I am frequently asked to review engineering reports and documents. And I am continually baffled with how many engineers want to 'take their reader on a journey' as opposed to getting to the main point. These engineers write documents where you have to read every single word to work out what they are trying to say. But having a final reveal is never a good thing in any technical document.
People watch movies to be taken on journeys.
Managers read reports to get information.
This is not to say that you shouldn't try and make engineering reports as 'enjoyable' as possible. You should. But you need to understand the audience - which is not a typical movie-goer. And while you may be proud of your research and analysis, no one wants to read your four month diary on how you got there.
Getting reports and documents right is important. Managers will quickly work out which engineers give them the best information in the easiest way. And these engineers are the ones that managers remember (when it comes to promotions, special assignments, raises et cetera).
So here are some tips and guidance to help you to write better reports. Trust me … this will eventually pay off for you.
1. What does your reader need to hear?
Simple question … right? But I am constantly amazed by how often this crucial question is seemingly forgotten by engineers.
A report is written to give information about a decision that needs to be made. If this is not the case ... the report is a waste of time. A manufacturing company may (for example) need to decide whether to extend the warranty period of a consumer product (or not). So lets break this down into what information the manager making the decision needs to know:
That is pretty much it. Writing this out helps make it clear in your mind as the author. But it is still too 'high-level.' What information feeds into this? We need to break this information requirement down a little further:
Now we are getting somewhere. These are the key factors the meet our 'high-level' information need. But we need a little more:
How many additional units will sell based on an increased warranty period?
How much revenue is made on each sale?
What is the current warranty reliability of the product?
What is the proposed warranty reliability of the product?
How much does each warranty action cost?
Now you are cooking. You could go down further if you wish, but you see what I mean. By doing this, you know the hierarchy of the information that needs to be in your report. You will need this later.
This is the first thing that makes a good engineering report a 'bad movie.' You don't go to a movie to help you make a decision. Movie-goers are not looking for information that will help guide their lives. The movie audience's focus is the movie. But ... the focus of the manger who reads your engineering report is a decision he or she needs to make. Not the report. Remember this.
2. What do you want to say?
Haven't I just worked out what I want to say in the previous step?
No. Everyone wants to say something regardless of who is listening. It is human nature. This is where you get introspective. Because you need to work out what you want to say … and keep it out of your report if it isn't what the reader needs to hear.
For example you may want to:
explain how long and difficult it was to undertake the analysis
detail the precise method you used down to the extent you include every single formula in the body of your text.
include a chart showing a deterministic relationship between relative humidity and corrosion fatigue crack propagation.
provide a historical review of all products released to date (to ensure the reader knows you did a thorough job)
But I can hear many of you saying that we need to have things like formulae and historical reviews in reports for due diligence. And you might be right.
But put these things as annexes or appendices.
Don't make your reader have to read all of this stuff to get your point.
As with everything in life, there is a balance to be struck. But knowing your own tendencies is very helpful for making sure you don't make the report 'all about you.'
Don't be the Hollywood director who has a story to tell.
3. Get your 'BLUF'
BLUF stands for:
bottom line up front.
It means that the reader knows exactly what you are recommending after the read the first page … or ideally the first paragraph.
It took until the final scenes of 'The Sixth Sense' for the audience to realize that Bruce Willis had actually died and was a ghost (apologies for those yet to see the movie). This makes a great movie. But leaving everything to the last paragraph in an engineering report won't help a manager who is pressed for time and needs to constantly make quick decisions.
Everyone is busy. And while your report may need to be 'long' to contain all the information you identified in the first step, the manager who is reading it may not have that time.
The BLUF needs to address the high-level information requirement.
In the case of deciding to extend the warranty period for our consumer product, this needs to answer the question 'How much money will this decision make (or lose) for my organization?'
If this is not clear in your first paragraph … make it clear.
4. Executive summaries summarize what you are trying to communicate … not how the document is structured
Too many executive summaries have come across my desk explaining the sections of the document it covers. These executive summaries say (for example) that the engineering report includes a literature review, followed by data analysis, and then recommendations. But it does not include what the report is trying to say!!
The executive summary should be complete enough for the manager to be able to make the decision. If it can't … you need to redraft it. Hollywood trailers may leave you wondering if the four teenagers will survive their weekend alone at Murder Lake. Executive summaries should not. (The teenagers all died horrible deaths by the way).
