How to Improve Content Marketing with IQR: Part 4

How do you know whether your content game is getting better or worse? It’s easy to rely on stock analytics tools, and for the beginning content marketer, tools such as Google Analytics are more than enough. For the veteran marketer who is creating content, how can we know with greater precision whether our content is getting better or worse? How can we more quickly diagnose the bad, double down on the good, and make our program sing?

In part 4 in this series, we look at the mathematical gulf separating our best content from our worst, and what that gulf means. If you haven’t read and done the exercises in part 1, part 2, and part 3, go back and do those first.

The difference between the best and worst performing content is the difference between the boundaries of the upper and lower quartiles. This is called the interquartile range, the IQR. The IQR (which we’re finally getting to in part 4 of the series) is a way to mathematically express how far apart our best and worst are. If the IQR were very tiny, it would mean very little variation between our best and worst content. If the IQR were gigantic, it would mean extreme variation between our hits and misses.

Let’s look at the IQR in our sample spreadsheet. What do we see?

Screenshot_7_6_15__6_24_AM.jpg

We see that except for a significant anomaly in the middle of the chart, the IQR otherwise appears relatively constant. Let’s add a trendline:

trendlineadded.jpg

We see the trendline remaining flat as well. A flat trendline indicates our content popularity isn’t increasing or decreasing.

What does this tell us? The IQR helps us to understand our relative content popularity. An increasing spread means the popularity of our good and bad content has diverged. Something’s changed which has changed how people perceive our content. A decreasing spread means our content is becoming more homogenous. The difference between a great post and a bad post, in terms of popularity, is shrinking.

The IQR lacks context, however. For that, we’ll need to refer to the previous chart:

5c0291bb-0da5-4eaa-a29e-dd33c31e0f6d_copy.jpg

The spread/IQR tells us how fast things are diverging (in examples 1, 4, and 7) or converging (in examples 3, 6, and 9).

If both your great stuff quartile and your bad stuff quartile are ascending (see the previous post and examples 1, 2, and 3 above), then a converging IQR may not necessarily be a bad thing. The rising tide of popularity will lift all content.

If your great stuff and bad stuff quartiles are both descending (examples 7, 8, and 9 above), a converging IQR is really bad. Everything is losing ground, and you need to pivot immediately.

This concludes our in-depth look at interquartile ranges and how to use them to measure your content marketing. Try these techniques out. You don’t need to do them more than once a month, but you should test to determine how well your content is doing, using any relevant content marketing metric.


If you enjoyed this, please share it with your network!


Want to read more like this from ? Get updates here:


Marketing Blue Belt Preorder

Subscribe to my free newsletter!


How to Improve Content Marketing with IQR: Part 3

How do you know whether your content game is getting better or worse? It’s easy to rely on stock analytics tools, and for the beginning content marketer, tools such as Google Analytics are more than enough. For the veteran marketer who is creating content, how can we know with greater precision whether our content is getting better or worse? How can we more quickly diagnose the bad, double down on the good, and make our program sing?

In part 3 in this series, we begin to extract more meaning from the data we’ve collected in part 1 and charted in part 2. Be sure you’ve read and done the steps in those early parts first.

Let’s now consolidate the graphs of the good stuff and the bad stuff into one picture, so as to see everything more comprehensively. I’ll reapply the trend lines as well:

Screenshot_7_2_15__6_34_AM.jpg

We can see that both trendiness are going up. Let’s start with that basic form of analysis and examine some different combinations and what they could mean.

5c0291bb-0da5-4eaa-a29e-dd33c31e0f6d_copy.jpg

Above, there are a total of 9 scenarios you might see in your trendlines. Let’s explore what they are and what they might mean. I’ll remind you from yesterday that good stuff refers to the best 25% of your content, the most popular content. Bad stuff refers to the lowest 25% of your content, the least performing content.

Scenarios 1-3: Content Marketing working well

[1]: Good stuff ascending faster than bad stuff. This is the best possible situation. All your content is improving, but your headliners, your big content, is punching above its weight. Keep doing what you’re doing, and double down on your best ideas.

[2]: Good stuff ascending at the same rate as bad stuff. This is a sign of an overall strong content marketing program, steady improvement across the board. The next important thing would be to develop some big ideas and amplify the great hits you’ve already got.

