DreamFoundry — Digital Products for Working Creatives

From Prompt
to Publish

The exact workflow used in paid client work across brand, website, and positioning projects

You don't need more prompts. You need better output — and a system that tells you what to do with it. This is the six months of iteration, already done, so you can ship sharper brand, positioning, and web copy faster without sounding like AI.

8 Categories
28 Tested Prompts
15+ Years in Practice
$97 $147

Founding price ($97). Moves to $147 after initial release.

Get the System →

One-time purchase. Instant download.
Works in any browser. No subscription.

8 categories, judgment-annotated throughout
28 prompts with copy buttons — ready to run
Before/after examples from real client work
Multi-AI research workflow (Claude + ChatGPT + Gemini)
Lifetime access — free updates included

Works with Claude, ChatGPT, and Gemini. Free tiers are enough to start — full breakdown below.

The Problem

AI can write. That doesn't mean it can write for your clients.

You get generic output that strips the voice. Every AI draft sounds like it was written for the same brand. You spend as much time fixing it as you would have writing it yourself.

Prompt packs aren't built for practitioners. "500 ChatGPT prompts" means 500 starting points with no judgment about when to use them, when to skip them, or what to do with the output.

You're rebuilding the workflow from scratch every time. A custom GPT here, a saved prompt there, a Notion doc you've already forgotten — nothing connects into a system you can actually rely on.

You spend more time fixing AI output than writing yourself. Every draft comes back generic. You rewrite the whole thing. The tool that was supposed to save hours is creating them.

The issue isn't AI. It's that building a system takes time you don't have.

You could build this yourself. Most senior practitioners intend to. A custom GPT here, a saved prompt there, a Notion doc started in January and abandoned by February when a client project took over. Sound familiar?

Building a workflow like this properly — running prompts across multiple models, noticing what fails and why, developing the judgment layer, connecting the modules into a sequence that actually holds — takes 40 to 60 hours of focused, unbillable work. That's the real cost of building it yourself.

This is that work, already done. This is the exact workflow used in paid client engagements across brand voice, website copy, and positioning work. Every module tells you when to use it, what good output looks like, what to ignore, and what the next step is. You bring the expertise. This gives you the system.

If you're already using AI

This isn't about generating more content. It's about getting to usable output faster — and knowing what to do with it when you get there. The system is built for practitioners who already know AI produces the first draft, not the final one.

"AI is not the writer. It's the sparring partner."
The core principle — from the product introduction
What This Looks Like in Practice

Tomorrow morning. A real client. The system in use.

8am

You drop a client's messy homepage copy into Module 02. You get a sharp positioning teardown in under 2 minutes — specific problems, not vague feedback.

8:15

You run Module 06 with the client's brand voice principles already loaded. Three homepage hero directions come back. One is close. You push it.

9am

You run the Quality Filter on the version you've refined. It comes back a 7.5. You know exactly which line to fix to make it an 8. You fix it yourself.

10am

You send the client three homepage options with your recommendation. A homepage rewrite that would have taken all day took two hours — because you skipped the blank page, not the thinking.

What didn't change

✓

Your judgment still drove every decision

✓

The copy sounds like you wrote it — because you did

✓

The client has no idea how fast you moved

✓

You billed for the expertise, not the hours

A typical homepage rewrite: 6–8 hours. With this system: 2–3. The time savings goes into thinking, not typing.

If you're earlier in your career, this gives you a structure you haven't built yet. If you're experienced, this replaces hours of setup work you don't have time for. Either way, you still do the work that matters.

What the System Produces

The difference isn't the prompt.
It's the process behind it.

❌ Generic AI output (no system)

"We help businesses grow with professional, approachable brand messaging."

Sounds like it was written for everyone. Converts no one. The brand voice is gone. The positioning is indistinct. A competitor could say this word-for-word without lying. This is what happens when you skip Modules 1 and 2 and go straight to writing.

✓ Output from this workflow

"Your pitch is clean. Your website still sounds like everyone else's."

A specific person reads this and thinks: "That's me." The positioning pressure test ran first. The voice fingerprint informed every word. The scenario work made the emotional register precise. This is what happens when the judgment runs the process — not the prompt.

What's Inside

8 categories. Every stage of the workflow.

From brand voice extraction to client outreach — each category covers a critical stage of the copywriting and brand strategy process, with annotated prompts designed to keep your judgment in the loop.