5. Less is more
When it comes to writing a technical document (… and lets face it, they are not 'fun' reads) don't use lots of or big words or long sentences. The simpler the better. For example:
It can be concluded based on analysis of historical and actual test data that the median estimate of the increase in profit based on extending the warranty by 1 year will be 15 per cent.
This is a horrible sentence. There is no need to say things like 'based on analysis.' Of course anything you say is based on analysis. Instead, try:
Extending the warranty by 1 year will likely increase profit by 15 per cent.
Hollywood movies tend to be 90 to 120 minutes. There is no such guidance for an engineering report. If you can say all that needs to be said in 1 page … do it!
6. If you can take it out without it affecting your conclusion … take it out
Simple rule here and it follows from the previous point. If your manager knows the consumer product well, you don't need to include a detailed explanation of its design. Even if your manager doesn't know the product well … will explaining to them the different modules help your argument about extending a warranty period? It might make your report feel more 'complete.' But as Obi-Wan Kenobi says to Luke Skywalker:
Bury your feelings deep down.
Including material that makes you feel better dilutes your key message.
Engineering reports are not textbooks. You are not trying to educate the reader (in many cases). You are trying to help them make the right decision.
Where Hollywood scriptwriters can elicit audience responses based on dramatic pauses and so on … this doesn't help your report. Remove the dialogue!
7. If you don't have all the information … say what needs to happen to get it
This is where step 1 is important. A reliability engineer may not understand how warranty periods impact sales. We know that products with longer warranty periods are more attractive to customers. But how many additional sales will this translate to? This is where the marketing department needs to help. But lets say that you are a reliability engineer tasked with writing this report, but can't speak to the marketing team in time.
You need to make it easy for the manager.
For example, you might be able to work out:
Increasing the warranty by 1 year will likely need an 8.8 per cent increase in sales to be cost neutral.
Brilliant! This means the manager knows that if his marketing team estimates a 12 per cent increase in sales … he or she should extend the warranty period. Conversely, if the estimate comes back at 4 per cent, then the warranty period should stay the same.
You could also include a basic chart that plots potential profit versus increase in sales, which would then communicate all the information you have. The manager can not have his or her task made simpler for them.
While many Hollywood franchises are trying to create the need for sequels … your engineering report should not. If you don't have all the information, then make the 'sequel' short and clear by focusing on what you need to know … and now how to interpret it.
8. Don't forget risk
As a rule, movies need to end with a definitive resolution on something. We need to know if the romantic leads could work through their issues and live happily after. Arnold Schwarzenegger has to always beat the bad guy. And there needs to be no doubt about this.
This is not always possible in the real world. Reliability engineers in particular have to deal with uncertainties. We will never know the true reliability of a consumer product because we don't have a lot of data. We don't throw our hands up in the air and walk off set.
So we need to communicate it. We may need to say things like:
Increasing the warranty by 1 year will likely need an 8.8 per cent increase in sales to be cost neutral.
A 9.1 per cent increase in sales is needed to be 95 per cent confident of cost neutrality.
So now we have uncertainty covered as well. It is simply not good enough to include your 'best guess.' The decision maker needs to understand what chance there is of losing money.
9. But Hollywood does get it right when it comes to drafts
Well not all the time. Movies that have had their screenplays reviewed and redrafted are typically better for it (unless the reviewer is the lead actor who is trying to increase their screen time … see Tom Cruise in 'The Mummy').
So get your report finished one week before it is due. Then don't think about it for a couple of days. I will guarantee you that when you look at it again, there will be parts you can't believe you wrote. Go through all the considerations above. And once you have done this, get someone else to read it. Someone who is not an expert on the topic.
I know that in my first drafts, my sentences tend to be too long. It is important for me to write down the gist of what I am trying to say, so I don't worry about sentence structure straight away. I don't want a focus on sentence length to interrupt my train of thought.
Sentence structure comes later for me. And I continually astound myself at how stupid I sound in my first draft of anything (my wife doesn't get astounded by my stupidity anymore … but that is for another article/counselling session).
10. And last but not least … practice
Gary Player is a former golfer who once said:
The harder you practice, the luckier you get.
Player did not come up with the quote - it has existed in one form or another since at least 1900.
Don't shy away from practicing. Don't shy away from having your work reviewed by someone who is better than you at writing. And don't be devastated by constructive criticism.
I get it all the time. Mostly from my wife. And I am getting better at not being devastated by it.