How to Review What You Shipped and Learn From It
How to review what you shipped: check it against the outcome you predicted, then review your reasoning, so your next guess is better than your last one.
Review what you shipped by checking it against the outcome you predicted before you built it, on a date you set in advance. Then review your reasoning, not just your output. Without both, you never learn whether your judgment is any good.
How to review what you shipped
Quick test. Think of the last three things you released.
Which one worked?
If you hesitated, that's the finding. Most solo operators ship, feel relief, and move on to the next thing within the hour. The release is "done". Whether it did anything is a question for nobody, because there's nobody else to ask.
That habit has a price. It's the difference between ten years of experience and one year lived ten times. You keep making the same estimation errors, with the same confidence, because nothing ever told you they were errors.
Review one: did the outcome move?
This one is about the thing you shipped.
When a release goes out, discovery's work isn't finished. There is a prediction attached to it: we expect this to produce that. Now you check.
The sequence is simple. The goal said what should be different. Discovery said this change would get you there. You built it. Now the measurement says whether the reasoning held.
For this to work you need two things in place before you ship.
- A stated outcome. Not "improve onboarding". Something you could look at and say yes or no to. The handoff note from discovery should already carry it.
- A date. Put the check in the calendar when you ship, not "sometime". A check without a date is the same as no check.
Then, on the date, look. Did the outcome move? Don't hunt for a flattering number. Look at the one you named.
Free guide · 21 pagesDiscovery and delivery, one person.The Dual-Track Playbook: how to keep discovering while you keep building.Get the free guide →What to do with the answer
There are only three results, and each has a job.
| What you see when you review what you shipped | What it tells you | What you do next |
|---|---|---|
| The outcome moved as predicted | The reasoning held | Note what made the prediction good, and feed it into the next discovery cycle |
| The outcome did not move | An assumption was wrong, or the change was the wrong route | Go back to the assumptions. Which one failed? Decide whether to fix, replace or remove it |
| You can't tell | The outcome was too vague, or you never set up a way to see it | Fix the measurement first. Next time, define it before shipping |
The middle row is where the learning lives. A miss is not a verdict on you. It is the cheapest lesson available, because the work is done and the information is free.
Results should also flow back into the quarter's goal and into the next round of questions. If you ship, measure, and file the answer where nobody will see it again, you've done half the loop.
Review two: was your thinking any good?
This one is not about the thing. It's about you.
Reviewing only what you shipped makes you better at shipping things nobody needed. So once a cycle, set aside half an hour and ask about the quality of your decisions instead of the volume of your output.
- Which assumptions turned out wrong?
- Where was I overconfident?
- Which tests actually changed a decision, and which were theatre?
- How long did discovery really take, against what I expected?
That last question is the sneaky one. Compare your estimates with what happened. If every "two days" turns into a week, that's not bad luck. That's a pattern you can correct, but only if you've written down both numbers.
This half hour produces the calibration that makes the whole practice compound. It's also the one nobody schedules, because it has no deliverable. Schedule it anyway.
Where the answers live
Here's a trap. You do the review, learn something sharp, and by the next cycle it has evaporated.
Working alone, your records are your institutional memory. They are the only version you get. So keep it short and keep it findable. Four lines per question: what I asked, what I did, what I found, what I decided. The handoff note and the post-ship check slot into the same place. Over a year, this record shows you something you can't see in the moment: where your assumptions are systematically wrong. Maybe you always overrate how much people will change their habits. Maybe you under-test whether anyone cares. A year of notes shows the pattern. A year of memory doesn't.
If you're not sure how to run the questions behind those notes, the post on running discovery and delivery as one person lays out the day to day.
The three excuses that skip the review
You will want to skip this. Here are the usual reasons, and why none of them hold.
"I know whether it worked." Do you? Or do you remember how it felt? A launch that felt flat can have moved the outcome. A launch that felt great can have moved nothing. Feelings are a poor instrument and you are the only one reading it.
"There's no time." The check takes minutes. The thing it prevents is repeating an estimation error for a decade. Compare the costs.
"The result is bad and I don't want to look." Fair. Nobody enjoys it. But the bad result is the most useful one, because it names an assumption you got wrong while the cost of finding out is still small. The version where you find out in a year costs far more.
Say you ship a change to how people get started with your offer, expecting more of them to finish setup unaided. A week later you look at the one outcome you named. It hasn't moved. Now you know the assumption "they will understand it" was wrong, you know it early, and you can go back to a cheap test instead of building the next thing on top of a bad guess. That is the whole value of the habit in one example.
Free assessmentHow trapped are you, really?Take the Corporate Suffocation Index, the free solopreneur readiness assessment.Take the assessment →Keep the review light
I'll say it plainly. This can go wrong in the other direction.
A review that acquires a template, a scoring sheet and a quarterly ritual is not a review. It's bureaucracy, performed by one person for an audience of nobody. You'll quietly drop it within two months, and you'll be right to.
The test is simple. If doing the check takes longer than running a small test, the process has become the point. Four lines and a decision. That's the standard.
Also, do not skip it because you're busy. Busy is exactly when you ship the most, which is when the loop teaches the most.
A simple monthly rhythm to review what you shipped
Here is a rhythm that fits a business with limited hours.
- When you ship: write the predicted outcome and book the check date.
- On the date: look at the outcome you named. Record the result in the same place as your notes.
- Once a month: ask what discovery decided and what you shipped. Two blank months in a row is a warning.
- Once a cycle: spend the half hour on your reasoning. Which assumptions failed? Where were you overconfident?
- Carry the answers into the next round of questions, and into the decision about what moves from discovery to building.
That fifth step is what turns a review into an engine. The learning isn't a report. It's an input. If you want to see the other side of that handoff, look at deciding when an idea is validated.
Free guide · 21 pagesDiscovery and delivery, one person.The Dual-Track Playbook: how to keep discovering while you keep building.Get the free guide →What you get when you review what you shipped
You will find out which of last quarter's releases actually worked. Not which ones felt good. Which ones moved something. That sounds small. It isn't. It means your next guess is better than your last one. Compounded over a year, that is the difference between a business that drifts and one that gets sharper.
Free assessmentHow trapped are you, really?Take the Corporate Suffocation Index, the free solopreneur readiness assessment.Take the assessment →Frequently asked questions
How soon after shipping should I check the result?
Choose the date when you ship, based on how quickly the outcome could show up. The key is that it exists and sits in your calendar. A vague "I'll look at it soon" is how the loop never closes.
What if I didn't write down a prediction before shipping?
Write one now, honestly, and mark it as written late. Then start attaching a stated outcome to every item from now on. A late prediction is flawed, but it's better than nothing.
Is reviewing the reasoning different from reviewing the results?
Yes. Results tell you what happened. The reasoning review tells you why you expected something else, which assumptions failed and where you were overconfident. It takes about half an hour once a cycle.
What if the outcome moved but I'm not sure my release caused it?
Say so in your notes. Note what else changed and treat the result as weaker evidence. If the call matters, run a small follow-up test instead of assuming credit.
Do I need a tool to review what you shipped?
No. Four lines per item in a place you can find later is enough. If the process takes longer than a small test, it has become too heavy.
The full method, with the checklists, is in The Dual-Track Playbook, a free 21-page playbook.
How trapped are you, really?
Take the Corporate Suffocation Index, the free solopreneur readiness assessment.