Research software picks products. Go-to-market planning decides what you will do when the product underperforms, and no subscription writes that for you. The real alternative is a two page written plan with dated checkpoints and stop rules agreed before launch. Build it in the seven steps below.
The short version
- A research suite answers what to sell. A go-to-market plan answers what happens in week nine. Two different documents, and only one of them protects your capital.
- Write your stop rules before the purchase order goes out. Written afterwards they are negotiable, and you will negotiate with yourself.
- Every step should produce an artifact. A number, a document, or a dated decision. If a step produces only a feeling, cut it.
- There are three endings: scale, fix, kill. A plan with no kill branch is a hope with a spreadsheet attached.
- The cheapest alternative to any software is a calendar invite with a decision written into it.
Write the plan before you buy the tool
Open a blank document and write the plan first. Buy software afterwards, if the finished plan turns out to need it. I put those in that order because I once did them in the other order, and it cost me real money.
I had a product that was not working. Advertising was running, the rating was mediocre, conversion was flat, and every week I told myself the next campaign restructure would turn it around. I poured money into it for three months. It never turned around. What I was missing was not data, because I had plenty of data. I was missing a line, written down in advance, that said: if these numbers look like this on this date, we stop. The whole method below came out of that.
Jungle Scout is a product research suite and this page is not a review of it. What I can describe is the part of go-to-market planning that tools of that class cannot perform, because that part is made of decisions, and a decision needs an owner with a name.
The seven steps, and the artifact each one owes you
- The one line thesis. Why your product beats a specific named incumbent. Artifact: one sentence naming the competitor and the defect you fix, drawn from their one and two star reviews rather than from a brainstorm.
- The unit economics sheet. Landed cost, marketplace fees, freight, expected return rate, and contribution per unit at three price points. Artifact: a sheet where the margin still survives a realistic cost of acquiring a customer.
- The capital plan. What the launch costs before it earns anything. A single product launch runs about $8,000 to $15,000 all in. A five product brand runs $25,000 to $50,000. Artifact: a figure you can lose without changing your life.
- The channel plan. Name the two or three sources of traffic you will run, in what order, and name the ones you are deliberately not running this year. Artifact: a dated order of operations.
- The evidence run. A first quantity of around 200 units with $5,000 to $10,000 behind it. Up to four products can be tested in parallel if capital allows. Artifact: rating, conversion rate, and acquisition cost measured at real volume.
- The decision date. A real date, set now, when you sit down with the numbers from step five and choose. Artifact: a calendar invite with the thresholds written in the body.
- The three branch plans. Write the scale plan, the fix plan, and the wind down plan today, while you are calm and nothing is at stake. Artifact: three short plans, exactly one of which you will run.
Step six is the one everybody skips, and it is the only step that makes the other six binding.
Scale, fix, or kill, decided in advance
| Signal | Scale looks like | Fix looks like | Kill looks like |
|---|---|---|---|
| Rating trend | Stable at or above the category norm | Slipping from a fixable cause such as packaging, sizing, or instructions | Slipping because of the product itself |
| Return rate | Flat and inside your model | Elevated, cause identified, correction costed | Elevated, cause is the design, and retooling is uneconomic |
| Conversion rate | At or above category and rising with review count | Below category with a main image that was never tested | Below category after image, price, and copy have each been tested |
| Acquisition cost | Falling as organic placement builds | Flat, but the campaign structure has not been rebuilt yet | Rising while spend rises, across the full window |
Two rules govern this table. Write it before launch, and attach a window to every row. A metric with no window is not a criterion, it is a mood: a bad fortnight should not trigger a kill, and a comfortable quarter should not delay one.
Where software still earns its money
Research suites are good at breadth. They tell you a category exists, about how big it is, and who is winning it. That is worth a subscription during selection.
What no suite does is name the defect worth fixing, judge whether a factory can hold the tolerance your fix requires, or tell you the truth on the decision date. So the honest alternative set is short: the written plan on its own, the written plan plus specialists hired for sourcing and creative, or a managed operator who owns both the plan and its execution. Put the same question to all three. Ask what would make you tell me to stop, then ask what that answer costs them.
What most agencies will not tell you
Kill criteria are commercially awkward for anyone who bills monthly. A dead product means a smaller account, and a smaller account means less revenue for the person advising you. That conflict is real and you should price it into how you read any recommendation to keep spending. It is also why I would rather publish our criteria than be asked about them: rating trend, return rate, conversion rate, and acquisition cost trajectory, judged over a defined window.
The second omission is your own time. Software is quoted in dollars per month while the seller's hours are quietly counted as free. A launch takes about four to six hours a week of the owner's attention, settling to about two hours a month once a brand is steady. Multiply your own hourly value by that and compare the total, not the subscription line.
Related answers
- Alternatives to Viral Launch for FBA
- Ranking best tools for Amazon launch management
- How to test product-market fit on Amazon
- Amazon brand launch timeline stages
- Amazon launch services: the complete guide
Send us the plan and we will tell you which step is missing, free, at Flapen.
