What should a product launch plan include?

An end state with numbers, set months after launch day. A launch condition you can measure. A backward chain of what must be true, each row with a check date. A money check. And this week's task in the top row.

Is a launch checklist enough for a new product launch?

No. A checklist lists tasks, and a task can't warn you. Add a number to each row and a date to check it. Then a slip shows up in week 6, not on launch day.

Why does a product launch fail?

This post shows three common breaks: nobody waiting on the day, no number the launch must clear, and no plan for the months after. The book adds three places where a plan breaks before work starts: the three offsets.

Product Launch Plan: Work Backwards From the Day After Launch

Nearly six out of ten Kickstarter campaigns miss their goal. One in eleven doesn't get a single backer.

You have a date. The countdown page is live. The mock-ups look good, and the task list runs forward from Monday: design, build, film, post, launch.

Now look for one number on that list. The number that must be true the day before launch for the launch to work. Is it there?

Here is the irony. A launch gets a date to the hour and a budget to the cent. The one question that decides it gets a feeling: will anyone come?

A launch date with no launch condition is a hope with a countdown. The Jirgenson Model answers it as a control system: an end state in numbers, a plan built backwards. It is scientifically proven that the Jirgenson Model works (doi.org/10.5281/zenodo.23964934).

Kickstarter campaign or shop opening: start from the day after

The line at the top is about launches that reach the big day with nobody waiting. Fix that from the far end.

Write down what must be true after launch, not on launch day. Then walk backwards. At every step ask one question: what must already hold right before this?

A navigation app works this way. It knows where you are and where you are going. Miss a turn, and it plots a fresh route from the spot where you stand. It never asks you to guess the road.

How to avoid an empty launch day: put a row for the people who must be waiting before the launch row. Give it a number and a check date. If the number isn't there on that date, move the launch, not the goal.

In the Jirgenson Model, the backward move starts at the finished result and asks, row by row, what must already hold. It stops when it reaches this week. Planning forwards multiplies the options. Planning backwards hands you one task.

The product launch plan template, filled in

This is the tool. It takes about twenty minutes and a sheet of paper. Five steps, each with a real entry beside it.

The entries come from Andrew, who runs a coffee shop. His launch was a second location. He planned it from the end.

1. Write the end state, after launch day

Not "open the second shop". Write the full state, with numbers.

Andrew's entry, for month 30, counted from the night he first wrote his goals:

Field	Andrew's entry
Launch	Second location open at M30
First shop	Margin of at least 14%, including a manager's cost
Money	Reserve of at least two months of expenses, plus a source of extra reserve
Owner	No more than 45 hours a week, only from M24 to M30
After launch	The new shop reaches a 15% margin within six months of opening

The last row matters most. The plan does not end at the ribbon.

2. Tie the launch to a condition, not a date

Write the event that must happen before you start. A date is a wish. A condition is a measurement.

Andrew's entry: the reserve holds at 2.0 months or more for three monthly measurements in a row. At month 12 he had two: 2.09 and 2.27. One more to go.

One rule sits under this step. Never start a growth launch while any number sits in the red.

3. Walk backwards, one row at a time

Ask of each row: what must be true right before this? Here is Andrew's chain.

Time before opening	When	What must be true
18 months	M12 (today)	Reserve at 2.0 or above for two months; a third is needed
12 months	M18	Premises found; lease signed with a demand-test clause
9 months	M21	Manager for the first location hired and trained
6 months	M24	Equipment bought; second team trained
3 months	M27	Renovation and fit-out complete
1 month	M29	Test week
0	M30	Opening

Look at month 29. A whole test week before the real opening. The launch checklist gets a dress rehearsal while mistakes are still cheap.

4. Check that the money closes the chain

A chain without money is a story. Andrew's rough sum: €4,500 of profit a month at a 15% margin. Twelve months bring €54,000. A manager from M21 to M24 takes €8,000.

That leaves €46,000 against the €45,000 the new shop needs. Almost nothing to spare.

Treat it as a rough estimate. A full one adds taxes, debt payments, the owner's drawings and a reserve for overruns.

5. Read the top row and do it this week

Andrew's top row said: hold the reserve at two months. Not "find premises". Not "design the sign". His launch started with the dullest number in the shop.

That is the whole template: end state, launch condition, chain, money, top row. A blank five-column version sits in the backward planning template.