Start Here — Common Workflow Paths

New client website rewrite

01 → 06 → 08 → 09 → 10

Voice extraction → hero → services → CTA audit → local SEO layer

Brand voice project

01 → 02 → 03 → 04 → 05

Extraction → audit → channel calibration → competitor mapping → style guide

Research-heavy engagement

13 → 11 → 12

Industry immersion first → parallel research → multi-AI synthesis

New business outreach

25 → 23 or 24

Prospect research first, then Loom script or cold email — always in that order

LinkedIn content sprint

19 → 21 → 22

POV post from real work → client story → 30-day calendar from one core idea

🎙
CATEGORY 01

Brand Voice Architecture

5 prompts covering voice extraction, consistency audits, channel calibration, competitor mapping, and style guide documentation. The foundation everything else builds from.

Best: Claude5 Prompts
🏠
CATEGORY 02

Website Copy

5 prompts for hero rewrites, About pages, services positioning, CTA audits, and local SEO layering. The DIY version of the HereHere Creative website rewrite process.

Claude / ChatGPT5 Prompts
🔄
CATEGORY 03

Multi-AI Research Workflow

4 prompts for parallel research across Claude, ChatGPT, and Gemini — with a synthesis prompt to reconcile all three into a single authoritative brief. The highest-leverage research method in the system.

All Three Models4 Prompts
❤️
CATEGORY 04

Mission-Driven Messaging

4 prompts for nonprofit and advocacy clients: impact statements, crisis communications, donor copy, and advocacy message architecture. Built from real mission-driven engagements.

Best: Claude4 Prompts
💼
CATEGORY 05

LinkedIn & Content Strategy

4 prompts for turning real client work into LinkedIn content — POV posts, contrarian takes, client stories, and a 30-day calendar from one core idea. No generic tips content.

Claude / ChatGPT4 Prompts
📬
CATEGORY 06

Client Outreach & Prospecting

3 prompts for top-of-funnel new business: pre-outreach prospect research, personalized Loom audit scripts, and cold emails built for one person — not a list.

Gemini → Claude3 Prompts
✉️
CATEGORY 07

Email Copy

3 prompts for campaign sequences, newsletter content, and re-engagement — the email use cases that come up most in client work, with failure mode notes for each.

Best: Claude3 Prompts
🤝
BONUS

Talking to Clients About AI

The question every copywriter is dreading right now. A framework for three client types — Skeptic, Curious, Enthusiast — with verbatim language and a conversation prep prompt. No competitor has this.

Best: ClaudeFramework + Prompt
Honest Assessment

Built for practitioners.
Not beginners.

This is for you if —

  • You're a senior freelance copywriter or brand strategist with real client work in flight
  • You've started building your own AI workflow at least once and abandoned it when a client project took over
  • You're already paying for Claude Pro or ChatGPT Plus but getting outputs that need full rewrites
  • You can tell the difference between good AI output and bad AI output
  • You charge for your expertise and judgment — not your hours — and need to protect that
  • You want a connected workflow, not a list of prompts with no context

This is not for you if —

  • You want AI to write the copy for you, start to finish, without editing
  • You're looking for 500 prompts that cover every possible scenario
  • You're new to copywriting or brand strategy and still building the fundamentals
  • You want a course on how to write copy — this assumes you already know
  • You already have a refined AI workflow you're genuinely happy with — this won't add much
Alex Stahlmann

Alex Stahlmann

Founder — Studio Low Five / HereHere Creative

15+ years in brand voice, positioning, and copywriting
Boutique studio serving restaurants, nonprofits, tech companies, and mission-driven orgs
Built real multi-AI research workflows on live client engagements
Co-founder, HereHere Creative — website copy rewrite service with local SEO layer
Based in Minneapolis. Every prompt in this system has been run on actual client work.
Why This Exists

You could build this yourself. Here's why most don't.

I built this system because it's how I actually work. Not to package something for sale — but because after years of running prompts across multiple models on real client engagements, I had something that worked consistently and was worth documenting.

The build-your-own path is slower and more painful than it looks. I know because I did it. You have to run the prompts, notice when they fail and why, develop the judgment layer that tells you when to skip AI entirely, and connect everything into a sequence that holds across different clients and use cases. That's the work. It takes time you're not billing for.

Every annotation in this system — every "what good looks like," every "what to ignore," every "judgment note" — came from a moment where the output was wrong and I had to figure out why. That's what you're buying. Not the prompts. The 15 years of knowing when they're working and when they're not.

Founding Offer

One price. Lifetime access.

This is the founding price — $97, moving to $147 after initial release. No countdown timer. No fake urgency. When it moves, this page updates and that's the new price.

