a fantastic experience.
• Get SSH access to your server – You need to contact your web host and ask for this access which usually requires a faxed in a copy of your drivers license
• Install WinSCP or any FTP client to transfer files to and from your webserver
• Install PUTTY for SSH terminal access to your server
• Have your web browser ready
• Download Open Cart for free from www.opencart.com

Installing Your Open Source Shopping Cart

Once you’ve downloaded all of the necessary programs above and have your web host all set up, you are ready to go!

Quick Note: All of the instructions mentioned in the video below are actually not necessary if you sign up for a hosting account at Blue Host. The reason is because Blue Host offers a program called “Simple scripts” which allows you to install many programs at the click of a button.

In any case, the tutorial below will teach you how to install a shopping cart from complete scratch which applies to any web hosting account. Enjoy the video!

Help!! I need help!! Kyle pooped in the tub again!! Come quick!!!

Naturally, my wife stopped her shopping spree and helped me clean things up. Yuck!! By the time she got back to her computer a good 45 minutes had elapsed.

Much to her dismay, when she got back to the computer to resume shopping, her shopping cart was empty!!! All of that hard work and she had lost it all because the online store she was shopping at had timed out and emptied out her cart. I will now attempt to recreate our conversation afterwards.

Wife: S#$%! F#$&! What the hell! I lost everything ! I had so many cute outfits in my cart and it took me forever to find all that stuff.

Me: Yeah that sucks….How much was in your shopping cart?

Wife: I don’t know, 700 maybe?!? I hate this store. Never shopping here again

Me: WHAT!?!?!? 700 dollars?? My entire wardrobe doesn’t cost that much!!!! What the hell!!! Do you think we are made of money?!?!

Wife: Well don’t worry about it. I’m not buying anything from this store anymore. It’ll take forever to find the stuff again

Me: (Feeling a sigh of relief) Oh….darn…..I know how much you liked that stuff (feigning concern). I feel horrible…absolutely terrible…

Wife: Shut up. Just for that, I’m spending more at the next shop

Persistent Shopping Carts

The reality is that many shopping carts behave this way. If you close your browser session or if you wait too long, many online stores will empty the contents of the cart. As a result, if a customer returns after a long hiatus of shopping, they will lose what they’ve added.

In many cases, this can cost you a sale because not everyone shops and makes a purchase in a single sitting. Just as an experiment, I asked a bunch of my friends whether they shop this way. And surprisingly, many of them use an online store like a wishlist. They put what they want in their cart, think about it and then return later. And more often than not, they shut down their browser and let things sit a bit before making a purchase.

In fact, looking at our online store statistics, a good 25 percent of customers visit our store more than once before making a purchase. The best way to deal with these customers is to make sure that whatever they put in their shopping cart stays in their shopping cart until they return and are ready to make a purchase. This is what is called a persistent shopping cart.

Implementing A Persistent Shopping Cart

This feature has been on my list for a very long time but I’ve been dragging my feet because strangely enough, there’s no plugin that does this for my shopping cart. Now this feature is very important for our online store for one main reason, personalization. When people shop at our store, then often put up to 8-10 items in their carts at a time and many of these products involve personalization with a special message.

What also ends up happening is that customers take a long time to shop because they want to think about what they want to write and sometimes they have to confer with their significant other before deciding on a message. Can you imagine how frustrating it would be to lose everything you typed in just because you took an extended break?

So it was a no brainer, I had to spin my own solution. Here are 3 possible solutions that I came up with. I apologize in advance if this is next section is a bit more technical than usual.

Extend The Session Timeout Period

By default, the session length for my server is about 15 minutes long. What this means is that my server will hold the contents of a customer’s cart in system memory for 15 minutes of idle time before emptying everything out. So, a logical solution to this problem would be to simply extend this timeout period.

The problem with this solution is that it greatly increases the memory requirements for your online store. By increasing the timeout period for your shop, you essentially have to hold a customer’s shopping cart information in main memory for much longer periods of time which effectively increases the amount of main memory you need to run your website. This solution is clearly not scalable because main memory is a precious resource.

Not only that, but it also doesn’t solve the problem when a customer shuts down their browser or waits for 2 weeks before returning. Do you really want to keep their shopping cart lying around using up precious compute resources for 2 weeks? Probably not…

Store The Cart Contents In A Cookie

Another solution is to use cookies to store the contents of their cart on their computer. Cookies are little files that websites put on the client’s computer in order to store certain pieces of information so a website can recognize a customer when they return. By placing a cookie with the shopping cart contents on the client computer, you can effectively store their shopping cart for as long as you like!

The problem with this solution is that the cookie size is limited to 4K on most browsers. Therefore, if a customer’s shopping cart is large and exceeds this value, certain products in their shopping cart will be lost.

The Best Solution

The best solution and the way I ultimately decided to solve this problem was to use a combination of cookies and my shopping cart database. Here’s what I came up with.

Whenever a customer enters my store, I place a cookie on their computer that is set to expire whenever I feel like it. This cookie contains a secret id which I use to tag certain products in my database. Whenever a customer places an item in their shopping cart, I also store the product in my database and tag it with this secret id.

