The brainstorming

What story do I want to tell?

This is the most important aspect which you should define before entering the (virtual) stage: What do you want to achieve with your talk? This can change during your preparation, for sure, but it is the most important thing to always have in the back of your head to align everything you are doing:

Should this one aspect be in the summary? In which order do I have to explain basic concepts? Know the story you want to tell and what you want your audience to take away.

Let me do a short example: When I’m talking about a new technology, I want the audience to be as excited about it as I am, and I want them to be able to immediately try it out after the talk. To achieve that, I tell the story how I got excited about it by telling how this technology solved a problem for me. Even if the people in your audience don’t have the same problem, they can put themselves in your shoes and appreciate how your pain was relieved. In the end, I give them an example or a pointer how to proceed on their own.

If you got your motivation and story straight, you should be clear about the circumstances and setting: What is your time budget? What (basic) knowledge can you expect the audience to have? Have your story and these answered questions always present. It helps with many decisions during your preparation process.

After we are done with the more explorative brainstorming phase, let’s dive into how to form the actual talk.

Creating an outline

I have been a heavy user of mind maps for many years, and use them for the early planning phases of all my projects. They allow to quickly and iteratively develop an outline.

The tool…

As a tool, I use the open-source application Freeplane. It works on macOS and has all the important features you need (and many more… Please try to ignore the slightly clunky Java Swing user interface. The devs are doing their best to make it look nice!). The most important features of dedicated outlining applications are folding (temporarily hiding child nodes) and fast reordering of subtrees. Get familiar with your tool of choice. The most mind mapping and out liner tools work similarly, and most of them even use the same keyboard shortcuts[1].

… and how to use it.

If you are comfortable with your tool of choice, you can start with the main work of the outline phase. For talks ranging from 30 to 45 minutes, I suggest going with a maximum of 5 main parts. You can start with this generic schema and then align the headlines to your current topic:

Use your outline to shift topics around and find headlines. Then go one level deeper and begin to flesh out what key facts you want to transport to your audience for every headline. I again try to limit myself to 3–5 topics for the sub-parts. You will understand why when we come to prioritization and time budget later.

I would suggest not to go into more detail at this stage by adding a third level to your outline. But if you have thoughts that you don’t want to lose, just add them and fold the parent node so that everything still fits onto your screen (Use the Space bar).

Example mind map for outline development
Example mind map for outline development

When you are structuring the individual parts of a talk, proceed with a similar pattern as with the talk as a whole. Think of it like a recursive function:

In the intro, motivate the audience by explaining why this part is important. Mention later parts where the knowledge from this part will be relevant. In that, later parts also do the back reference to the preview as the icing on the cake.

What you tell in the beginning and in the end will be remembered. This is also true for the individual parts of the talk.

After deciding how to introduce the topic of the current part, go into the actual topic and decide which aspects you want to talk about to explain it. Add them as child nodes to your outline as a single term or (very) short sentence, and then proceed. As with the whole talk, try to limit yourself to 3–5 aspects.

For the summary, explain and present the most important aspect(s) of the current part of the whole story.

When you are reaching the end of the talk, a dedicated summary should contain some additional things compared to the short wrap-ups between the individual parts: asking for feedback and a call to action.

Most topics cannot be explained 100% in 30 or 45 minutes. Tell your audience that you know that, and that you intentionally and probably also unintentionally skipped some aspects. This can lead to interesting conversations and valuable feedback afterward. Explicitly ask for experiences and feedback.

A talk should not replace in-depth studies, but give an impulse to look into something. That leads us to the „call to action“.

If the people leave your talk, they should know how to proceed if you got them interested. Give pointers, examples, links on how to continue and getting started.

With this, you should be able to get to your first outline draft. Always keep the three steps in mind for everything you do:

You get the point. Remember the recursive function analogy.

How do we continue now? We have to find out how we actually want to explain the topics that we decided are important for our story.

The research phase

Now is the time to fill knowledge gaps and decide on the details. If you look at your outline and are certain that you have enough knowledge and confidence to discuss every part for an hour, you could skip to the next section already. I strongly suggest not to do this. You will need much more time during the „practice & prioritize“ phase then.

During research, you can be liberal with your outline and add all the info you want to talk about to the sections. Don’t limit yourself at this point. If you fall into a rabbit hole for a topic, collect everything you think is important. The outline (or, better, mind map) will visually indicate importance via nesting depth and amount of (sub-)sections. Don’t worry, we will clean that mess up in the next and last phase.

Your outline will look more like the next example now. You are free to go down as deep as you like to feel confident. In the beginning, you will probably collect far too much information, but you will get a hang of it while practicing. I’m still on my quest to finding the „right“ nesting depth at this stage. If you are uncertain if something is important, you should always think: Does it contribute something valuable to my story?

