Be a Better Web Analyst with Actionable Intent
December 28th, 2009 in Web Analytics
There are many ways to become a better Web Analyst, but what I’ve been most interested in lately is examining my actionable intent. Yes, that sounds weird. But the way to look at it is that actionable intent explores the way we look at core standard web metrics and the action we intend to take. As analysts, I believe the actions we need to take are most clear when the metrics fall significantly outside the norm. The problem is that we see the analytics most often inside the norm and we lose sight of opportunities. Analyzing a “spike” in visits is much easier than analyzing a “small bump”.
The way to best examine your actionable intent is to walk through an extreme scenario so you can develop some fresh insights. The output of this exercise is to strengthen your ability to make strong actionable decisions in circumstances when it’s not all that clear (or when it’s just the small bumps).
Scenario – Joe’s Hat Outlet (website)
![]() |
![]() |
| Figure 1 – August ‘09 | Figure 2 – Sept ‘09 |
Joe’s Hat Outlet is a small ecommerce website that features a small selection of baseball hats. As you analyze the data, Figure 1 shows 862 visits (29 visits/day) for the month of August. The trend line next to the number shows that 862 is pretty average for the month.
Now look at Figure 2. What would happen if you checked the metrics for Joe’s Hat Outlet one month later and noticed a major increase of 1000 visits? That’s over a 100% increase for one month! Before you go outside to hug the mailman you’d probably want to do some double checking. Time to investigate some possible causes and effects. At this point, you’re now exploring your actionable intent and this is where you can start taking down notes. You’ve created the extreme scenario, now create some hypothetical responses…it’s OK to get a little creative here.
Investigate the following when walking through the scenario:
- What questions am I formulating?
- Did Joe’s Hats go viral?
- Was there an article or blog post written about Joe’s Hats?
- Could the website handle an increase in visitors like that?
- Was the spike instantaneous or was it gradual over the month? - How would I investigate this (metrics)?
- Examine where visits are coming from.
- Examine bounce rate
- Examine conversions: increase or decrease?
- Visits coming from returning or new visitors? - What are some potential causes?
- Joe’s Hat Outlet was featured in an external source
- The outlet is carrying a rare hat
- Website got positioned high for a competitive keyword - What does this value mean?
- Visits more than doubled to the website in 1 month - What action would you recommend be taken?
- Make sure the website can handle the spike in visits
- Ensure that top entry pages have strong call-to-actions matching visitor intent
- Apply A/B testing to top entry pages - What outcome would you expect from that action?
- Server load testing
- Increased conversion rate
And there you have it. What you’re doing here is exercising your analytical mind to come up with some new ideas for taking action (actionable intent). What you’re taking away is a list of clear insights to apply to your smaller analysis needs. Take these insights and examine how you can apply them to the current situation whether it be in small does or on a bigger scale. This is a great exercise to break out every 6 months for proactive approach to developing fresh ideas and insights.
Bonus: Once you’ve completed that exercise for a potential spike, create the scenario from the other side of the spectrum – a major plunge in a core metric. Go through the same process of questions.


Twitter
Linked-In
Facebook
Digg
And eehhh what happened to Joe’s in that month ?