Game launch maths: the number that pays for the launch

Four out of ten new games on Steam don't even earn back the 100 dollar fee to list them.

Write the number the launch must clear before you write the date. Then turn it into the traffic it needs.

Elena, a freelance designer, did this for her first online book. She priced it at €29 and needed 21 sales a month. At a conversion rate of 1.5%, that meant 1,400 page visits a month. Her launch row sat at month 4.

Now her launch had a leading indicator: page visits, counted every week. A leading indicator points to what will likely happen next. Revenue only tells you what already happened.

How to avoid a launch that can't pay its own fee: put the break-even number in the end-state row. If the leading indicator stays far below its line, you see it weeks before the bank does.

New product failure rate: plan past the opening

Three out of four new grocery products lose their shelf spot within a year.

A launch is a row in the plan, not its last line. Plan the months after opening as rows of the same chain.

Andrew wrote one: the new shop reaches its 15% margin within six months of opening. Any earlier, he could not tell a stable margin from a lucky month. So the judging date sits at six months, and the plan keeps running.

Then write down, in advance, what makes you stop, sell or revise. Do it while your head is clear, before launch week.

How to avoid the empty shelf: give every launch a post-launch target and a date to judge it. Without that date, a slow start and a dead product look the same for months.

The usual launch list against a backward plan

An illustration: one product, two plans.

The usual way reads "Week 1, design. Week 4, build. Week 8, film the video. Week 10, launch." Every row is a task. None is a number. So in week 6 it can't tell you that the launch is already in trouble.

The backward way reads "Six months after launch: a margin. Launch day: a sign-up count. One month before: a test week. This week: the one task that moves the first number." Every row can fail. That is what makes it useful.

This is one piece of a loop: goal, state, gap, levers, plan, forecast. The backward move turns the goal into this week's plan.

"My product is nothing like a coffee shop." Fair. The chain doesn't care. Elena used it for a book, Andrew for a shop, and a thirteen-year-old for a bicycle fund. Any launch with an end state and a date fits on the same sheet.

What a launch without a condition costs

Andrew ran the numbers on the rushed version. Say he had signed the lease at once, with the reserve still below two months. Within a month, rent for the second shop would have pushed the reserve back into the red. Six months of building, gone.

The timing matters even with money in the bank. The opening eats about 1.8 months of reserve. Start at 2.3, and he holds 0.5 the day after opening: red in the first month. At the 1.35 months he had when he first dreamed of it, the money would not even cover the opening.

The slow cost is quieter. If nothing changes in the first shop, its margin stays at 9% and the reserve reaches two months at M16. With the plan working, at M10. Six extra months before the launch condition can even begin to count.

The human cost is plain. The owner who launches on a date spends opening week on the phone to the bank.

Where the template stops

The template gives you the route. It treats a symptom: you stop launching on a wish.

The deeper cause stays in place. A chain built on a wrong starting number runs flawlessly to the wrong place. A fading market sinks the best route inside it. And the chain shows where you are today, not where each row will be in three months.

Two steps close that gap:

The three offsets. A goal that contradicts itself, a starting point measured wrong, a market going under. Each one breaks a plan before work starts. Chapter 11.
Two forecasts and control by forecast. "If nothing changes" against "if the plan works", in dates, so you move the launch before a row slips. Chapters 8 and 9.

Without them, the template tells you what to do this week. It doesn't tell you whether the ship under the launch is sound.

Key takeaways

Write the end state months after launch day, with numbers.
Tie the launch to a measured condition. A date is a wish; a condition is a measurement.
Walk back one row at a time until a row lands on this week.
Put the break-even number and a weekly leading indicator into the chain.
A launch is a row in the plan, not its last line.

Your launch date, or your launch condition

You have two choices tonight. Keep the countdown running and hope the backers show up on the day. Or turn the date into a condition and walk the chain back to this week.

The chain gets you to launch day with a number in hand. The three offsets and the two forecasts keep the launch standing after it.

Test the logic for free: Chapter 1, with the refusal test. Or get the full control system on Kindle, on Amazon. It comes as a Kindle book, with a task to try in every chapter.

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

This topic is covered in Chapter 15, The backward move: from the goal to today of Steer, Don't Guess: Setting Priorities, Reading the Numbers and Correcting Course in Business and Life. The numbers in the examples are illustrative; the book states its limits in Appendix G.