Example outline after research phase
Example outline after research phase

Practice & Prioritize 🔃

Now is the time to be agile™️ and iterate. During this phase, you will fit your talk to your time budget. To spend your time – and more important: the time of your audience – wisely, you have to practice and prioritize.

Time budget

Go through your outline level by level and write down the time budget (in minutes) that you want to dedicate to the parts and the (sub-)sections. I use graphical icons for that. It is a good practice to go all the way down to the last level of your outline for this. It will become obvious now if you went too deep into some rabbit hole(s) and need to prioritize. You can start to shift things around or remove details right now, or better wait for your first practice run and do it afterward. Do wait, especially if you don’t have much experience how fast you are speaking and how precisely you can formulate.

You will have a clear picture of how you plan to spend the time now. The following example is for a 30-minute talk. See how I went all the way down to one minute slots:

Outline with time budget in minutes as icons
Outline with time budget in minutes as icons

Are we ready to hit the stage now? Soon, I promise! But first, practice! We have to validate if our time budget fits our plan.


Ideally, you do the talk once in its entirety and record yourself[2]. It sometimes hurts to see yourself talking on video, but you can learn so much from it. Do it!

You can practice individual parts isolated to tweak them from a content and time budget perspective after that later to save some time. I usually identify some parts where I have difficulties finding the right words and then iterate on these a few times before doing a rehearsal of the whole talk again.

This is important: Don’t try to hit the time budget at your first try! Talk about everything you planned. It is totally fine if you need three times the budget you planned. There will be many options for improvement to finally hit your target in the end.


The obvious things first: If you have difficulties doing a transition between two parts: Is the order right, or should some parts be switched? Remember: Important things should go into the beginning and the end. If a part is too long: Do all aspects add something valuable to your story, and are they really needed for your desired result? No? Remove. If you constantly can’t fit an aspect in your time budget: it seems very important and should get more time or visual support!


Maybe you are wondering that you are already practicing your talk before we even talked about slides[3]. That was completely intentional. You get a feeling for what is essential and where your explanation would really benefit from support on a visual level. If you are struggling to verbalize something: Use a diagram. If it takes very long to describe some facts: use a diagram.

Mark things in your outline (With a check mark icon, for instance) when you put some information onto a slide or into your moderator notes. This way, you make sure that everything you collected during your research is part of your story.

The last thing I recommend doing early in your rehearsal process: Plainly transfer your outline to a slide deck. Just copy the headlines to individual slides, and maybe copy your detailed notes into the moderator notes. Use this as a starting point and then …

Iterate 🔃

You probably remember that I talked about „iteration“ at the beginning of this section, and this is the place where it starts. When you have your time budget spent, did your first practice(s), changed priorities, already removed some details and have your first slides[3] start over by looking at your outline again. Do you think the time budget is still valid? Then start another practice run and so on. There is no right or wrong here. Tweak your slides, fine-tune your wording and change order until you can tell your story in time (budget).

Closing words

Chances are good, that there are people in the audience that are more experienced at giving talks and have more knowledge about the topic. That is totally fine and actually a good thing. Don’t be afraid of it. Most people are much humbler, constructive and forgiving face-to-face than on the internet. If someone decides to criticize a missing aspect: be humble and thankful. That’s actually free coaching and from my experience this barely happens in front of the whole audience but in a personal conversation afterward.

I hope this rough pattern helps to improve or start your journey of transferring knowledge. Tell a story to motivate, summarize in between, and leave the audience with something to do or think about in the end. And most importantly: You don’t need to know everything to give valuable talks. Talks should spark interest, give impulses and start conversations.

Thanks to Kaja, Jan and Joachim for making this post much better!

  1. The most important features you need to know for an outline phase with Freeplane: Add a new node: Just hit Return / Enter to create a new node on the same level and start typing to add text | Add new child / sub node: use Tab | Quickly navigate between the nodes using the arrow keys | Move nodes between levels & reorder by holding Command (CTRL on Windows) and using the arrow keys. This feature is your best friend. Use it heavily for reordering while you think about your structure and prioritize! | If your map grows, use the Space bar to fold / hide child–nodes you currently don't need.  ↩

  2. On macOS you can use the QuickTime Player to do a screen recording with your microphone enabled and also have an overlay with your webcam. Click on „File“ → „New movie recording“ first! This opens a window with your webcam. Leave that window open, but don't start that recording. Click „File“ → „New screen recording“ instead and select your microphone via the „Options“ menu before starting the screen recording. If doing a recording is not an option, you can always rely on a (stop) watch and simple notes (remove this, shorten that, …). How do you do this on Windows / Linux? Please reach out so that I can add that info.  ↩

  3. Slides are just a placeholder for „visual support“. You could also use a mind map, or show code in your editor.  ↩