How to write a media measurement report in 4 hours or less

PR pros know that how to write a readable PR report hardly gets mentioned compared to how to write a press release or a story pitch. Katie Paine fills that big gap step-by-step below.

As we’ve said elsewhere in this publication, the key to great reports is: Use your data to tell a story.

But teasing that story out of a pile of data is not straightforward. Whether you sum up your accomplishments for the year or summarize a bad day in the media, your report writing will be faster and better if you work in a strict sequence with intermediate goals.

Once you gain a little experience, you’ll be able to write a clear, effective report in four hours or less. Here’s how to plan your time and the steps you must take.

Hour 1: Get your mind around the problem

Who’s the audience? What kind of presentation will they prefer?

Let’s assume the report request came from higher up in your organization. Put yourself in the head of the boss of the person who requested the report. Ask yourself what type of person he or she is:

  • Are they a “word” person—a former journalist, editor, or PR person—who is most comfortable getting descriptive data?
  • Are they a “numbers” person, who wants just the facts and numbers and a chart or two?
  • Are they a “visual” person, who needs to see data in graphs and word clouds?

And if you don’t know what kind of person they are, make sure you use all three forms of presentation in your report.

What questions does this report address? What actions will result from this report? What data do you need?

  • If you report in the middle of a crisis, what data will people need to make the right responses?
  • If you report after an event, what data do you need to make the event better next time? If it was dreadful, what data do you need to not make the same mistakes?
  • If you prepare a monthly or quarterly report, what decisions do you want people to take after reading it? What data will help your team make better decisions?

Assume the worst about your data and plan to deal with it.

There are bound to be some inaccuracies in your data. Chances are good that:

  • You won’t agree with the coding
  • You can’t find a clip that you know should be in there
  • There’s a key publication missing

And if you use a vendor whose data you haven’t used before, assume there will be more than one thing wrong. There’s always a learning curve. Don’t panic. Very likely one or two problems won’t invalidate all the data.

Read over the data with a skeptical eye.

We’re not suggesting that you read thousands of clips, posts, and Tweets. But you must spend an hour digging in the data to see what you’re dealing with, not just to make sure it’s accurate, but also to get a feeling for what is happening and why.

  • Start with what everyone will look at first, the negative mentions. Make sure they are accurate
  • Then scan the positive clips. What’s happening to make the media or customers say good things about you?
  • Now filter out the dreck. Use the Excel filter to search for phrases like Viagra, Free, and PRNewswire (reprints of press releases should not be included in your metrics)
  • Read the clips from key publications, and review the writing of bloggers or authors with the most followers
  • What topics captured editors’ interest? What did people comment on?

The purpose is not to check every clip, it’s to become familiar with the media in the time you’re reporting on. Your goal for this first hour is to have a good idea of the story your data tells.

Hour 2: Illustrate the story

We might title this “Analyze the Data.” Unless you have days instead of hours to work, be careful not to get lost in the analysis. Any analytics will generate dozens of reports in all shapes and sizes. You don’t need dozens, so resist delving too deeply into the crosstabs. You only need charts that illustrate the story you will tell.

Every scenario is different. The following options will help you decide which charts or graphs to generate:

Scenario: Reporting in the middle of a crisis

What you’re trying to say, i.e., questions the report will answer:

  • How bad is it?
  • Do we need to respond?
  • Do we need to respond differently?
  • Is it a crisis or just an internal panic?
  • Which authors and media should we worry about?

Charts you’ll need:

  • Increase or decrease in coverage by hour or day
  • Increase in negative sentiment by hour
  • Sentiment by media outlet (for instance, Twitter vs. mainstream news)
  • Share of voice vs. the competition
  • Share of undesirable voice vs. the competition
  • What’s the gist of the conversation? (e. g., a word cloud)
  • Top authors by sentiment

Scenario: Reporting after a crisis or an event

What you’re trying to say and questions the report will answer:

  • Should we devote resources to this in the future?
  • What worked?
  • What didn’t?
  • What should we do differently next time?

Charts you’ll need:

  • Share of desirable voice vs. the competition
  • Share of desired positioning vs. the completion
  • Increase in online or website engagement (pre/post event)
  • % of coverage in top-tier vs. lower-tier media
  • Share of spokespeople quoted vs. the competition
  • What’s the gist of the conversation? (e.g., a word cloud)

Scenario: Reporting after a major product launch

What you’re trying to say and questions the report will answer:

  • Did we get our messages and position across?
  • Did we outperform the competition when they launched their product?
  • How well did our spokespeople get messages across?
  • Did we reach the audience we aimed for?
  • Did the reporters we briefed report accurately?

Charts you’ll need:

  • Share of voice vs. the competition
  • % of coverage containing one or more key messages
  • % of coverage containing desired positioning
  • % of spokespeople quoted
  • % stories that contain key messages
  • Breakout of coverage by media outlet, source and author
  • Ratio of desirable to undesirable coverage

Scenario: Monthly or quarterly report

What you’re trying to say and questions the report will answer:

  • Did our campaigns pay off?
  • What worked, what didn’t?
  • What should we do differently next month or quarter?

Charts you’ll need:

  • Share of desirable coverage vs. the competition
  • Share of undesirable coverage vs. the competition
  • Message communication by message
  • % of items containing one or more key messages
  • % of top-tier vs. lower-tier media
  • % of items containing one or more quotes from internal spokespeople
  • % increase in social engagement
  • % increase in unique visits to targeted URL
  • % change in awareness and preference compared to last reporting period
  • Coverage of programs or initiatives by tone and message (ranked from best to worst)
  • Desirable vs. undesirable coverage by geographic region

Throw out any charts that don’t contribute to your story line. Eliminate any charts that have insufficient data. Never put a chart in just because it was there before.

I recommend no more than 10 charts. If you include more than 10, you’ll need more than an hour for the next step, which is to write your headlines and summary.

Hour 3: Write the headlines and summary

Your third hour should be spent writing. First, write a conclusion headline for each chart:

  • Conclusion headlines should inform and enlighten. Never write a headline like, “There was a big spike in coverage in June.” Instead, say, “Because we (or the media) did x, y, or z, media coverage increased by xx% in June.”
  • Explain in one short sentence or phrase what conclusions readers should draw from the pictures or charts.
  • The titles on the individual charts should describe the data they represent.

Once you’re done with the conclusion headlines, write your executive summary and methodology.

  • The executive summary should capture the 5-7 points you want your audience to remember.
  • The methodology should be as detailed as necessary and goes at the end of your report.

Hour 4: Proof, edit, revise and rehearse

Clean up any graphical inconsistencies and proof your report thoroughly. Grammar counts. No typos, no misspellings.

Now, sit down and read the report aloud to someone. Give it to someone else to read aloud to you:

  • Does it sound smooth and reasonably eloquent? Is it readily understandable? If not, refine as necessary.
  • If you were presenting this in the board room, would everyone understand all of it? If not, refine as necessary.

Congrats; you’ve written a solid and effective report in four hours. Take a well-earned break.

A version of this article originally appeared on The Measurement Advisor.

COMMENT Daily Headlines

Sign up to receive the latest articles from directly in your inbox.