7 E-Commerce Metrics Every SMB Should Track

If you're in the processing of building out your e-commerce software

I spoke with Jimmy Rodriguez, Chief Operating Officer at shopping cart software provider 3dcart

1. Conversion Rate
Conversion rate is the most important statistic to measure when trying to determine how effective your website is at driving sales. Conversion rate math is simple: What percentage of website visitors actually purchase a product? If 100 people come to your website every hour, then industry math suggests two people should make a purchase, according to Rodriguez. Sure, it's nice to drive significant traffic to your website but, without measuring your conversion rate, it's difficult to determine if your website is optimized for sales. Similarly, conversion rate can tell you if your marketing is effectively recruiting likely buyers (as opposed to just generating clicks from people who would never have actually purchased a product).

Rodriguez suggests plugging a tool such as Google Analyticsemail marketing

The Snapchat Copying Continues With Facebook 'Messenger Day'

Hey Elon, Don't Make Me Spend 30 Years in a Pizza Hut on Mars

9 Tips for Designing a Great Web Storefront

Robert Herjavec's 5 Tips for Small Business Success

---

2. Bounce Rate
Driving traffic to your website can be difficult, especially for smaller businesses with unflashy products. That's why it's important for you to determine if you're driving the right traffic or if you're bringing people over who don't actually care about what you're selling. Bounce rate is measured as the percentage of visitors who leave your website without clicking into a second page. For every 100 website visitors, 64 should navigate to a second page, Rodriguez said. If you're averaging a bounce rate that's higher than 36 percent, then your marketing efforts might be misplaced.

However, just because you're seeing bounce rates that are higher than 36 percent doesn't mean you need to start from scratch. A high bounce rate isn't bad if you've been paying for untargeted advertising. There's no way for you to know who these customers are that saw your ad and there's no way to know if they had any interest in your products. Conversely, 36 percent would be an awfully high bounce rate if you were conducting highly targeted advertising (such as an email blast/sales promotion to a list of your former customers). Using this number to determine how effective your ad spend and your website landing pages are can be crucial for future success.

3. Average Order Value
You probably already know how much it costs you to recruit one customer. You should use your e-commerce software's analytics tool to determine the average order value to determine if you're spending too much money gaining new customers.

"This is a unique metric depending on what you sell," said Rodriguez. "As a business, you need to know this value to find out what is an acceptable cost of customer acquisition." If your average order value is $30, then you don't want to spend $20 to acquire each new customer. You would be dedicating way too much of your revenue to customer acquisition. Figuring out this number could save you tons of money in the long run or it could help you to determine how better to allocate resources for customer acquisition.

Additionally, you should use this data to project what the customer lifetime value for each customer could be. "If I learn that client will spend $40 per order every month, I can spend a bit more to acquire that customer," said Rodriguez.

Image via Statista

4. Navigation Behavior
Analytics tools such as FullStory

The heat maps will show you where people are spending the most time on your website, on which links they are likely to click, where they are scrolling, and even what they're reading. "Most people, when they read a paragraph of text, they mouse over the text," said Rodriguez.

If your heat maps reveal patterns in how people navigate your website, use these patterns to funnel visitors where you'd like them to go. Or, if necessary, completely rebuild your navigation to make the purchase path clearer.

5. Shopping Cart Abandonment
An entire article could be written about the tactics used to re-acquire customers who have abandoned items in checkout. The average rate of abandonment compared to completed transactions is 74.5 percent, according to a SalesCycle report

Common abandonment rate issues include complex checkout paths, lack of payment options, and long load times. You can use GA and your e-commerce software to determine what your abandonment rate is and then use a tool such as FullStory to track the path of abandonment to find out where the problem is.


Image via OneUpWeb

6. Multivariate Testing
It's important that you constantly test your website's layout to determine what works and what doesn't in order to maximize clicks and conversions. Multivariate testing helps you determine the best combination of variables out of all possibilities. For example, do you convert more customers by using product videos than you do using images? Do shorter headlines work better than long headlines? Do product videos and long headlines work better than images and short headlines?

You can send multiple versions of the same page into the field to see which elements perform best and then finalize your optimal design for mass consumption. There's no limit to the amount of testing that can be done, so make sure you constantly test, tweak, and test again.

7. Mobile Versus Desktop
Herein lies the rub: Everything I previously mentioned needs to be tested on a desktop browser, but it also needs to be tested on smartphones and tablets. That's because even responsive

"Tablets and phones [are] a completely different behavior than desktop," said Rodriguez. "Split these metrics and determine if you need to improve the mobile or desktop experience." Most e-commerce tools, coupled with GA, will give you a mobile versus desktop overview, so it's not necessary to go out and buy a new tool to conduct this research. However, you'll need to manually run these tests to gain the proper perspective for each medium.

This article originally appearedPCMag.com