Q&A on Writing for Execs
Last week, I had the opportunity to run a fun workshop on writing for executives at TrustCon (hello to new trust and safety subscribers!) I got great questions and comments, which reflected similar ones I hear whenever I speak about this stuff. So I thought I’d share some common Q&A here. Note that while this recent workshop was for online trust and safety professionals, this Q&A reflects general questions I get and is not specific to any industry.
For context, the workshop was I based on tips I distilled from my time writing for and editing the President’s Daily Brief. You can read the basics here:
Ok here goes, in no particular order…
Q: How Detailed Should a Bottom Line Be?
A: Just detailed enough to convey what’s happening and why it matters.
Perhaps the most important tip I give is to have a “bottom line up front” (or “BLUF”)—meaning, have a title and topic sentence with all the information a busy reader might need. But how much information do you need, and can you realistically cram, into this small space?
To answer this, I go back to intelligence analysis basics, which is that your bottom line should have a “what” and a “so what,” meaning you’re telling the reader the basics of a new development and ALSO why it matters. And therefore, you need enough detail to make the “what” and “so what” clear, and no more.
For example, say you’re telling executives about a 0.05 percentage point increase in the past month in a metric your company monitors, like number of customer complaints. Your BLUF could be very detailed, like “0.05 Percentage Point Increase in Complaints Last Month.” But as an exec I might read this and have the following questions:
Is this a large increase or relatively small in the scheme of things?
Is this a trend or just a spike in the past month?
Do we know anything about why there was an increase?
Are we doing anything to address it?
Do I have any decisions to make?
In general, why is this information worth my time?
Therefore for a situation like this, I’d err on providing less detail in the BLUF about what’s happening and more about why it matters, depending on why that is.
One more thing to keep in mind is how much context your audience already has. Are you regularly providing updates on complaint trends, and do execs already know some of the answers to the questions above? If so, maybe you don’t need to be as explicit about the context and can get into more detail. This just goes to show that there’s no one approach for every situation.
Q: Does avoiding “value-laden” language mean having no values?
A: No!
Another tip I give is to avoid “value-laden” language that conveys a subjective judgment. But this is NOT the same as saying one should not have values, for yourself or your organization.
For example, here’s an example of what to do and not do to avoid value-laden language:
GOOD: “The increase in customer complaints seems to reflect lack of understanding of our product features.”
NOT GOOD: “We have an increase in customers making stupid complaints because they’re misunderstanding our awesome product features.”
You can see with the better option, you’re setting a sober discussion about customer response to your product and options. Whereas with the second, you’ve set a combative tone and perhaps cut off some useful discussion.
This is completely different than having values and referencing them in your writing. For example, in this case, perhaps you or your company values accessibility for customers with disabilities, some of these features are for them, and you think the increase in complaints is an acceptable trade-off. You could say:
“Although we’ve seen an increase in customer confusion with our new features, we recommend keeping them to provide accessibility for customers with disabilities.”
There is no “value-laden” language in that sentence, but you are still advancing your values. (Also I’m sure I’m setting up a false trade off between customer confusion and accessibility features but just bear with me for this example.)
Q: Don’t I need value-laden language to convince people I’m right?
A: No!
I’ve heard this question from people in government, the private sector, NGOs, everywhere. People sometimes find themselves in a situation where they think their organization is not doing what they think it should, on whatever issue. In this situation, it’s temping to try and shock executives into seeing things your way by using over-the-top language.
While I understand and sympathize with this view, the reality is that this approach backfires. Doing this will undermine your credibility because you’ll be seen as only caring about your issue, not the big picture. And more than that, it’s likely to actually skew your analysis and recommendation—because you won’t yourself be thinking about tradeoffs clearly, but rather focused on your own view. While it can be frustrating to feel like your issue is not going the way you want, it’s more effective to present the issue in neutral language and tee up the practical or principled implications for your organization.
Q: Isn’t Some Probabilistic Language Clear Enough, like “Probably” or “Very Likely”?
A: Yes, with caveats.
One tip I’ve stressed is to avoid probabilistic language to convey uncertainty if you can, because these terms can be interpreted in different ways. But I want to clarify you can’t and generally shouldn’t avoid these completely. Sometimes you just need to convey how likely you think something is.
When you do this, just make sure to keep the language simple and clear to all. For example, what is the difference between “likely” and “very likely”? I don’t know, and your audience might not either.
A caveat here is your organization may have a common terminology you can reference, so you know people always understand these terms in the same way. If this is the case, you can use more specific terms with confidence that people will understand them. The intelligence community does this sometimes to decrease confusion.
The caveat to the caveat is, I just advise making sure the terms don’t get so precise that they convey certainty you don’t have. For example, if you say “likely” means 51-75% chance, and “very likely” means 76-90% chance, how are you calculating the odds? If you’re not doing it with statistics based on data, you might be over-interpreting your own assessment.
Check out more on this here, if interested:
Ok I’m going to stop here. Are there other questions you’d like more on? Let me know in the comments!