Poldermarketing
What should a freelance strategy and execution package include?
A freelance strategy and execution package is one named person who chooses a narrow bet and ships the work inside a fixed window. Agree the buyer and the message, run one channel on a written cadence, and set a date to keep, change, or stop. A slide deck that nobody builds from is a different product.
One named person ships the bet
A freelance strategy and execution package is one named person who writes a narrow bet and then puts a live artifact in an account the company owns. A slide deck, a leader who only directs, and an agency team are different purchases.
Miniloop, dated 11 May 2026, sells a demand-generation system. A fractional lead directs and does not write pages, run ads, or build campaigns. An agency executes a brief and does not own the strategy. A specialist does a task and does not choose the bet. This package is the person who chooses and ships. See who executes, not only advises.
The U.S. Small Business Administration says a marketing plan turns strategy into action. Most plans cover the target market, the advantage, the sales path, the goals, the channels, a budget, and a date to measure. That document is not this hire. Y Combinator’s essential advice, by Geoff Ralston on 25 September 2017, is principles, not a scope: talk to users, do work that does not scale, pick one or two metrics, and do not grow before you have made something people want. The founder stays in those conversations.
You can contract the actions only
You can contract the actions a person will take. You cannot contract that customers will arrive, or that a post will spread.
Count the actions, reserve time for advice, or stack both. A hybrid stacks counted outputs and reserved time for advice. With no bench, buy the hybrid.
Searchlab puts hourly risk on the buyer and fixed-fee risk on the freelancer once the deliverable is defined. That page sells an AI tool, so leave the pitch. Use hourly only for a fuzzy diagnostic, with a ceiling. Ask for a fixed scope, a maximum number of hours, and an end date.
Media, a custom build, and tools you must keep after the end date sit beside the fee. Stackmatix, which manages Google, LinkedIn, Meta, Reddit, TikTok, and X ads, says an honest package itemizes media, creative, setup, and senior time, and keeps media separate. A bundle that folds the ads into the fee is a reason to ask where the markup sits. You should be able to pause spend without rewriting the person’s time. Stackmatix tells the buyer to push for about ninety days and a dated pause.
Three gates across about ninety days
Buy about ninety days while the channel is still a guess. Treat each gate as a decision you can fail, and sign the table as a scope, not as a standard.
| Window | In the scope | Gate |
|---|---|---|
| Days 1 to 14 | Conversations you join, including a small set of customer interviews, plus a pass over existing data and the closest alternatives. The note names the buyer, the offer, the message, what you will stop, one channel, one measure, the baseline, and what stays with you. | You can say the bet, and you agree the stop rule. |
| Days 15 to 45 | Link tags, tracking, and a weekly note, in accounts you own and inside the fee. One page, sequence, or campaign goes live. | One artifact is live, and the note can force a decision. |
| Days 46 to 90 | A few tests, then a written keep, change, or stop, and notes a next person can use. | A next quarter exists only if the signal is real. |
Stackmatix says most agencies sell a specialist, a full-funnel bundle, or a sprint. The sprint fits while the channel is a guess, a package fits once the lane is known, and the bundle is harder to attribute to one win. Miniloop’s stage order is a specific job very early, guidance later, and a leader plus other people only when there is something to direct. The SBA says the owner should keep a marketing plan current at least once a year. In this window the review is the written keep, change, or stop at day 90.
Book five to eight hours a week for briefs and reviews, Searchlab’s figure, or the gates slip. Name the motions only a founder can run, such as outbound, recruiting, and the story only you can tell in a demo. Stackmatix says packages rarely coach those. Screen the person on what to look for at seed. Searchlab gives a tool no interviews and no strategic judgment. Keep the interviews, the message, and the stop with people. Drafts can sit in a workflow you own.
The same words hide six purchases
The words marketing package are used for a PDF, a monthly quota, a part-time leader, and a team. Who chooses the work, and who ships it, decides the buy.
| What you buy | Who chooses | Who ships | Buy it when |
|---|---|---|---|
| Strategy document | The writer, then you | You, or someone you still must find | Builders are already on the team |
| Ninety-day sprint | One person, with you | That same person | The channel is still a hypothesis |
| Monthly hybrid | That person, with you | That same person, on a counted cadence | The lane is chosen for a quarter |
| Fractional lead | The part-time lead | Other people | You can fund execution and nobody is choosing |
| Agency | You, if the brief is real | A team inside one function | One channel works and you need volume |
| Full-time hire | The hire | The hire, later a team | The motion is stable and should live inside |
Miniloop says most seed teams should keep strategy inside and outsource execution. Positioning needs knowledge an outsider rarely has, so outsource the writing and keep the verdict on the buyer. The channel choice and the first live run can still be one hired person. Without a brief, an agency produces activity and a fractional lead produces a plan.
One channel, maybe outbound or a page, is the freelance shape. Compare seats on fractional lead or agency. Whether anyone outside should start is on when to hire.
