For the first year of performing my stage show, I operated on a simple feedback model. I would perform the show, notice what did not work, fix those things, and perform again. It was binary: something either worked or it did not. If it worked, I kept it. If it did not, I changed it.
This approach produced improvement, but it was crude. It caught the obvious problems — the joke that bombed, the transition that created dead time, the routine that consistently fell flat. What it missed was everything else. The subtle dynamics. The small opportunities. The moments where something was working well enough that I never thought to examine whether it could work better.
The refinement cycle that John Graham describes in Stage By Stage is far more sophisticated than my binary approach, and adopting it transformed how I develop my show. The cycle has five elements: Perform, Think, Observe, Review, Repeat. Each element feeds the others in a continuous loop, and together they produce a kind of compound improvement that my simple fix-what-is-broken model could never achieve.
Perform
The first element is obvious but not trivial. You have to perform. Regularly. In front of real audiences. There is no substitute for this and no shortcut around it.
But performing within the refinement cycle is different from performing in isolation. When you are running the full cycle, each performance is not just a show — it is a data-gathering operation. You are performing and simultaneously collecting information that will feed the other four elements of the cycle.
This does not mean you perform with a clinical detachment. The show has to be genuine. The audience has to get your full energy and attention. But in the background of your mind, a part of you is taking notes. That line got a bigger laugh than usual tonight — why? That transition felt smoother — what did I do differently? The volunteer in the third routine seemed uncomfortable — was it something about how I positioned them, or was it the specific person?
This background awareness develops naturally with flight time. In the early stages, you are too consumed by execution to notice anything else. But as the show becomes more automatic, your capacity for simultaneous awareness expands. You can deliver the show and observe it at the same time. Not perfectly, not comprehensively, but enough to generate useful data.
Think
The second element is the one that surprised me most. Graham writes that you should be thinking about your show before, during, and after performances. The thinking phase is not limited to the post-show debrief. It is ongoing, and some of the best ideas come at unexpected moments.
I have experienced this repeatedly. A perfect line to cover an awkward transition pops into my head while walking through an airport. A new callback idea surfaces during a long train ride to a consulting engagement. A structural solution to a pacing problem crystallizes while I am in the shower after a morning run.
The thinking element of the cycle works because your subconscious mind processes performing experiences even when you are not consciously working on the show. When you perform regularly and pay attention during performances, you feed your subconscious a steady stream of raw material. The thinking that happens in off-hours is your subconscious processing that material and surfacing connections, solutions, and ideas that your conscious mind would not have reached through deliberate analysis.
The practical requirement is simple: capture the thoughts when they come. I use the notes app on my phone. Every idea related to the show — no matter how small, no matter how half-formed — gets typed in immediately. A word change. A timing adjustment. A new piece of music to try. An observation about how a particular type of audience responds to a particular moment. Some of these notes lead to changes. Most do not. But the act of capturing them keeps the thinking process active and gives me a library of ideas to draw from when I sit down to work on the show deliberately.
Observe
The third element is about watching and listening to the audience with deliberate attention. Not just monitoring whether they are engaged, but studying how they engage. The quality of their reactions. The specific moments where their attention peaks and dips. The things they say during and after the show.
Graham makes a point that reshaped how I process audience feedback. He says that polite applause may mean you need more animation, or that the audience was not fooled. This distinction is critical. In my early performing days, I treated all applause as positive feedback. Applause meant the trick worked. What I did not distinguish was the difference between polite applause — the social obligation to clap when something is clearly finished — and genuine applause that comes from being truly moved or astonished.
Learning to read the quality of audience reactions is a skill that develops through observation over many performances. You start to notice that a particular moment consistently gets applause that is prompt and perfunctory — the audience is acknowledging the trick, not celebrating it. You notice that another moment consistently produces a beat of silence before the applause, which is the audience processing something unexpected. You notice that a joke gets laughter from the front rows but not the back, which might be a volume issue or might be a sightline problem that prevents people in the back from seeing the visual element that makes the joke land.
The observation element also includes listening to what audience members say after the show. Graham notes that an unexpected audience comment that gets a laugh can become your own line. I have experienced this firsthand. Some of the best lines in my show originated not from my scriptwriting but from something a spectator said during a performance that was so perfect I incorporated it. The audience, when you observe them carefully, will write your show for you.
Review
The fourth element is the one I resisted longest: recording and reviewing your performances on video.
Graham is unequivocal about this. He writes that you will learn things from video review that you would never have learned, even after hundreds of shows. He also warns that reviewing footage will make you cringe, and that this is an excellent sign.
He is right on both counts.
