Evaluating Success: Key Performance Questions to Ask Yourself

We research our competitors, create robust content strategies, set goals, produce content, and track as much data as we can. However, what good is all that data without the insights and recommendations to help move your business forward?

“Once you understand your customers’ preferences and tastes, you are in a much better position to create a better experience for them,” says Daniel Waisberg, author of Google Analytics Integrations: Centralizing Digital Marketing. “And a better website experience will inevitably lead to a happier customer.”

All good analytics start with great questions. To help you improve your digital marketing efforts, here are some of my top key performance questions (KPQs) for you to begin asking yourself and/or your analytics team that will help you work your way to deeper findings.

What are my big-picture content goals?

• This is the necessary first step to get your mind where it needs to be to guide you through the process and help focus on primary insights.

Where can I gather my data?

• If you’re looking only at website data within Google Analytics or Adobe Analytics, you may not be seeing the whole story and can trap yourself in an echo chamber.
• Tip: Remember to leverage tools like the analytics within your social channels, social listening keyword monitoring reports, targeted activation reporting, etc., so you can see the big picture.

What data models should I use?

• The Easiest Analysis: Regression Models — This is a simple statistical process that allows us to estimate and interpret relationships between different performance metrics. Regression models use historical data, allowing us to learn from our past and make better decisions about the future.
• For Testing Analysis: Predictive & Prescriptive Models — This is when you gather and review current and historical data sets to predict future possibilities, including alternative scenarios and risk assessment. This is great for building recommendations around, “If we do this…this will happen.”
• For Audience Insights: Cohort Analysis — When you want to quickly compare how different groups of users behave over time.
• For Content-Specific Analysis: Composite Scoring — Best if you’re looking only at one group of content and you gather enough data for statistical significance. You determine what metrics to consider, prioritize them and then factor the weighted scores to give all pieces of content a single score in the end to judge effectiveness for your goals.

What sources are sending traffic?

• Pay attention to the behavior for each traffic channel (source and medium). Just because some sources provide a large amount of traffic doesn’t mean they’re the best, most relevant source for your products and/or content. For example, your paid media team could be driving lots of traffic to your site, but if the traffic is not as relevant as it should be, or the landing pages served aren’t what users are expecting, metrics like bounce rate, time on site, scroll rate and other engagement factors will be poor and skew entire data sets at an aggregate level.

How are most of the new visitors coming to my site?

• Always ask this question, but don’t stop there! I recommend looking at where most of your new traffic exits the site as well. This is the content you need to tweak to better meet the needs of this audience.

What terms are driving organic traffic to my site?

• Make sure to inspect pages driving non-branded organic traffic to your site. Look at their URL structure, metadata, keyword density, text to html ratio, and inbound links pointing at the site to learn what you need to do to other pieces of content not as visible organically.

Which content is the most popular?

• Review both top pages by page views as well as top landing pages by page views to see the different engagement metrics associated with each report.

What content types or categories perform best?

• • This is where content tagging comes in handy because you can create custom dimensions to track content in specific ways. This saves you time so you don’t find yourself down the rabbit hole of advanced segments and potentially missing content here and there from cluster groups.

How successful is the content?

• While the bounce rate is an important KPI, don’t judge the success solely on this metric. Instead, include things like scroll rate and time on page.

How are users reacting to slow load times?

• You may see this information within GA if you have your Google Search Console connected to your account. This is a good sanity check to make sure the bounce rate for great content is really the result of restless users—and not the content or imagery you use on the page. For example, it’s often not the content or products that make users bounce; it’s the slow load time for that image that wasn’t properly compressed before uploading.

What are users actively searching for?

• If users are actively searching for content and/or products on your site, it’s important to learn what they want.
• Tip: If you aren’t already tracking the internal site search data, do so! I also recommend you set up event tracking for clicks to all primary navigational links.

Are users doing what you want them to do on each page?

• Meaning, are users clicking on my buttons, watching my videos, downloading my assets or engaging with various other custom events and/or goals?
• Tip: If you discover through event tracking that most users start videos but pause or exit them after the 50-second mark, tell your video team to make shorter clips in the future. No need to pay for a 120-second spot; you could possibly stretch your creative budget and be more effective at engaging your audience.

What are the most critical goal abandonment points?

• Whether it’s the cart process for an e-commerce site or an online registration or sign-up, anything where a user must follow certain steps should be tracked so you know what’s stopping them from completing the task.

–––

Big data is a big deal for all companies, and in 2017 I expect them to continue to spend more and more money to acquire information. Yet regardless of how advanced your analytics and data-gathering infrastructure is, it won’t provide you with obvious insights and solutions to lead to data-driven recommendations unless you’re asking the right questions.

Keep digging—the answers are there.