The Substack Writer's Tech Stack: What Top Newsletter Writers Actually Use

Ask a successful Substack writer what tools they use. Most will say "Substack." Press harder. Ask where they draft. Where they research. Where they brainstorm headlines. Where they check grammar before hitting publish.

The answer is never just Substack. It is always a stack. A collection of tools, each doing one job well, wired together into a workflow. The newsletter you read in two minutes took four or five tools to produce.

I spent the last year studying how top newsletter writers work. Interviewing them. Reading their process posts. Watching their workflows. Here is what I found.

Why Substack Alone Is Not Enough

Substack is a great publishing platform. It handles subscriptions, email delivery, payment processing, and audience growth. It does these things well. What it does not do well is help you write.

The Substack editor is basic. It handles text, images, headings, and links. That is about it. No AI assistance. No grammar checking beyond your browser's spell check. No research tools. No version history worth mentioning. No way to try three different openings and compare them side by side.

This is fine if you write short posts off the top of your head. It is not fine if you write researched, edited, polished newsletters that people pay $10 a month to read. For that, you need more tools.

Every top Substack writer I talked to drafts somewhere else and publishes on Substack. The question is not whether you need a stack. It is which stack.

The Typical Stack (What Most Writers Use)

Here is what most newsletter writers have cobbled together over time. It works. It is not optimal.

Research: Scattered Across Five Places

Most writers collect research in a mess of browser tabs, bookmarks, and saved tweets. The common tools are Perplexity or Google for web research. Twitter/X bookmarks for ideas and threads. Pocket or Instapaper for saved articles. Apple Notes or a random Google Doc for rough notes.

The problem is obvious. Research lives in five different places. When you sit down to write, you spend the first 30 minutes hunting for that article you saved last Tuesday. Or that tweet thread someone posted about retention rates. You know you saved it somewhere. You just cannot find it.

Drafting: Notion, Google Docs, or a Plain Text Editor

Notion is the most popular drafting tool among newsletter writers I surveyed. Google Docs comes second. A few writers use iA Writer or Apple Notes. One used Scrivener. Several draft directly in Substack, though they all said they wished they did not.

Notion is popular because writers already use it for everything else. Content calendars, subscriber databases, sponsor tracking. Adding a drafts database feels natural. But Notion's editor is built for team wikis, not long-form writing. It is slow on large documents. It does not support Markdown shortcuts well. And its AI features rewrite entire blocks instead of making surgical edits.

Google Docs works because it is familiar and collaborative. If you have an editor who reviews your drafts, Google Docs is the path of least resistance. But its AI features (Gemini) are shallow. And exporting from Google Docs to Substack means fighting with formatting every single time.

AI Assistance: ChatGPT in a Separate Tab

Most writers using AI have ChatGPT or Claude open in a separate browser tab. They copy a paragraph, paste it into the chat, ask for a rewrite, then copy the result back into their draft. Some use it for brainstorming headlines. Some for outlining. Some for punching up a weak conclusion.

The copy-paste loop is slow. It breaks formatting. It loses context. After 20 messages, the AI forgets what your newsletter is about. And you never get a clean view of what the AI actually changed. You just get a wall of new text and have to spot the differences yourself.

Grammarly sits on top of this as a browser extension. It catches typos and basic grammar issues everywhere you type. Most writers have the free tier. A few pay for Premium.

Publishing: Copy-Paste Into Substack

The final step is moving the finished draft into Substack. If you drafted in Notion, you copy and paste. Some formatting survives. Some does not. If you drafted in Google Docs, you copy and paste. Bullet lists sometimes break. Bold text sometimes vanishes. You spend 10 to 15 minutes fixing formatting issues every single time you publish.

If you are curious about the specifics of this formatting problem, we wrote a detailed guide on ChatGPT to Substack formatting that covers the common pitfalls.

