Reverse JournalReverse Goal Setting: How to Plan Backward From Done
Start at the finish, ask what had to be true before it, and stop when you reach something you can do today.
Reverse goal setting, also called backward planning, means defining the finished goal first. You then work backward in time, asking what had to be true just before it, and before that, until you reach an action you can take today. It replaces the usual forward guess ("what should I do first?") with a chain of concrete milestones that each lead to the next.
What backward planning is
Forward planning starts where you stand and lists what might come next. The list grows quickly, most items feel equally urgent, and it is hard to tell which ones matter.
Backward planning fixes the end point. If the goal is a published book, you ask what had to be true the week before publication: a final file, a cover, a launch list. Then the month before that, and so on. Each answer is small enough to picture and points to the next one. Related approaches include backward design (curriculum planning) and backward chaining (teaching a skill step by step). The core move is the same: begin at the outcome and step back.
Reverse Journaling uses this move as one step of a loop that also includes writing the goal as already done, naming your main obstacle and logging what you did. See Reverse Journaling for the whole loop.
Why the planning fallacy makes this worth doing carefully
People are optimistic about how long things take. Roger Buehler, Dale Griffin and Michael Ross (Journal of Personality and Social Psychology, 1994) called this the planning fallacy. In one study, students predicted finishing their thesis in an average of 33.9 days. It took 55.5 days, and only 29.7 percent finished by the time they predicted. Across other studies, the pattern held for academic and non-academic projects: predictions ran shorter than reality, and fewer than half of participants finished on time (source).
The same paper reported two things you can use. First, people focused on plan-based scenarios instead of their own past experience. Second, in one of their experiments, the optimistic bias disappeared when participants were asked to connect relevant past experience to their predictions.
The practical lesson: a backward plan is only as honest as the durations you put in it. Check each milestone against a similar thing you have done before, and add a buffer.
The reverse goal setting method in six steps
- Write the goal as done. Past tense, checkable, with a date. "I published my book by 1 June and 10 readers have sent reviews."
- List what had to be true at the finish. Three to five items. These are your acceptance criteria.
- Step back one year, 90 days, one month, one week. At each step ask: "What had to be done before this?" Write one sentence each.
- Sanity check against your past. For each milestone, recall the last time you did something similar and how long it really took. Adjust. This is the Buehler step.
- Name the obstacle at the tightest link. Find the milestone most likely to slip, and write an if-then plan for it. Peter Gollwitzer and Paschal Sheeran's 2006 meta-analysis of 94 tests found a medium-to-large effect of if-then plans on goal achievement (source).
- Pick today's action. One thing you can do in about 45 minutes that moves the nearest milestone. Do it, and log it on your done list.
Three worked examples
Example 1: publish a book
| Horizon | What had to be true |
|---|---|
| Done | Book is live, 10 readers have written reviews |
| 1 month before | Files proofed, cover final, launch list of 50 readers ready |
| 90 days before | Edited manuscript back from editor |
| 6 months before | Full draft complete |
| This month | Chapters 1 to 4 drafted |
| Today | 45 minutes on chapter 2 |
Obstacle plan: "If I start rereading earlier chapters, then I write the next paragraph first."
Example 2: run a 10k
| Horizon | What had to be true |
|---|---|
| Done | Finished the 10k, felt strong in the last kilometer |
| 2 weeks before | Longest run of 8 km completed, then an easier week |
| 8 weeks before | Running three times a week, 5 km without stopping |
| This week | Three runs on the calendar |
| Today | 30-minute easy run, logged |
Obstacle plan: "If it is cold at 6 am, then I put on my running clothes before I decide anything."
Example 3: launch a business
| Horizon | What had to be true |
|---|---|
| Done | Launched, first 10 paying customers |
| 1 month before | Working product, pricing page, 20 warm leads |
| 90 days before | Five customer interviews done, offer defined |
| This week | Three interviews booked |
| Today | Send 10 interview requests |
Obstacle plan: "If I want to polish the logo, then I make one outreach call first."
How to write a good milestone
A milestone is only useful if you can tell whether it happened. Three habits help.
- Use a state, not an activity. "Working on the draft" is an activity. "Chapters 1 to 4 drafted" is a state. You can look at a state and say yes or no.
- Keep it inside your control. "Editor returns manuscript" depends on someone else. "Manuscript sent to editor" depends on you. Where you must depend on others, add a buffer and a follow-up date.
- Make each one small enough to picture. If you cannot imagine what you would do on a Tuesday to reach it, break it in two.
When a milestone slips, do not scrap the plan. Ask which link was too optimistic, fix that duration, and move the later dates. Slipping once is information. Slipping every time at the same link usually means you named the wrong obstacle, so write a new if-then plan for it.
Forward vs backward planning
| Question | Forward planning | Backward planning |
|---|---|---|
| Starting point | Where you are | The finished outcome |
| Typical output | A growing task list | A short chain of milestones |
| Main risk | Busywork with no end | Overly optimistic timing |
| Fix for the risk | Define the end point | Check durations against past experience, add buffer |
Where reverse planning can go wrong
- Fixed dates with no slack. Given the planning fallacy, build in extra time. If the task feels like two weeks, write three.
- Too many milestones. Four or five steps back is enough. More than that and you are guessing at details that will change.
- Treating the plan as fixed. Review weekly. The date can move. What should not move is the habit of checking.
- Skipping the action. A plan that ends at "chapter 2 by Friday" without today's action is still a wish. Pick the 45-minute task.
Keep the chain honest with evidence
A log of what you actually did is the best defense against optimistic planning. It shows how long things really took, which improves your next estimate. Teresa Amabile and Steven Kramer reported that making progress in meaningful work contributed most to a positive inner work life among the workers they studied (source). A nightly line is a cheap way to see that progress.
Keep the log short. One line a night is enough, and a short honest entry beats a long one you stop writing by Thursday.
If you want a guided version, Start From Done builds a backward milestone plan from your done statement, from the finish back to today. The core is free. For more depth on the thinking behind it, see how to manifest your dreams and the WOOP method.
Questions
What is reverse goal setting?
It is planning that starts from the finished goal and works backward in time. At each milestone you ask what had to be true just before it, until you reach an action you can take today.
What is backward planning?
Backward planning is another name for the same idea: fix the end point first, then step back through milestones. Related approaches include backward chaining and backward design.
What is the planning fallacy?
It is the tendency to underestimate how long tasks will take. In Buehler, Griffin and Ross's 1994 studies, students predicted 33.9 days for a thesis and needed 55.5 on average.
How far back should I plan?
Four or five steps is usually enough: one year, 90 days, one month, one week, then today. Beyond that you are guessing.
How do I stop backward plans from being too optimistic?
Compare each milestone to a similar task you finished before, and add a buffer. Buehler and colleagues found that in one of their experiments, linking predictions to past experience removed the optimistic bias.