The first time I watched a video of my stage show, I was genuinely taken aback by the gap between what I thought I was doing and what I was actually doing. Gestures I thought were natural looked stiff. Pauses I thought were dramatic were barely pauses at all — I was rushing through them. Eye contact I thought was consistent was sporadic and unfocused. My posture during transitions was different from my posture during routines, which created a visible shift between “performing” and “not performing” that the audience could see even if they could not articulate it.
The video also revealed things I was doing well that I was not aware of. A particular physical gesture during a reveal was more effective than I realized. A casual aside that I barely noticed in the moment was getting a stronger reaction than my scripted jokes. My energy in the opening sixty seconds was higher and more engaging than at any other point in the show, which suggested that I was losing energy as the show progressed rather than building it.
None of this was visible from the inside. All of it was visible on video.
The review process I developed is straightforward. I record the show from a fixed camera angle that captures both me and the audience. After the show, I watch the entire recording once through without stopping, just to get a general impression. Then I watch it again with a notepad, stopping at any moment that catches my attention — positive or negative. I note the timestamp and what I observed. Then I compare my notes from the video review with my notes from the live performance. The discrepancies between what I noticed during the show and what the video reveals are the most valuable data points.
I do not review every performance. That would be unsustainable. But I review regularly enough to catch trends and patterns, and I always review after a show where something felt different — either better or worse than usual.
Repeat
The fifth element is the one that ties the cycle together. You take what you have learned from thinking, observing, and reviewing, and you apply it to the next performance. Then the cycle begins again.
The power of the cycle is not in any single iteration. It is in the compounding effect of running the cycle continuously over dozens, then hundreds, of performances. Each iteration produces small improvements. A word change here. A timing adjustment there. A physical repositioning. A different music cue. Individually, these changes are minor. Collectively, over time, they transform the show.
Graham describes this compounding effect when he writes about finding the flow. All moments should blend into one seamless whole, he says. Transitions should not feel like “here is a trick, okay that one is done.” Cut all hesitation from your speech. “There is no time to vacillate, hesitate, or complicate.”
This seamlessness is not something you can design on paper. It emerges from the refinement cycle. Each iteration removes a tiny friction point, smooths a tiny rough edge, tightens a tiny moment of slack. After enough iterations, the show flows in a way that feels effortless and inevitable — not because it was designed that way from the start but because it has been refined toward that quality through hundreds of small adjustments.
The Compound Effect
I want to be specific about the compound effect because it is the most important aspect of the cycle and the least intuitive.
In my first twenty performances of the show, the cycle produced large, obvious improvements. Structural changes. Routine swaps. Major script rewrites. The show looked noticeably different from performance to performance because the problems were large and the fixes were substantial.
Between performances twenty and fifty, the improvements got smaller. Word-level script changes. Timing adjustments measured in fractions of a second. Repositioning of physical actions by a few centimeters. Each individual change was barely noticeable.
Between performances fifty and a hundred, the improvements got smaller still. But the cumulative effect was dramatic. When I compared a video of performance ninety-five with a video of performance fifteen, the difference was striking. The show was tighter, smoother, funnier, and more engaging in ways that could not be attributed to any single change. It was the compound effect of eighty iterations of the refinement cycle, each one producing an increment of improvement so small that it was invisible in isolation but transformative in aggregate.
This is why Graham writes that the more you know your show, the more you can focus on everything else during your show. The cycle frees you. By continuously refining the mechanical and structural elements of the show through the cycle, you gradually reduce the cognitive load of performing, which frees your attention for the things that matter most — the audience, the energy, the relationship, the moment.
Running the Cycle Honestly
There is one prerequisite for the cycle to work that I want to address directly, because it is the element I struggled with most: honesty.
The cycle only produces real improvement if you are honest about what you observe. This means acknowledging when a piece you love is not working. It means accepting that a joke you are proud of is not landing. It means recognizing that a moment you thought was your strongest might actually be your weakest, because the audience’s polite applause was masking their disengagement.
Self-deception is the enemy of the refinement cycle. It is remarkably easy to watch a video of your performance and see what you want to see rather than what is actually there. It is easy to interpret polite laughter as genuine enjoyment. It is easy to blame a flat audience reaction on the audience rather than on your material or delivery.
The discipline of the cycle requires a commitment to seeing clearly, even when what you see is uncomfortable. The cringe that Graham describes when you first review your footage is not a one-time event. It recurs every time you reach a new level of awareness, because each new level reveals imperfections that were invisible at the previous level.
This is why I view the refinement cycle not just as a methodology for improving a show but as a practice in honest self-assessment. Every iteration of the cycle is an opportunity to see yourself more clearly — as a performer, as a communicator, as a person standing in front of other people and asking for their attention. The willingness to look honestly at what you see is what separates performers who continuously improve from performers who plateau.
Perform. Think. Observe. Review. Repeat. The cycle is simple. Running it honestly is the hard part. And the results, over time, are everything.