[3]: Good stuff ascending slower than bad stuff. Your least performing content is making strides to become better. Now’s the time to start dreaming up some big ideas to take your best stuff to the next level.

Scenarios 4-6: Content Marketing might be in trouble

[4]: Good stuff ascending while bad stuff descends. You still have great hero content, but your maintenance content is suffering. Either you’ve tapped out your audience or your content simply isn’t of interest most of the time. Find someone to do a better job with the topics and content formats you’re not good at.

[5]: Good stuff and bad stuff remain neutral. Your content marketing is working okay, but not improving. This is a sign that you need a jolt of creativity and different thinking.

[6]: Good stuff descending while bad stuff ascends. Often, this is a sign that you’ve spent so much time shoring up your weak areas that you’ve let the important areas go. Get your big ideas back on track.

Scenarios 7-9: Content Marketing definitely in trouble

[7]: Good stuff descending slower than bad stuff. Both areas are declining, but your top content still holds some influence. Use it to reboot your program. Do thorough analysis and throw overboard the types of content, ideas, and topics that are least performing.

[8]: Good stuff descending at the same rate as bad stuff. This is general bad news. You’re headed for the bottom. This is when you reboot everything.

[9]: Good stuff descending faster than bad stuff. This is the worst possible situation. Your best content is losing ground rapidly, and whatever traction your bad content has is probably so small that the rate of decline is meaningless, bottoming out.

If we take the chart from earlier and compress the axes down to just the ranges where the trendlines are, we can see which scenario is at work in my own data.

Screenshot_7_2_15__7_25_AM.jpg

What we see above is scenario 2. I now know what I need to do in order to move my content marketing program ahead.

In the next post in this series, we’ll look at measuring the distance between your good and bad stuff, and how to interpret that measurement.


If you enjoyed this, please share it with your network!


Want to read more like this from ? Get updates here:


Marketing Blue Belt Preorder

Subscribe to my free newsletter!


How to Improve Content Marketing with IQR: Part 2

How do you know whether your content game is getting better or worse? It’s easy to rely on stock analytics tools, and for the beginning content marketer, tools such as Google Analytics are more than enough. For the veteran marketer who is creating content, how can we know with greater precision whether our content is getting better or worse? How can we more quickly diagnose the bad, double down on the good, and make our program sing?

In part 2 in this series, we start to dig into the data we’ve collected and identify early opportunities. If you haven’t read part 1 to get your data, go and do it first.

We left off with all of our data in columns. Let’s tackle the bad news. How bad is the bad? Because this is social media data, I’ll prune out replies, leaving only the content I want to share. Once I’ve correctly sorted and cleaned my data, I’m ready to analyze.

Take the bad stuff column and chart a simple line graph. Depending on how much data you have, this may be a taller order than it sounds. Below, I’ve taken the bad stuff – my lowest quartile – and charted it out:

Screenshot_7_1_15__5_59_AM.jpg

This is tough to interpret, so let’s right click and add a trendline:

Screenshot_7_1_15__6_01_AM.jpg

In general, we can see that the worst of my posts, the posts that got the least amount of exposure, have still been on the rise. If we zoom in a bit, we can see that the trend in the lowest quartile has gone from about 1,800 impressions to a little over 2,200 over the span of 6 months:

Screenshot_7_1_15__6_07_AM.jpg

This is a solid improvement in the least well-performing content. The next step for me would be to go back over the data and identify when things changed. Was the improvement consistent over the same period of time?

What about my best stuff? How’s the boundary between good to great? Let’s repeat the same process, from making a chart to applying a trendline:

Screenshot_7_1_15__6_21_AM.jpg

We see improvement… but look carefully. The improvement from beginning to end in this six month timespan is shallower than we saw in the bad stuff. This tells me that the best stuff resonated more, but didn’t grow as fast as the bad stuff.

So, we know what the good stuff did. We know what the bad stuff did. Is there a relationship between the two? Is there some insight we can glean from both of them together? Stay tuned; tomorrow, we’ll look at the difference between good stuff and bad stuff, and how to interpret it.


If you enjoyed this, please share it with your network!


Want to read more like this from ? Get updates here:


Marketing Blue Belt Preorder

Subscribe to my free newsletter!