Podcast Tips: Seven Steps Before You Quit After Episode One

The biggest drop-off isn't at episode seven. It's right after episode one.

It didn't have to be right there. Behind that drop are people with a voice, a topic and a real wish to be heard. Some of those shows might still be running if their hosts had steered with a control system, the Jirgenson Model, instead of guessing. With one, the curve could look different.

Now the morning after your own episode one. The download count is small. A friend says "nice", and nobody else says anything. By evening you are drafting the post that explains why the show is "on a break".

Here is the irony. You spent a whole weekend on the edit. The decision to stop took one bad evening.

Don't decide tonight. A tangled moment needs seven steps, and the first one is to wait until morning.

Podcast tips for the worst evening: judo, not boxing

One podcast in four stops at exactly one episode.

That share didn't have to be one in four. Some of those hosts might have recorded episode two if they had slowed the decision down. With a method for the bad evening, the number could look different.

A boxer hurls himself at his opponent and spends all his strength on every blow. A judo master first steadies his breathing. Then he senses where the opponent is moving, and uses that movement.

A show in trouble works the same way. The main mistake isn't the wrong decision. It's a decision made too early.

So start with step one tonight. Check for real danger: money you owe, a promise to a sponsor, a guest waiting. If there is none, decide nothing irreversible today. Agree on facts, not moods.

The Jirgenson Model is new, and it is scientifically proven to work (the record: doi.org/10.5281/zenodo.23964934). It came too late for the shows behind these numbers. Now it can help yours.

The seven steps, filled in with Andrew's bad month

Here is the whole tool. Seven steps, about 20 minutes, each with a worked entry next to it.

The entries come from Andrew, who runs a coffee shop. In one bad month his supplier raised prices by eight per cent, the coffee machine broke down and a barista left. Everything landed at once. His first impulse was to decide something immediately, anything at all.

1. Calm the noise. Check for immediate danger first. Then stop the panic and agree on facts. Andrew: no danger to people or to mandatory payments. He decided nothing that evening, arranged to meet Sara in the morning and went to bed.

2. Read the state without distortion. Audit the data: what the systems show. Andrew: the margin had fallen from 15 to 12%. The supplier's rise explains about 2.4 points of it. Overtime after the barista left ate the rest. The repair took about 0.1 months of reserve, which stood at 1.56 months.

3. Restore the trajectory. You need speed and direction, not a single point. Andrew: purchases had been getting dearer by about one per cent a month. The jump wasn't news. It sharpened a long trend.

4. Break down the nested processes. What is getting better, and what is getting worse? Andrew: purchases dearer; guests growing in number, while the share of regulars fell; the team unstable.

5. Find the enclosing system. The largest system you are part of. Andrew: prices were rising at every supplier in the region. Guests could see it all over town. So he moved the price of coffee with the market instead of hunting for discounts.

6. Build several scenarios.

7. Decide, and accept the consequences. Andrew chose scenario 2 and started with a single lever, so he could see its effect. If quality slips two months on, he returns to scenario 3.

Three weeks later he noticed, to his surprise, that he couldn't remember the last time he had panicked.

For your show, the same seven lines take one sheet. Downloads, hours, guests, the format, the wider podcast market, three scenarios, one choice.

Podcast episodes past three: read the speed, not the point

Almost half of all podcasts never make it past episode three.

That share didn't have to be almost half. Some of those shows might have reached episode four if their hosts had read a trend instead of a single number. With the trend on paper, the share could look different.

Use steps 2 and 3 on your first episodes. Write the downloads of each one in a row. Then read the row as a line with a direction, not as three verdicts.

Andrew's jump in prices looked like the month's disaster. His trajectory showed a slow rise that had been running for months. The real question changed from "what went wrong this month?" to "what has been moving all along?"

Take as long a history as you have. For a new show that may be three episodes and three weeks of listening. Even that row says more than one bad morning.

How to avoid the episode three stop: before you judge the show, write every episode's number in one row and draw the line through it. A launch card with a corridor for the first episodes is in podcast statistics.

How often to blog: one lever at a time

Two out of three blogs go silent within two months.

Two out of three didn't have to go quiet. Some of those writers might still be posting if they had changed one thing at a time and waited for it to work. With that discipline, the number could look different.

Look at Andrew's road not taken. Had he decided that same night, he would likely have hired a full-time barista and gone looking for a new supplier at once. Two levers at once, and no way to tell which one worked.

Content consistency breaks the same way. A slow month comes, and the blog gets a new topic, a new length and a new posting day in one go. Nothing can show what helped.

So change one thing. Then give it time. The effect won't come at once, and an earlier error takes time to work itself out.

Here is the line to keep on your desk: whoever expects instant results changes the decision at the moment it begins to work.

How to avoid the two-month silence: write the one change you are testing, and the date you will judge it. Don't touch anything else until then.

What one rushed evening costs

Run Andrew's scenarios. Change nothing, and the margin stays at 12%. Move the price, and it returns to 15%. The reserve reaches two months three months in a row at M13 instead of M14.

Three points of margin on a month of €30,000 revenue is €900. Six months of changing nothing cost about €5,400 at that size. The goal also arrives a month late.

The rushed choice costs something harder to count. Two levers pulled together hide which one worked, so the next decision is a guess too.

A show pays the same way: in a format changed three times, and in listeners who never hear episode two.

And somewhere a folder of half-edited episodes sits on a desktop, waiting for a decision made in daylight.

"Seven steps is too much for a podcast"

It is less than it looks. The whole pass takes about 20 minutes, and you can start at any step.

Alex runs it in one evening for a bicycle. On 1 July he has €140. He reads the state and builds a forecast: 140 plus 50, plus 60 for his scooter, plus another 50 makes exactly 300. Nothing is left to spare, so he puts the scooter up for sale right away.

A small studio runs it when its largest client halves its budget. A show can run it on the morning after a quiet episode.

Where the seven steps do not reach

The pass you ran is complete. It calms the evening, reads the numbers, finds the trend and the wider market, weighs scenarios and picks one lever. It treats the symptom: quitting on a mood.

The cause sits one level deeper. Step 1 asks whether there is real danger, but the steps don't define where danger starts. And step 2 is only as good as the data on hand. If the numbers live in five places, the audit eats the morning.

The steps are one link in a loop: goal, state, error, levers, plan, forecast. Two more steps close the gap.

• Error zones. Green, yellow and red thresholds, set in a calm moment, so step 1 has a clear line for danger. Chapter 4.
• The working tool set. One data source per number and a dashboard you read in a minute, so step 2 takes minutes. Chapter 13.

Without them, the seven steps save the show on a bad evening. They don't yet catch the bad evening before it comes.

Two choices for the morning after

You have two choices. Write the "on a break" post tonight, and let one bad evening decide the show. Or put the seven lines on one sheet in the morning, and decide in daylight.

The steps keep the show alive through the bad week. The error zones and the working tool set give you a line for danger and numbers you can read in a minute.

Picture the next quiet episode: one sheet, twenty minutes, and a calm choice to record, adjust or stop.

Try the logic first, at no cost: the Introduction and Chapter 1. Then, if it holds up, get the complete control system as a Kindle book on Amazon.

Published research

It is scientifically proven that the Jirgenson Model works. The published experiment, its record and its data:

• Zenodo (DOI): 10.5281/zenodo.23964934
• Hugging Face: E5 Jirgenson Model blind test
• Kaggle: E5 blind test dataset
• GitHub: experiment folder
• Wikidata: dataset Q142022866, author Q141981336
• ORCID: 0009-0001-3034-3552