When you buy, you get the full interactive system immediately — 8 categories, 28 prompts with one-click copy buttons, before/after examples, judgment annotations, and the multi-AI research workflow. Every future update is included.

If you've already paid $2,997 for a copy course that's been sitting unopened, you understand the value of a focused, practical system you'll actually use.

FOUNDING PRICE — Early Buyers
$97 $147

One-time. No subscription. Instant access.

8 categories, judgment-annotated throughout
28 prompts with one-click copy buttons
Real before/after examples from client work
Multi-AI parallel research system
Model recommendations for each prompt
Judgment annotations — when to use, when not to
Lifetime access + all future updates
Get the System — $97 →

Instant download. Works in any browser.
Questions? hello@dreamfoundry.io

Common Questions

Answered directly.

🛠

Before you buy — the toolkit this requires

This system is built to run across Claude, ChatGPT, and Gemini. You don't need all three for every module — but you'll get the most from it if you have access to at least Claude and ChatGPT.

CLAUDE

Free tier works for most modules. Claude Pro ($20/mo) is a meaningful upgrade for multi-AI synthesis and crisis communications — those two specifically require context depth the free tier often can't hold. If you only pay for one subscription, make it this one.

CHATGPT

Free tier works for most modules. ChatGPT Plus ($20/mo) gives better results on positioning pressure tests and headline generation — the difference is noticeable but not essential to getting value from the system.

GEMINI

Free tier works. Gemini Advanced (~$20/mo via Google One AI Premium) gives meaningfully deeper research output — better competitive analysis, sharper prospect intelligence, more thorough industry immersion. Worth it if research modules are a core part of your workflow.

Free tiers across all three will get you somewhere useful. If you're investing in subscriptions: Claude Pro first (synthesis and crisis modules), Gemini Advanced second (research and competitive intelligence), ChatGPT Plus third (positioning and headlines). Full stack runs about $60/month — one solid client engagement covers it.

Do I need paid subscriptions to all three models?

No — free tiers across all three will get you somewhere useful. That said, paid tiers do improve specific modules. Claude Pro ($20/mo) makes the biggest difference: multi-AI synthesis and crisis communications both require context depth and reasoning precision the free tier often can't sustain. Gemini Advanced (~$20/mo via Google One AI Premium) gives meaningfully deeper research output — better competitive analysis, sharper prospect intelligence. ChatGPT Plus ($20/mo) helps on positioning and headline modules but is the most optional of the three. Priority order if you're choosing: Claude first, Gemini second, ChatGPT third.

Why wouldn't I just build my own system?

You could — and some people will. But building a system like this properly means running prompts across multiple models, noticing what fails and why, developing the judgment layer, and connecting everything into a sequence that holds across different clients and use cases. That's 40 to 60 hours of focused, unbillable work. Most practitioners intend to do it and don't — client work wins every time. This is that work, already done. If you already have a refined AI workflow you're genuinely happy with, this probably won't add much. If you've started building one and abandoned it, this is the version that's finished.

Is this a course or a reference system?

A reference system. It's not a course on copywriting — it assumes you already know how to write. Think of it as a practitioner's annotated workflow: you open it when you're working on a client, run the relevant prompt, and use your judgment to take the output somewhere useful.

How is this different from a generic prompt pack?

Every prompt includes three things a generic pack doesn't: a "when to use it" judgment layer, a "what good looks like / what to ignore" annotation, and a clear next step that connects it to the rest of the workflow. The prompts are also built from specific real-world use cases — not written to cover every theoretical scenario.

I'm an in-house marketer, not a freelancer. Will this work for me?

Yes — Modules 1 through 7, Module 10 (LinkedIn), and the Multi-AI Research system all apply directly to in-house work. The outreach and client-facing modules are more relevant to agency and freelance contexts, but everything else translates cleanly to internal brand and content work.

What format is the product delivered in?

An interactive HTML file that works in any browser — no app, no login, no subscription required. It has a sidebar navigation, one-click copy buttons on every prompt, and a clean reading experience. Download it once and use it offline anytime.

What's included in future updates?

New prompts as they're developed and tested on real client work, refinements to existing modules based on user feedback, and any new workflow integrations (new AI models, new use cases). Founding-price buyers get all updates free, forever.

Ready to Work Differently

Your judgment.
Amplified.

28 prompts. 8 categories. One system that keeps your expertise in the loop at every step. The founding price won't last long — and it shouldn't.

Get the System — $97 →

One-time. Lifetime access. Updates included.