Graphics: Canva or Midjourney

Header images matter for click-through rates in email and social sharing. Most writers use Canva for clean, template-based graphics. A growing number use Midjourney or DALL-E for AI-generated header images. Some skip graphics entirely and let the text speak for itself.

The Problem with This Stack

Count the handoffs. Research in Perplexity. Notes in Apple Notes. Draft in Notion. AI rewrites in ChatGPT. Grammar check in Grammarly. Publish in Substack. That is six tools and five handoff points.

Every handoff is a friction point. Formatting breaks when you move between tools. Context is lost when you switch windows. Time drains away in the transitions. A writer producing two newsletters per week might spend three to four hours just managing the handoffs between tools. That is three to four hours not spent writing.

The bigger problem is cognitive. When you are drafting in Notion and switch to ChatGPT to rework a paragraph, you break your flow. You shift from writing mode to prompting mode. You think about how to phrase instructions to an AI instead of thinking about what you are trying to say. Then you switch back and try to pick up where you left off. That context switch is expensive.

The Better Stack

A better stack reduces handoffs. Fewer tools. Fewer transitions. More time in flow state. Here is what I recommend after testing dozens of combinations.

Research: Perplexity + NotebookLM

Perplexity handles outbound research. You ask it questions and it returns answers with inline citations to real sources. Not hallucinated URLs. Actual links to actual pages. For a newsletter writer researching a topic, this replaces the 30-tab Google session that used to eat up an entire morning.

NotebookLM handles inbound research. Upload your saved articles, interview transcripts, PDFs, and notes. It creates a searchable knowledge base from your sources. Ask it questions and it answers using only your material, with citations to the exact source. No hallucinations. No information leaking in from training data.

Together, these two tools replace the scattered mess of bookmarks, saved tweets, and random notes. Perplexity finds new information. NotebookLM organizes the information you already have. Both are free to start. Perplexity Pro at $20 per month is worth it if you write research-heavy newsletters.

Writing + AI Editing: Athens

This is where the biggest improvement happens. Athens replaces both your drafting tool and your AI assistant with a single editor. You write in a clean Markdown editor. When you want AI help, you highlight text or type a natural-language instruction. The AI proposes edits directly in your document. Green highlights for additions. Red strikethrough for deletions. You see every change and accept or reject each one.

No copy-pasting between windows. No losing formatting. No trying to spot what changed in a wall of rewritten text. The AI works inside your document, not in a separate chat.

For newsletter writers specifically, this solves three problems at once. First, you draft and edit in the same place. No handoff between a drafting tool and an AI tool. Second, you keep full control. The AI suggests. You decide. Every change is visible before you accept it. Third, your document is always in Markdown. When you move it to Substack, formatting transfers cleanly.

Athens costs $99 per year. For comparison, Notion AI costs $96 per year and only rewrites entire blocks. ChatGPT Plus costs $240 per year and still requires copy-pasting. Athens gives you the drafting environment and the AI editing in one tool for less money.

For a deeper look at how AI editing differs from AI writing, see our guide on AI writing tools vs. editing tools.

Grammar: Grammarly Free Tier

Grammarly's free tier catches typos, subject-verb agreement errors, and basic punctuation issues. It runs as a browser extension, so it works everywhere you type. The free tier is all most newsletter writers need.

You do not need Grammarly Premium if you are already using Athens for AI editing. Athens handles the higher-level edits: tightening prose, cutting filler words, improving structure. Grammarly handles the mechanical layer: catching the typo you missed, flagging the comma splice you did not notice.

Two layers of checking. One free. One built into your editor. Between them, very little slips through.

Publishing: Substack

Substack stays in the stack because it is the best at what it does: building an audience, managing subscriptions, and delivering newsletters. You do not need to replace it. You just need to stop using it as your writing tool.

The workflow is simple. Write and edit in Athens. Export or copy the finished Markdown. Paste into Substack. Because the source is clean Markdown, formatting transfers without the usual breakage. No more spending 15 minutes fixing bullet lists and bold text after every paste.

