Blog post

The Great #Macaroni Experiment

By Martin Kihn | September 11, 2017 | 5 Comments


It is impossible to know if anyone is actually reading your long-form content. You can count how many people looked at it, but not how many gave it real attention. Even — in the best case — finished it.

Which is why I did the #macaroni experiment.

On January 12 and 14th, 2016, I unleased a duet of long blog posts called “Top 10 Amazing Secrets of DMPs” — Part 1 and Part 2 — and combined they came to about 6,300 words of science. That’s the equivalent of two substantial New Yorker articles. And they’re by far my most popular blog posts of the past four years.

They provided evidence that the advice I offered in a different post was right: “The #1 Secret to a Successful Blog Post.” The #1 Secret: “Write longer posts.” In which I argue for depth, originality and more time in composition — not quantity over quality.

These vivid titles (“Top 10 …” and “#1 Secret”) happened at a time when I was experimenting with BuzzFeed-like clickbait to see its impact on views. My conclusion was that — at least in my own little sphere — readers are looking for information, not insults.

So what is The Great #Macaroni Experiment? At the end of part 2 of the DMP series, I inserted this request:

“Favor: If you’ve read down to here and got something out of this 6,300-word DMP series, please tweet me @martykihn. Just tweet the simple hashtag #macaroni.”

The idea came from a film by the waifish performance artist Miranda July. The film, “Me and You and Everyone We Know,” is an oddly charming rom-com wherein Miranda falls for a shoe salesman and submits VCR demo tapes — it’s 2005 — to performance art curators. Frustrated by the cavernous silence of these blind submissions, she puts a simple request near the end of the tape: If you’ve watched to here, please call my number and say the word “macaroni.”

Beautiful. I did the same. Here are the results.


A total of 42 #macaroni responses were tweeted to @martykihn between January 14, 2016, the day part 2 was posted, and last week. That represents less than 0.5% of the total number of people who actually viewed part 2 in that period, according to Google Analytics. Remember this is a very long post about a rather obscure, even technical, topic, and I am very happy to have engaged almost four dozen souls. My private goal was 10, so my own mission was accomplished.

Content marketers out there may be interested in the #macaroni tweeting pattern. This chart shows the dates and volume of #macaronis (blue line) as well as the cumulative running total #macaronis (red line):


As expected, there was a higher concentration of readers in the beginning — 30% of the #macaronis occurred within the first 30 days. I actually had six #macaronis on the day part 2 was published; my best day. Within 90 days, I was up to 50% of total #macaronis. You can see from the slope of the red line that the rate of #macaronis is gradually — although not rapidly — declining.

But what’s most interesting to me is that after the first 30 days, the rate of readership stayed fairly steady. It took 28 weeks to go from 50% to 75% of total #macaronis and another 28 weeks to get to 100% (today). After a few weeks, I stopped promoting the posts on my own Twitter feed and LinkedIn profile and @Gartner_Inc stopped as well, so any readers I’m getting after that period are finding me on their own, via search.

So what can we learn from this experiment?

  • First, that people will indeed read long content with close attention if it is interesting to them. Do not despair the attention span of the people. We do crave real news.
  • Second, that longer, more substantial content has a longer, more substantial life. Here we are at the 20 month mark and still receiving #macaronis. Given that the average tweet has a half-life of 24 minutes, a 3 month half-life for a 6,300-word series isn’t bad.
  • Third, you get out what you put in. These posts took me forever to research and write and were not really part of my job duties. The time showed in the material, and contributed to its longer shelf life. I can’t prove this; but I know it.

So there you have it. Thanks to all who participated. #macaroni

The Gartner Blog Network provides an opportunity for Gartner analysts to test ideas and move research forward. Because the content posted by Gartner analysts on this site does not undergo our standard editorial review, all comments or opinions expressed hereunder are those of the individual contributors and do not represent the views of Gartner, Inc. or its management.

Comments are closed


  • #macaroni

    I said that first, so you’d be sure I’ve read at least half of your article. (I read it to the very end.) And I gotta say it was fun and interesting.

    In our own blog, we’re used to writing materials always exceeding the 1000 words threshold and in average, getting up to 2000 words. And the feedback? some hate us… they really do… they say we’re just doing it for SEO or because we’re plain stupid.

    But that’s not the whole story: some read our posts to the end and even interact with us. And these positive feedback is way more than the negative one.

    I’ve never questioned our way, because I thought that we want true readers for our blog. Because they’re the ones who are really interested and who will spread the words about us. And now I’m happy there’s someone who agrees with us on the matter.

    Thanks Again for the great post

  • Interesting technique.

    And yes, it (partially) confirms the thought leadership measurement KPI’s I had put in place when presiding over the BearingPoint Institute over 6 years.

    One of the KPI’s was resilience, which was a measure of whether a topic was addressed with enough original substance and depth to make it relevant over time. Typically, market shares score best because readers seek validation (of what they instinctively feel), seconded by normative representation of a reality (e.g, frameworks such as Magic Quadrants).

    The other scoring keys were relevance, newsworthiness, etc.

    Thanks for posting insights that are fun to read, relevant, credible and resilient.

  • Chris OHara says:

    Just came across this analysis, Marty. I believe I gave you a day 1 #Macaroni. I thought your series was the best explanation yet for that a DMP does. #TheFourDozen