All too often, I’ll ask a client a question, and they’ll readily have an answer. But when I question the answer and ask for a source, they’ll tell me, “We just know.”
I don’t have to explain why this is SO wrong, and yet we all do it all the time. Instead of finding proof for our assumptions, we base many of our decisions on what we think we know. This is a terrible way to do business: It will negatively affect our decision-making. Backing up decisions with real data is critical in digital communications, where assumptions could translate into daily mistakes because of the speed at which our customers consume information.
Applying what we know to what we don’t know
How can we fix this? The answer is in our analytics. The problems with analytics are most people don’t:
- Understand them
- Want to understand them
- Have the ability to translate what they really mean into facts that inform decision making
Important note: Understanding your analytics is only one pillar to informing your decisions. But it’s a very important pillar, because if you DO make a change, you will be able to track if that change really makes a difference in user behavior patterns. Scott Frangos just outlined a few other optimization tests you can run to get answers to your questions.
Pretend they are five: Workaround the analytics problem
If you want to get people in your organization using analytics instead of gut instincts to make decisions, you need to sell the importance of them and show people how to use them.
There’s a classic episode of The Office where the Scranton branch has a budget surplus they can use at their discretion. Oscar, the head of accounting, is trying to explain this concept to Michael Scott, the oh-so-clueless branch manager. Michael tells Oscar, “Pretend that I’m five.” Oscar, without missing a beat, says “Let’s say your parents give you money for a lemonade stand….”
While I trust your team is savvier than Michael, most people really don’t understand analytics and struggle with some of the most basic concepts such as, what’s the difference between a page view and a unique visit? How do you calculate a bounce rate? What does it even mean?
Step #1: Ask what they need
People only care to learn about something if it will help them. So, start by asking your stakeholders what questions they:
- Need to know
- Want to know
- Would help them make better decisions
It’s then your job to find the best metrics within your analytics to answer those questions.
Example:
Let’s say you are a wholesale online plumbing parts company, and a certain product line is being completely ignored. Your team knows these parts are superior to another product line you carry, and they want to push these superior parts. The team is exploring lowering the price, using customer testimonials and supplying extra content on this line of parts. They have questions such as these.
Question: How popular is our website? How popular are those products?
Metric: General traffic to those product pages (page views, unique visits)
Question: How are people finding out about our products online?
Metric: Search terms that help users land on those pages (keywords)
Question: What’s motivating people to buy? Are they reading the copy about the products?
Metric: User behavior on those pages (bounce rates, time spent on those pages, where users go next after those pages, shopping cart entries, etc.)
Once you answer these questions, more questions will inevitably arise such as:
- What is the traffic on the other product’s pages?
- Are those pages set up differently?
- Do we use other techniques beside written copy to drive business?
- What have we done in the past to highlight products?
Step #2: Train your team to understand the different parts of analytics
Once you know the questions your team has, you can start by providing them with answers with some basic analytics– in a very easy to understand format.
Try running a few workshops like “Understanding our analytics in 10 minutes.” There are three benefits to training your team to understand the terms around analytics and how powerful they can be, including:
- Understanding how to get at the answers to their above questions
- Finding answers to questions they didn’t even know they had
- Setting their own goals to metrics found in analytics
How to Train
Most people will not admit they don’t really understand, so hammer the basics of analytics over and over.
I recommend using real-life analogies to illustrate concepts, such as:
- Visit: When a user visited the site one time, like a shopper who walks into a store one time.
- Page Views: When a person is reading a book and reads one page.
- Bounce rates: When a person opens a book, reads one page, and closes the book.
- Unique visitors: How many people came to the site—like visitors to a hotel. They don’t count the number of people through the door every day, but rather they count the number of rooms they sell to how many visitors, excluding those who visit more than once.
- Pages/Visit: How many pages a person viewed/read spent during one visit to the site.
- New Visit: What percentage (number out of 100) came to the site for the first time. This could be compared to season ticket holders versus those who come just for one game.
- Traffic sources: Where your users came from. This is similar to media tracking—did they find you on TV, the radio, from a print ad? Your analytics can tell you if they came from other referring sites (sites with links to your sites), from paid keyword ads and from organic searches.
- Keywords: Which words they used and clicked on to find your site.
Once your team understands these concepts and ties them into key questions, they can start making decisions about how to push the superior plumbing products. For instance, if they know visitors come from search engines with certain keywords that the other product line is using in their descriptions, they can change the content accordingly. If they know the home page or landing pages get a lot of traffic, they can move those products up higher on those pages.
Next week, we’ll look at three more steps you can implement to achieve a strong analytics program.
Do you use analytics in your organization? Do you find some of the same challenges we discussed above? Or are you experiencing something quite different? Please share them, along with your favorite episode of The Office.