Comparing the Two Stacks

The typical stack uses six or more tools with five handoff points. Research is scattered. AI lives in a separate tab. Every transition costs time and formatting.

The better stack uses four tools with three handoff points. Research is consolidated. Writing and AI editing happen in one place. Formatting survives the transitions.

The cost difference is small. Perplexity Pro at $20 per month. Athens at $99 per year. Grammarly free. Substack free (or taking a cut of paid subscriptions). Total: roughly $340 per year. That is less than ChatGPT Plus alone.

The time difference is significant. Consolidating your drafting and AI editing into one tool saves 30 to 60 minutes per newsletter. Over two posts per week, that is four to eight hours per month. Over a year, that is two to four full working weeks reclaimed.

What About Drafting Directly in Substack?

Some writers skip the stack entirely and draft directly in Substack. This is tempting. Zero handoffs. No formatting issues. Everything in one place.

It works for short, casual posts. It does not work for polished, researched newsletters. You have no AI assistance beyond what a browser extension provides. No version history. No way to experiment with multiple drafts. No Markdown support. No offline access. And if Substack ever goes down or changes their editor, your entire workflow breaks.

Drafting in Substack also makes it harder to repurpose content. If you want to turn a newsletter into a blog post, a Twitter thread, or a chapter in a book, you need the source in a portable format. Substack's editor does not give you that. Markdown does.

How to Transition to the Better Stack

You do not need to change everything at once. Here is a phased approach.

Week one: Sign up for Perplexity and try using it for research instead of opening 30 browser tabs. Keep everything else the same. See if your research phase gets faster.

Week two: Try drafting your next newsletter in Athens. Write the draft, use the AI editing to tighten your prose, then paste the result into Substack. See if the editing phase feels better than copy-pasting with ChatGPT.

Week three: If weeks one and two felt good, drop NotebookLM into your research workflow. Upload your sources for your next newsletter and try querying them instead of re-reading everything manually.

Week four: By now you have the full stack running. Audit the result. Are you faster? Is the writing better? Are there fewer formatting issues when you publish? Most writers who make this transition never go back.

The Stack in Practice: A Real Workflow

Here is what a typical newsletter production looks like with the better stack.

Monday morning (30 minutes): Open Perplexity. Research the topic for this week's newsletter. Save the key sources and findings. If you have existing material (interview notes, saved articles), upload them to NotebookLM.

Monday afternoon (60 minutes): Open Athens. Write a rough draft. Do not edit as you go. Just get the ideas down. Use your Perplexity research and NotebookLM queries to fill in facts and citations.

Tuesday morning (45 minutes): Edit the draft in Athens. Use the AI editing to tighten each section. Accept the changes that improve the prose. Reject anything that changes your voice. Run through the piece one more time yourself.

Tuesday afternoon (15 minutes): Copy the finished piece into Substack. Add the header image. Set the subtitle and preview text. Schedule or publish.

Total time: about 2.5 hours. The old stack with scattered research, drafting in Notion, AI rewrites in ChatGPT, and formatting fixes in Substack took closer to four hours for the same quality of output.

What the Best Newsletter Writers Have in Common

After studying dozens of successful Substack writers, one pattern stands out. The best writers are not the ones with the most tools. They are the ones with the fewest handoffs. They have a tight, intentional stack where each tool does one job and does it well.

They do not draft in their publishing tool. They do not use a general-purpose chatbot for editing. They do not scatter their research across five apps. They have a system. They trust it. They spend their energy on the writing, not the workflow.

The tools are not the point. The writing is the point. But the right tools remove enough friction that the writing comes easier. And when writing comes easier, you write more. You publish more consistently. Your newsletter grows.

Build your stack. Keep it tight. Spend your time writing, not managing tools.

For more on building the ideal writing stack in 2026, check out our guide on the best writing apps for Substack, or learn about writing newsletters with AI.