Score the scope before the fee
Two proposals can both be honest and still be different objects. Score the scope, and compare the fee only after the objects match.
- One primary channel, and why this buyer is there. Stackmatix’s test: no named channel means a content farm.
- A count of what ships, and who writes, approves, and publishes it. Ongoing is not a count.
- The weekly measure, set before production. One or two metrics, not a vanity total.
- The name on the senior hours. If that time is an extra, the pitch is senior and the work is junior.
- Accounts, link tags, lists, and dashboards in the company’s name. You leave with the pixels, the lists, and the dashboards. Stackmatix’s exit test adds a short notice period.
- A written keep, change, or stop, and about ninety days, not a lock of many months.
- Outside the fee: media, specialist design or development, and tools you keep.
- Your hours booked, including briefing and review, and the motions that stay on your calendar.
- Language and overlap, per task. Searchlab says copy a buyer will read needs a writer who fits that buyer and your working hours. A specified production task can sit further away.
- A revision cap per artifact, and a written out-of-scope line. Work past that line is a change, not a silent extra.
- A named approver and a response window of a few days. Without both, the gates slip.
If the work touches customer lists or pixels, the company decides why that data is used. Where the freelancer only follows that instruction, the Dutch data protection authority treats the company as the controller and the freelancer as the processor, and the scope includes a processing agreement. If the freelancer uses the data for their own purposes, they are a controller too, and that is a reason to walk.
Miniloop’s red flags on that page are a guarantee of a specific result, such as a ranking by a date, no example you can read, no discovery, a high-pressure sale, and a price or scope that will not stay still. Separately, avoid a long lock until you have worked together, and ask who will actually do the work. A report that stops at traffic or likes, and never names the pipeline step that would make you stop, fails the weekly measure. A list, a page, or a calendar is in this package only when that same person also ships the one live artifact.
After the sprint, the operating rhythm is a different design: a repeatable process for a small team.
Hold the package until you can name the buyer
More production will not repair a buyer or an offer you cannot say out loud.
Y Combinator says not to scale before you have made something people want, and that a startup does one problem well at a time. Miniloop calls outside help, before you can direct it, paid guessing, and says someone inside must own the relationship or the work can be technically fine and still sit off the business.
Also wait if nobody can approve within a few days, if month one covers every channel, or if the job is scaling a media budget rather than a small test. If the channel is already proved, hire the operator for that lane. If you still do every task because the bet is unclear, start on when a founder should stop.
The ninety days I would write down
I would name the buyer, the offer, the message, one channel, and the stop date, then build the artifacts in accounts that stay in your name.
I am Piet Baudoin. Poldermarketing is a one-person business. I work in Dutch and English, and I am an AI-native marketer. For freshly funded startups anywhere, I am the all-in GTM option: one remote freelancer for marketing, AI, and automation, from the first message to the first customers. I am strong in AI, content, automation, and workflows. Google Ads and Meta Ads are newer for me. I will set them up and review them with you. I am the wrong hire when the job is scaling a large media budget. You can hire me for that fixed project, or for set days each week. How a project runs is on how I work.
If the message is still fog, settle it on positioning and messaging before a campaign ships. For a read on what a stranger sees, start with the free growth scan.
Questions people ask
Is a slide deck a strategy and execution package?
No. A deck can be one deliverable inside the window. It is this package only when a named person then builds the work in your accounts. The U.S. Small Business Administration calls a marketing plan the document that turns strategy into action: audience, advantage, sales path, goals, channels, and a date to measure. Miniloop says a fractional lead usually does not write the pages or run the campaigns. If the proposal ends when the document is accepted, you bought advice.
Should media spend sit inside the fee?
Keep the professional fee and the media spend apart. Stackmatix, an agency that sells advertising, says an honest scope itemizes media, creative, setup, and senior time, and treats a bundle that folds the ads into the fee as a reason to ask where the markup sits. You should be able to pause spend without rewriting the contract for the person's time. Tools you must keep after the end date sit beside the fee too.
Who should own the accounts and the files?
Your company should. Stackmatix tells founders to leave with the ad accounts, the pixels, the lists, and the dashboards, and to treat a contract that holds them hostage as a reason to walk. The freelancer gets access, not ownership. Agree before the first week how files and feedback move, and what sits outside the scope. On the end date you should still be able to export the work and continue. A clean exit is part of the package.
When is a monthly retainer the wrong shape?
While the channel is still a hypothesis. Stackmatix argues that a fixed window of about ninety days, with a decision at the end, fits that uncertainty, and that a standing package fits once the lane is chosen. A hybrid of counted deliverables and reserved time can still be the shape, as long as the term stays short. Renew only after a written keep, change, or stop.
How involved does the founder stay?
A real share of the week. Y Combinator's 2017 note tells early teams to talk to users and not to grow a motion before they have made something people want, so the founder stays in those conversations. Miniloop says someone inside must own the relationship, or the work can be technically fine and still sit off the business. Searchlab puts briefing and review at five to eight hours a week. If that time is not free, the gates slip.