While PPC nerds certainly love their data, sometimes we can (admittedly) go overboard. We throw every metric, dimension, and number we can think of into a report, maybe add in some notes, sometimes changing the formatting, and call it done. Some clients like this strategy as they want to be able to see every single piece of the data. Others, not so much.
In today’s post, we’ll walk through how to cut out the clutter and get to the insights which drive our action items. We’ll also discuss how to find those insights from your data.
Don’t Include Too Much Data
How long is your report? Does every single item on it add value of some sort?
Your first step should be to remove all the unnecessary charts, graphs, screenshots, etc. I’ve seen reports that are pages and pages long with endless charts and visuals, and this really detracts from the main message. It’s harder to create action items when you have too much information.
Include the information that will help guide your decision or that is valuable to the client’s business. If you’re wondering whether or not to include something in a weekly/monthly/etc. report, think of these items:
- Does it answer a question or provide valuable information?
- Does the client/stakeholder care about it?
If not, it’s pretty safe to say that you can remove it without losing any sleep at night. This moves us to the next point, regarding how to use reports to answer key questions.
If you want to ensure that your report is giving you actionable takeaways, you may find it useful to work backwards. Think about what changes you’ll need to make in your account regularly. Then think about how you’d go about getting the data necessary to make those decisions. What metrics do you need to see?
- What question are we answering / which problem are we solving?
- What data do we need to make a decision?
- What is our decision criteria?
- How can we display this information in an easy-to-read format?
From there you’ll have a good idea of what to include in your report. Let’s take a look at another example.
Goal: where should we allocate more budget?
Information needed: conversions, impression share, CPA
Criteria: allocate more budget and bid increases to top converting, low CPA locations
Impact: estimated additional search conversions with increased spend
Of course these are just recommendations. A human touch is still necessary to make the final decision (maybe the client doesn’t want to spend more in one location due to business reasons).
Build Some Automation
If you are aiming to have your report just tell you explicitly what you should do, there are some areas that you can build in some automation.
Quality Score Example
In this example, we have used Supermetrics to pull in quality score factors from AdWords. These handy little charts auto-populate using query functions.
But so what? Why do we care about this? How can we improve quality score?
The next step in making this data actionable is to estimate potential QS gains by improving CTR, ad copy, or landing page relevance. To do this, we estimated potential gains by ad group by assigning a multiplier for the “average” and “below average” ratings. Then a simple IF statement is used to pick which will have the largest impact on the ad group.
The end result is an easy to digest table recommending which actions to take to increase quality score. For convenience, we’ve included priority/rank (based on total potential gain) and impression share.
Heat maps, while useful, can be a bit overwhelming for the user. How do we pull out specific account action items from a heatmap?
Here’s a standard dayparting heatmap, looking at conversion rates by hour of day and day of week.
Does it give you all of the information? Sure. Is it easy to pick out where you should actually make some changes in your account? Not so much. In this version of the heatmap, it recommends bid adjustments based on six-hour blocks of time (which is easier to digest and implement in your account).