This way, I can keep track of everything that the customer is interested in based on this secret id which is stored on their computer in the form of a cookie. If they choose to leave for a long time and come back, I look for the presence of this cookie and restore the contents of their shopping cart based on my database which is present for as long as I want it to be.

The downside to this solution is that if a customer abandons their shopping cart, I will accumulate a lot of junk in my database which needs to be periodically cleaned up. So to solve that problem, I set up a CRON job or automated task to automatically clean up the database every single night and remove items that exceed a certain age threshold.

I’m still in the throes of testing my additions to my store and hope to be live with this new feature by the end of the week. The best part is that by storing your abandoned shopping carts, you can also better observe your customers shopping patterns as well.

Photo By javYliz

In any case, the issues that I’m about to describe aren’t necessarily “bugs” per se but website usability issues that we discovered from talking to our customers directly.

9 times out of 10, a customer gets confused navigating through our online store because I’ve made certain assumptions about the way a user should interact with our site. I’ve made certain assumptions about how clear the instructions are written and how tech savvy the internet user is.

And you know what? I’ve found that whenever I “assume”, it makes an “ass” out of “u” and “me” (One of my professors used to say this to me all the time).

Now I’ll be the first one to admit that I have blinders are on when it comes to critiquing my own online store. The problem when you design your own website is that you often overlook obvious mistakes because you are too familiar with your own design. Of course you can ask your friends to provide their opinion as well, but most likely your friends won’t be an exact representation of your customer base either.

For example, all of my friends are pretty web savvy so they can usually navigate through any decently implemented website. The other problem is that your friends might be too polite to tell you how they truly feel. In all cases, the best way to get the proper feedback is straight from the customer’s mouth which is why I take all customer feedback extremely seriously.

Whenever a customer even hints at a usability issue with our site, I take a very close look and make changes where appropriate. Here are a few tweaks that I made because of a single customer call we received last week. You can be the judge about whether these problems were worth fixing.

Default Radio Button Selection During Payment

This particular customer called us because she could not figure out how to get past the payment screen during checkout. Here’s a snapshot of our website for the purposes of illustration. Can you tell what’s wrong with this picture?



What happened was that this customer entered in all of her credit card information, but then forgot to hit the radio button to indicate that she was paying by credit card. As a result, she kept getting the message “Please select from one of the following payment options” but couldn’t figure out what she was doing wrong.

Now most users who are even remotely web savvy would figure this out eventually and click on the “Pay By Credit Card” radio button. But the problem is that you have to design your website for the lowest common denominator. And more often than not, this includes people who don’t think like you, people who may be used to doing things in a different way, and people who are just plain clueless. In any case, as soon as she told me her problem, I was able to help her make it past the payment page.

Now I’m a firm believer that if a single customer reports a usability problem with my website, then chances are there are many others out there who are experiencing the same problem without you even knowing. So, I placed a piece of Google Analytics code on my store to track how many times people got the “Please select from one of the following payment options” error message for a period of 3 days. Turns out that over 40% of customers received this error message!!

The results indicated that my payment radio buttons were a clear usability problem with my website. So ultimately, I made a change to have the “Pay By Credit Card” radio button selected by default because paying by credit card is the most common payment method for our store.

Confirm Button Not Obvious Enough

Unfortunately, the troubles with this particular customer were not over. After making it past the payment page to the “confirmation” page, she simply abandoned her shopping cart and didn’t complete her purchase. After having just spoken to this customer on the phone, I knew that she wanted to buy so I was puzzled as to why she left without finishing the transaction.

So, I gave her a phone call to find out why. Turns out that she thought that she had already completed her order after entering in her credit card information. Now if you do a fair amount of shopping online, the way most shopping carts work is that after entering in your payment information, you are directed to a confirmation screen where you can verify your shipping address, your billing address and your shopping cart contents before you actually put the transaction through and pay.

This lady wasn’t aware that she actually had to click another button to complete her order so she simply left! After walking her through the ordering process over the phone, I decided to revisit my checkout process entirely and here’s what I changed.



First off, I duplicated the “checkout timeline” graphic so that it was present at the top and bottom of the page. This timeline informs customers exactly where they are in the checkout process. In addition, I decided to make the “confirm” button a little more prominent so it would be less likely to be missed. Hopefully, these 2 changes will prevent other customers from having the same problems.

The Drama Wasn’t Over Yet

After finally processing her order, I thought that the worst was over. In fact, we shipped out her order immediately and 3 weeks passed so I thought everything was all good. But then she called to complain that she never received her order. Once again, I was puzzled.

And it was only after confirming her address over the phone did I notice a major flaw in our shopping cart. It turns out that the address field in our checkout form was not long enough to hold her entire address correctly because her address was abnormally long. In fact, I don’t think I’d ever seen someone try to cram so much information into the address field ever! The upshot was that there wasn’t enough room to include the suite number for her building so a few digits got cut off at the end of her address. Once that issue was resolved, I made a change to the shopping cart to allow unlimited characters in the address!

The Bugs Never End

We’ve been using the same shopping cart design for a few years now and we still find small issues all the time. It just goes to show that you need to be constantly aware of usability issues with your shopping cart and nip them in the bud as soon as they are discovered. Make sure you listen to your customers because they are ones paying the bills.