Which Metrics Should Your Web Business Track?

John Tabita

Knowing your numbers is critical to the success of any web business. After all, if you can’t measure it, you can’t manage it isn’t just a weight-loss mantra: it also applies to web metrics. But knowing which metrics not to track is equally important. After all, if you can’t manage it, why measure it?

The metrics you measure will differ depending on the type of site you run. For the remainder of this article, however, I’m going to assume the purpose of your site is to generate a lead or sell a product.

Your Website is Your Sales Person — So Treat it Like One

For years, I supervised a telemarketing team whose job was to set appointments for our sales reps. In order to measure our effectiveness, I tracked these three metrics:

  1. Number of calls
  2. Number of decision-makers reached
  3. Number of appointments set

I monitored these particular numbers because they told a story:

  1. I knew exactly how many phone calls an appointment setter should be able to make each day. So if the number of calls was too low, it indicated a lack of effort.
  2. I also knew how many decision-makers the average appointment setter should be able to reach in a typical business day. Once you have a decision-maker’s name, reaching them was simply a matter of persistent calling over the next week or so. A low decision-maker to call ratio meant a lack of follow-up on the part of the appointment setter.
  3. If the number of calls and decision-makers reached were at goal, but not enough appointments were set, then the issue was lack of skill.

Once I had the story, I knew what to fix. It makes no sense to provide additional training to an appointment setter who is not making enough phone calls. Conversely, pushing an unskilled telemarketer to make more phone calls isn’t going to increase the amount of appointments they set.

So it stands to reason that driving more traffic to your website isn’t going to increase sales if 80 percent of your visitors leave because it’s a poorly-designed site. If you don’t know your numbers, you don’t know the story.

The Rest of the Story

I also tracked a number that was beyond the control of my appointment setters: how many appointments led to a sale.

Once an appointment was set, the appointment setter had no control over whether the rep would close the sale or not. So why track that?

For starters, a closed sale is the offline equivalent of a conversion, which is when someone visits your website and takes the action you intended them to take, such as making a purchase or calling your business. Whether you’re a sales person or a website, your conversion rate is a key performance indicator you should measure.

But, more importantly, I tracked our close/conversion rate to justify the value my department provided to the company, by showing return-on-investment. If a company pays $70,000 a year in salaries and commissions to a telemarketing team that only generates $50,000 in sales, that’s bad.

Likewise, if you spend $800 a month for advertising that only produces $650 a month in revenue, that’s also bad.

Keep in mind that your ultimate goal is not to increase revenue or sales. It’s to make a profit. So the most important metric is return-on-investment. Will your return exceed its cost?

Here’s my point. If you’re selling products directly from your website, it’s easy to track sales. But if your website’s a lead-generation site, you still ought to track actual sales offline to justify its cost.

Here’s a completely fictitious example of what I mean.

Click-Throughs, Conversions and ROI — Oh My!

Wally’s Widgets spends $150 a month for his online advertising and gets an average of 110 click-throughs a month. His conversion rate is 2.9 percent, so his advertising results in 3 sales a month.

His average sale is $85, so by deducting the $150 monthly cost we see he’s made $105 from his advertising:

3 sales @ $85 ea. = $255
$255 – $150 = $105

Gary’s Gadgets spends $400 a month for his advertising. His conversion rate is also 2.9 percent, with the same average sale—$85. But due to his higher ad spend, the monthly click-throughs from his increased exposure is 355, resulting in 10 sales.

So when we deduct his monthly advertising cost from $850 in sales, he makes $450 a month:

10 sales @ $85 ea. = $850
$850 – $400 = $450

Since both are getting a positive return on their investment, each advertising program could be considered a success.

But wait. There’s more …

The True Cost of Your Advertising

I’m sure Wally’s happy to be getting a positive return on investment. But would it surprise you to know that, although Wally is spending less per month than Gary, his advertising is actually more expensive? Here’s why:

Wally’s advertising resulted in 3 sales a month. At $150 a month divided by 3, his cost-per-sale is $50.

Gary, however, received 10 sales a month, for which he pays $400, so his cost-per-sale is only $40. Although both are getting a positive return on their investment, Wally’s advertising is more expensive, because he’s paying $10 more per sale than Gary, even though he’s paying less per month. Make sense?

Remember: Cost-Per-Sale (CPS) and Return-On-Investment (ROI) are the true measurement of what your advertising’s costing you, not cost per month.

Click-Through Rate (CTR), Cost-Per-Sale (CPS), Conversion Rate (CR) and Return-On-Investment (ROI) are all key performance indicators every web business should measure. Beyond that, what you chose to measure will depend on your site’s purpose. The nature of the web makes it easy to track numbers ad infinitum. But before you can know what to track, you must first decide why you want to track it.

Remember, if you can’t manage it — i.e., affect or change it — why measure it?

Free book: Jump Start HTML5 Basics

Grab a free copy of one our latest ebooks! Packed with hints and tips on HTML5's most powerful new features.

  • Aaron

    It should be noted that some of the numbers can be quite difficult to calculate for some business models. For instance, the company I work for has something like an 18-month sales cycle, and we operate on a subscription model. Individual purchases are huge. Boiling that all down to a straightforward ROI number is not a simple endeavor. Further, our CPS number needs to include potentially quite a few factors other than just web ads. A total CPS in the many, many thousands of dollars would likely be reasonable for us.