What's Important in Website Analytics

By Amir Glatt, Duda Co-Founder & CTO

There's no doubt that we live in the data era. Today, nearly everything we do online is trackable.

Even more importantly, new analytics tools appear every day to crunch this data and provide marketers and businesses with insights we simply didn't have in the past.

In the business world, companies that base decisions on data are more successful than those that do not. The culture inside the most successful companies in the world, like Google and Amazon, is data driven, and decisions are usually made based on what the numbers say rather than what the most senior person in the room thinks.

With the enormous amount of data that is now collected and crunched comes a new type of problem for businesses: How do you determine what's important and what's not? Internet professionals can spend hours, days even, looking at different metrics and stats but what should savvy Web workers focus on? Although all data is relevant in certain ways, the most important data, the figures which should draw the most attention from marketers (and the like), is the data that they can take actions on. If they want to discover something about their websites, but can't do anything about it, it's probably not worth their time. Let's look at the website statistics of a typical small business to demonstrate the difference between actionable data and excess analytics.

The business we will use in this article is a Phoenix-based restaurant called "The Gladly." As a restaurant, the main goal of its site is to drive new reservations, provide people with the information they need to make a dining decision and position the restaurant as a place people would like to visit again and again.

The Essentials

Image A provides a glimpse of The Gladly's actual website data as captured by Google Analytics. This kind of basic information is important for providing a general understanding of the website's overall performance. How many potential customers interact with this website? How soon do they leave the website?

A.

How many of them are new or previous visitors? Is the trend negative or positive? The most important insights we can gain from looking at this dashboard is the number of overall visits as it tells us whether or not the website, in its current form, is an important customer acquisition tool.

The bounce rate will tell us if people find what they're looking for when visiting the site. If the bounce rate is low, it's a good sign. If it's high, it means the website's content (or the layout/structure) should be improved, since people who landed on the site tend to leave without taking any action. A 27 percent bounce rate is considered very good in most cases.

Bounce Rate Too High?

Tip: Ask website visitors to answer a few questions when they leave to get a better understanding of why. See other ways to gather design feedback at wsm.co/desfeedback15

Device Details

One of the most important insights Web professionals can gather is on the devices used by visitors in order to ensure they are provided with the best possible experience. In The Gladly's case, 47 percent of site visitors are coming from a mobile device so it's very important that the restaurants audience is not walking into a UX nightmare (see Image B).

B.

Traffic Sources

There is a great deal of actionable information a website can extract from the traffic sources reported in its analytics. For example, as illustrated in Image C, it appears that 47 percent of the website traffic is coming from Google search, which indicates the site's SEO initiatives are effective and working well. Going down the list, we can also see 6 percent of visitors are coming from Yelp. Traffic coming from the local business review site generally indicates a high level of visitor interest in a business and there are a few ways The Gladly can take advantage of this information. For example, the restaurant could attempt to get more traffic from Yelp by buying a premium package or perhaps can convince (or incentivize) more people to write positive reviews on the platform to raise even more awareness and thus get more visits. Analyzing traffic sources can help enterprises better understand which channels are the most successful and determine where they need a greater investment in outbound marketing efforts.

C.

Information Organization & Optimization

Knowing which pages receive the most activity (visits, pageviews, etc.) is useful in optimizing the user experience as it enables enterprises to prioritize information that is in demand. In Google Analytics, companies can see which pages get the most views (e.g. thegladly.com/dinner, thegladly.com/contact, etc.). It is important to consider if there is important information on lower traffic pages that should be made more accessible to users. For example, if a large percentage of traffic is going to the contact page, moving a business's address to the homepage may benefit visitors, particularly those using mobile devices.

Goals, Events & Actions

Companies can use different techniques to track important actions taken on a website in order to discover the difference between first-time visitors and regular or repeat customers. For example, The Gladly sends events to Google Analytics when a visitor makes a reservation using OpenTable.

What is important about tracking this type of data is that it can help companies understand the actual dollar value of a website visitor, and further, determine an appropriate online marketing budget. Sound complicated? It's actually not. Let's assume that someone who reserves a table is worth $30 in profit to the restaurant, and 1 percent of visitors reserve a table. In that case, a website visitor is worth $30 for every 100. Therefore, if we can advertise on different platforms, like Google, for less than $0.30, this acquisition activity would be a profitable one for this business.

Data that can help make decisions depends heavily on the business, the situation and the goals. Try to focus on 5-10 things that matter the most and concentrate on them since, as we already discussed, there's just so much data and so little time.