Months later the radiologist would see that note, call the surgeon and ask how the patient was doing.
When his original diagnosis turned out to be correct he was happy. When he was wrong he would take another look at the x-ray to see what he missed. He learned a lot. Over time he became better and better at his work. Among surgeons, he's the most highly respected radiologist they know - all because of this feedback system.
Using B2B lead generation as an example, this post shows a clever way to set up a feedback system and obtain actionable insights from your web analytics app. With a little imagination this technique can be used for any type of online business.
This technique starts with anecdotal evidence. The trick is to start with individual cases and then use web analytics to see if there's a pattern. If you want a steady stream of new, productive ways to explore your analytics data, this is for you.
Summary Totals are often requested by people sporting fancy job titles, but for people who actually roll up their sleeves and optimize advertising and marketing for a living, it's valuable to look at things one x-ray at a time.
Too much is lost when data is totaled, averaged and rolled up. I want to see the exceptions that become invisible when looking at an average. I want to x-ray the data.
Step 1: Send Yourself an Email Every Time a New Lead Appears
I started doing this for purely personal reasons. I was making campaigns - AdWords, SEO, email marketing with Marketo. I wanted to keep track of results. Google Analytics shows the big picture and I could segment that data to look at smaller groups. Still, there was a lot I could not see.
Google Analytics Flaw: You Can't Isolate a Particular Person. People fill out forms, giving us their name, email address and other information, yet Google Analytics is severely flawed because the data is anonymous. I can't see reports for any specific person.
Segmenting the data - by Visitor Source, Engagement or Outcomes - is a step in the right direction, but the ultimate segment is made up of one person.
Who cares if a radiologist is 95% correct? To improve we must look at the 5% of cases that are wrong - one at a time - and discover the mistakes we are making.
I wanted something better, so I decided to sit right down and send myself an email every time someone filled in a form on our site.
With Marketo this is ridiculously easy.
However you implement it, each email can show:
- Everything the prospect filled in on the form.
- Often, the search phrase they typed into Google is available.
- If they arrived at our site by clicking a link in an email we sent them, I see that.
- If they arrived by clicking a link on some other site, I see that.
- I see the person's name and how they arrived at our site.
Google Analytics Flaw: The Attribution Cookie Keeps Getting Overwritten. After visiting five times, someone searches Google using our company name and fills in a form. With Google Analytics, we see lots of cases like this and the information is useless. With Marketo, it's easy to include information in my email about the campaign that originally brought them to our site. A clever programmer can use a cookie to make this information available even if you don't use a solution like Marketo.
This is so much better than Google Analytics.
Step 2: Make the Alert Email Exceptionally Clear
- I made labels and headings descriptive.
- I arranged the information in the email into logical groups.
- I customized the email headers:
- When they fill out a form from the "Summer Campaign" the top of the email is yellow and has an image of the sun.
- With color headings on the emails I can identify campaigns at a glance. Yellow for the Summer Campaign, Blue for AdWords, Green for Organic, Red for referrals from other sites.
Step 3: Share a Little
Once the email alerts were polished, I started sending them to other marketers. It's fun to get an email showing that a campaign you created is getting results. People love these emails, each one a snapshot of someone new contacting our company.
While statistics can be abstract, people love stories. Each of these emails tells a story about a person and how they found our company.
Step 4: But Wait
Set up a delay, so you receive the email two weeks after the visitor completes the form. Copying the radiologist extraordinaire, I have set up a feedback system!
Looking at these emails right away it's fun, but if I did that I would not have a clue about the outcome.
If I wait two weeks, I can look up the person in Salesforce and see what our Sales Team has discovered.
- There might be a note by the sales person saying, Not Interested.
- On the other hand, the prospect might have been immediately assigned to a top sales person. Oh my, this must be a hot one.
Don't naively cheer when a new lead comes in. Instead, wait a couple of weeks to look so you see whether it's a quality lead.
You will discover amazing things. Case by case, you will develop a feel for what is working and why.
- You'll see lot's of new prospects come from a Google search for a particular keyword, but then see they turn out to be duds.
- For every campaign, every landing page, every link bringing in people from various referring websites, you'll accumulate a visceral gut knowledge of what works and what doesn't.
- You'll start to rely less on uninformed opinions. Your daily actions increasingly will come out of a pragmatic assessment of what works. You will have killer marketing skills.
The great David Ogilvy - the best Ad Man to walk the streets of Madison Avenue - used this exact method. He got his start as a young man writing small classified ads and placing them in magazines.
Ogilvy wrote a variety of ads and tracked the results for each one. This feedback system improved his skills. Later Ogilvy remarked that at a glance he could spot advertising created by people who track results - and he knew they were bound to be successful.
The bits of information from this feedback system will lead you to explore your web analytics app in new ways. As you see example after example of what's working and what's not working, you'll sharpen your skills based not on hunches but on real examples.
True, this is anecdotal evidence, but now you can turn to your web analytics app to see if there is a pattern. You'll see views of data you might never have thought to look at, but for the curiosity that comes from seeing individual cases and wondering if there might be a pattern.
Step 5: Now Optimize
You see it every day now in your alert emails. You become a very pragmatic person. When you see something that works it becomes natural to check analytics and see if there is a pattern. When you see something that is not working it becomes natural to fix it and use the feedback system to see whether your fix is effective.
Roll ups are nice. Seeing the big picture is swell. Having your finger on the pulse of each marketing heartbeat is where the action is.
How I Did It: Doubling New, Incoming Business
I wish I could tell you this happened quickly. I was receiving lots of emails. It took months before I finally investigated something that seemed to be repeating.
Often I noticed people arrive at the web site from Google as a result of entering an odd query. Most companies have trophy keywords. These are the ones the CEO asks you about: Why do we not appear #1 in Google for this query?
But now I was seeing the opposite. Here was an odd search query that no one thought to optimize for. We appeared on the first page of Google when people searched using this query, but it was purely by accident.
Curious, I ran the query and clicked on our page. That's when I saw that the page did not speak to these people. Although it contained their search phrase, it provided no helpful information.
Next, I checked Google analytics to see if there was a pattern and there sure was. We were getting lots of these visitors but the bounce rate was 85% - of course - because we had no helpful content for them.
I've heard Matt Cutts say that people use search to find helpful information. So I wrote a helpful Ultimate Guide (five pages) that addressed the topic of their search query.
Two weeks later, there was a honking big orange button on the web page they used to bounce off, offering an Ultimate Guide. Thanks to an extraordinary marketer I was working with - Barbara Burlington - a click on the orange button led to an award winning landing page she created - it converted at 30%.
The button was large and orange, with a font so large you could read it from across the room, and it contained the exact search query people used to find this page. People love it when you mirror their language back to them. It shows you hear them and are offering what they want.
People clicked that orange button, causing the bounce rate to fall dramatically.
After clicking people arrived at a landing page that offered the Ultimate Guide. The free download was relevant to what they had been searching for. No wonder 30% of visitors filled out that form.
A few months later I heard that the monthly leads summary prepared by the Sales Team had a new category - just for leads from this project! Suddenly the number of leads had doubled and the sales people were loving them.
All this - just by doing what the radiologist and David Ogilvy had done to be successful: Looking at individual cases, finding a pattern and running with it.