Commentary

Task Completion: Why It's The Most Important Metric

by Jonathan Levitt , August 11, 2009

Only about 20% of visitors come to an e-commerce website to make a purchase - so why do most marketers cling to the conversion metric as the Holy Grail of success? The truth is, measuring task completion is a much more valuable metric. People have taken time out of their busy days to come to your site for very specific reasons, and yet many marketers assume everyone who arrives at their sites is there to make a purchase. We know this isn't true, because research shows that 80% of site visitors are there to do something else entirely: browse, research, comparison shop, look up a store location, read your company blog, or any other task.

That said, task completion is deeply connected to conversion. Figuring out how many of your site visitors were able to complete what they set out to do is not only a critical measure of customer satisfaction, it can be a powerful crystal ball into future purchases. If you optimize your site to enable visitors to easily complete the tasks they want, they'll be more engaged with your brand and more likely to return later to make a purchase.

A survey by iPerceptions of thousands of websites across sectors shows that 60% of visitors who complete their tasks report a higher future likelihood to purchase, versus only 14% for those who do not complete their tasks. And even if you don't sell online, optimizing your site to enable visitors to complete their tasks also positively impacts offline conversions. People who get exactly what they need from your site are much more likely to visit your store or call an 800-number to complete a purchase; our survey data has especially shown a strong correlation between task completion and offline purchases for big-ticket items, such as cars, vacation packages, and appliances.

I know what you're thinking - another thing to measure! With web analytics, conversion metrics, campaign measurement, and customer satisfaction measurement already taking a lot of your bandwidth, you might think task completion is one metric you don't need to add to your plate. Well, it's actually very easy to get started measuring task completion - and if you ignore it, you're leaving valuable revenues on the table. To measure task completion, you don't need to understand complicated formulas or algorithms, nor do you need to buy expensive analytics software. All you need is to post a couple of simple "yes" or "no" questions that site visitors answer via a free online survey like 4Q. Beware, task completion is going to hold you to account from day one. You won't be able to hide behind fuzzy definitions of satisfaction or engagement - instead, task completion is a simple, cut-and-dry barometer of each visitor's success.

In fact, for many years, companies placed a huge emphasis on their Net Promoter score as a barometer of their company's health, brand recognition, and even stock value. Wouldn't it be simpler and more illuminating to just to ask your potential customers if you delivered what they needed, and if not, why not?

Here are three tips to help you start measuring - and benefiting from - task completion today.
• Ask the right questions. You can tap into the power of website task completion simply by asking your visitors a few quick questions. Implement a free survey like 4Q to ask your visitors: "What are you here to do?"; "Did you complete what you set out to do?"; and "If not, why not?"
• Invite feedback. Simple surveys are great, but sometimes visitors want to leave more feedback than just clicking a button. You can integrated open-ended feedback into your task completion measurement program by using a more sophisticated survey like iPerception's webValidator Continuous Listening Solution. Linguistic analysis can identify phrases, tone, and patterns in open-ended qualitative feedback, and assign it a quantifiable task completion value.
• Act on the data. When you ask your site visitors if they were able to complete the tasks they set out to do, the answers may not be pretty. But all of the data, whether it's good or bad, is extremely useful. If you find out your site is failing visitors who set out to complete a certain task, by all means fix that problem quickly. If you find out that the site succeeds in another area of task completion, don't rest on your laurels, but instead analyze what's working about that area of the site so you can extend that experience site wide.

commentary, metrics

3 comments about "Task Completion: Why It's The Most Important Metric".

Check to receive email when comments are posted.

Priscilla Jessup from Denmark {the agency}, August 11, 2009 at 8:03 a.m.
Great article. Appreciate the solid insights on the importance of helping visitors get their task done. Our interactive division is integrating simple surveys with positive results and very actionable learnings.
Reply

Chris Murdough from The Allant Group, August 11, 2009 at 8:55 a.m.

Great piece, Jonathan. Nice to see you're still out there advocating for VOC measures. Task completion is extremely insightful, especially when viewed by visitor segment -- perhaps visit frequency to the site or referral source.

Kevin Ertell from ForeSee Results, August 12, 2009 at 2:30 p.m.

I’m in total agreement that there is a significant gap between those who come to a site to make and purchase and those who actually do. Site conversion rates as we measure them today are not particularly helpful, and they are getting less helpful as we use our sites for more and more purposes. However, I am also very wary of other metrics that are overly simplified. While the allure of a simple metric is strong, I fear overly simplified metrics ultimately end up being less than actionable. You seem to allude to this fact yourself when you suggest using complicated linguistic analysis as part of your recommendations at the end of your post. I find the reality is that our sites are complicated, and figuring out how to solve for the gap mentioned above requires the analysis of millions of variables. Such analysis – especially when it includes analyzing customer mindset -- requires the more complicated algorithms you seem to dismiss in your post. If we want our analyses of our problems to be precise, accurate and actionable, we’re collectively going to have to get over our fear of numbers and embrace the types of statistical analysis that have been proven over the years to provide truly reliable and actionable data. It’s not a bad thing to be complicated when accuracy is the result.