Eric Bogatin, Signal Integrity Journal Technical Editor
Eric Bogatin, Signal Integrity Journal Technical Editor RSS FeedRSS

Eric Bogatin_new headshot_100

Eric Bogatin is Technical Editor at Signal Integrity Journal and the Dean of the Teledyne LeCroy Signal Integrity Academy. Additionally, he is an Adjunct Professor at the University of Colorado - Boulder in the ECEE Dept. Eric improves the signal to noise ratio by sorting through all of the information available and finding the best quality content to publish on

Signal Integrity

Eight Tips on Giving a Presentation

November 16, 2021

Now that I am teaching full time and have a number of undergraduate and graduate students doing weekly project reviews, I listen to more presentations than I give. I find myself giving the same feedback over and over again. And I’ve noticed, when I review or edit papers for the SI Journal, I am providing the same feedback. So, I thought I would share my eight most common comments and tips to engineers giving presentations.

Tip #1. Every slide should have a “so what?” What is the message you want your audience to walk away with? The entire paper should have a “so what?” Then, break it down to smaller sections that support the main “so what?” and eventually down to each slide. This means you must identify the message in your head, first. Always ask yourself, what is the message I want this presentation to scream to the audience? Each slide should support this message and have its own “so what?” It is really ok to explicitly state the message on the slide. “The simulation matches the measurement up to 10 GHz, but it is not so good above this.” Don’t add more than one message per slide.

Tip #2. Always do an analysis of your measurement or simulation. Never show a figure, a measurement or a simulation, just to check off a box and say here is my result. If you are showing it, tell me what I am supposed to be looking for. Is this good or bad? If it does not add to your “so what” message, don’t show it. If you are saying the fixture that holds the cable in place is important, then show me the fixture and the cable. Don’t just show me the cable unless it clearly shows the feature that is important.

Tip #3: Don’t tell me the backstory, tell me the results. All projects are worked on in a linear flow. You do a series of experiments, some of which are dead ends, some of which are more important than others. This is all the backstory, the core dump of everything you did. But, unless your story is about your journey and all the dead ends you followed, I don’t want to hear about them in a short presentation. Focus on the message you want me to walk away with. Standing around eating pizza at parties is the time to tell the story of the five things you tried that did not work until you found the one thing that did.

Tip #4. Don’t make me work too hard. The only reason you show a plot or a figure is to make a specific point. If your graph does not speak for itself, and shout out your message, change the plot. Never use auto scale. Use a set of scales that are appropriate for your message. If you are looking at a large dynamic range, use a log scale. If you are showing an impedance, plot the impedance and frequency on log-log scales. Plot two items you want to compare on the same graph so I don’t have to go from graph to graph to compare. Use scales that are nice whole number or multiples of 1 or 2 or 5 or 10 (scope scales) so I can do estimates from your graph at a glance. Make it easy for your audience to see your message.

Tip #5: Apply rule #9 to each measurement or simulation example [1]. If you show me a result, don’t just describe what I can see on the graph, tell me why you expected it be that way. If it is not what you expected, then tell me why. Is your intuition off, is the simulation set up incorrectly, is the channel on the scope the wrong channel? Never show a result that you have not anticipated, and especially that is inconsistent with what you would expect, unless you have attempted to understand why it violates rule #9.

Tip #6: Perform every consistency test you can think of. Most of the time, I do not understand all the details of the experiment or project being presented. The presenter should know the project better than I do. However, I can do consistency tests. If your VCO has a sensitivity of 1 MHz/V and you measured 4 MHz, but your voltage rail to your DAC is only 3.3 V, this is not consistent. You need to explore why not. If I find one inconsistency, it makes we wonder if there are more, and this means you do not understand your system very well. What else might be wrong?

Tip #7: If your slide “begs the question,” answer the question. If in your presentation, you show a result where there is a simple to perform consistency test and you do not do it, I will wonder, why didn’t you do that test? If it is an obvious question, anticipate it and either do the test or tell me why you didn’t do it. If your measured impedance suggests a substrate thickness of 64 mils then either measure it and show me the results or tell me why you did not measure it. If is an obvious question I will ask- answer it before I have to ask it. The best presentations are the ones where there is an obvious question that would be asked, someone actually asks the question and the next slide has the answer.

Tip #8: Apply situational awareness: No instrument or simulator is perfect, or “ideal.” They all have limitations. These limitations create measurement or simulation artifacts. Be aware of them. Every scope has a bandwidth, including the probes. It has a sample rate with a shortest time resolution. When you perform a measurement, recognize how close to the limits of your measurement your instrument was. Likewise, every simulator has some limitations. If you are simulating a lossy interconnect, does your simulator have a causal model of the dielectric, or include the skin depth resistance of the return plane? Be aware of the limitations of your tools and the problem you are addressing and convince me your result does not have an artifact. This is situational awareness.

[1] Bogatin's Rule #9 is “Never do a measurement or simulation without first anticipating the results. If you see something unexpected, don't proceed with the result until you can understand why it doesn't match your expectation.


You must login or register in order to post a comment.