Digital Change

AI Masterclass: Newsletter structure - How to design a monthly customer newsletter with Gemini, ChatGPT and Claude

Written by Lars-Thorsten Sudmann | Mar 17, 2026 5:15:00 AM

Your company wants to communicate regularly with customers. So you start a newsletter. In the beginning, this is often still a motivated process: Collect topics, write an intro, include three updates, add a call-to-action, done.

After two or three issues, the same thing usually happens: The newsletter becomes irregular, too long, too random or sounds different every time. Sometimes it is too promotional, sometimes too dry, sometimes just an unsorted list of internal information.

This is exactly where AI can help - but not by "just writing a newsletter". The real leverage only comes when you specify a clear structure and use AI specifically for topic selection, structure, tonality and variants.

Why the problem remains

Many teams already use AI for content selectively, but without a system.

Typical mistakes are:

  • A complete newsletter is generated directly, without fixed sections.
  • Every employee prompts differently.
  • There is no reusable template for the subject, introduction, main topics and conclusion.
  • The content is based on internal priorities rather than genuine customer benefit.

The result: AI produces text, but no reliable communication structure.

So the real bottleneck is not the writing. The bottleneck is the structure.

The concrete solution: the 5-building block model for monthly customer newsletters

You don't need a complex editorial system for a functioning monthly customer newsletter. You need a simple, repeatable framework.

I recommend this 5-block model:

1. relevant start

Don't start with corporate phrases, but with a clear hook:

  • What's new?
  • What is important?
  • Why is it worth reading right now?

2. main topic of the month

A newsletter needs a focus in terms of content. This could be

  • a new solution
  • a customer case
  • an industry trend
  • a specific tip
  • a product improvement

3. two to three short update blocks

This is where you place smaller messages:

  • Team news
  • dates
  • events
  • New resources
  • References to blog, whitepaper or webinar

4. clear call to action

Every issue should know exactly what the next step is for the reader:

  • Book an appointment
  • Read article
  • Request a demo
  • Attend an event
  • Send a reply to the newsletter

5. recognizable conclusion

A short, consistent conclusion creates brand character and saves time in creation.

This turns "We need to send something to customers again" into a predictable communication process.

How to use Google Gemini, ChatGPT and Anthropic Claude for this process

It gets exciting when you don't just use one tool, but compare the same newsletter process with several AI systems.

This is precisely how you can quickly recognize which tool works best in your company for which work step.

Process step 1: Gather briefing and material

Before you open a tool, collect:

  • Target group
  • Aim of the newsletter
  • Topics of the month
  • Existing content such as blog posts, event information or product updates
  • Desired tonality
  • Desired length

Only then do you go into AI.

Process step 2: Design structure with Google Gemini

Gemini is well suited to this workflow if you are already working in the Google ecosystem. Officially, you can use Gemini directly in Gemini Apps, upload files and develop content or include sources in Google Docs with "Ask Gemini" or writing functions. Depending on the function, this requires registration, a suitable plan or activated workspace functions.

A sensible use in the newsletter process is

  • Upload topic material or provide it in Google Docs
  • Ask Gemini to build a monthly structure from this
  • Then generate several category suggestions
  • Vary subject lines and introductions

Example prompt:

Create a monthly newsletter structure for 
existing customers.
Goal: inform, strengthen trust, show relevance.
Target group: medium-sized companies.
Tone: professional, clear, approachable.
Build the following structure: introduction, main topic, three short
short updates, CTA, conclusion.
In addition, formulate three subject lines and two
variants for the introduction.

Typical strength of Gemini in the process:

If your team works with Google Docs anyway, the transition from input to draft text is often very direct.

Process step 3: Variants and fine-tuning with ChatGPT

ChatGPT is particularly strong for this use case if you want to systematize the workflow. According to OpenAI, ChatGPT allows you to create projects, bundle chats, upload files and store project-specific instructions. In addition, paying users can create their own GPTs for recurring tasks.

In practical terms, this means for your newsletter creation:

  • Create a "Monthly customer newsletter" project
  • store style rules, target group and fixed categories there
  • upload reference material
  • Use the same framework every month

It becomes even more efficient if you define your own GPT for newsletter production, for example with

  • fixed output structure
  • brand voice
  • typical CTA formulations
  • editorial rules
  • no-go formulations

Typical strength of ChatGPT in the process:

Repeatability. If you want to turn an individual task into a standard process, this is a great advantage.

Process step 4: Tonality and summarization with Claude

Claude is particularly suitable if you want to develop a calm, consistent text from several notes, documents or collections of input. Anthropic describes that Claude offers projects for organizing knowledge and chats; in addition, Claude can now create and edit files directly, while Artifacts are used for interactive or visual work states.

In the newsletter process you can use Claude for:

  • Condensing into clear, readable paragraphs
  • Reformulation into a consistent tone of voice
  • calmer, less promotional formulations
  • better transitions between topic blocks

Typical strength of Claude in the process:

Raw material often turns into a very clean, readable overall text - especially if you place value on style, reading flow and linguistic consistency. This assessment is a practical classification of the workflow, not an official product promise. It is supported by the fact that Claude offers projects, learning resources and file-based workflows for professional use.

Direct comparison: This is how you proceed in practice

A sensible testing process looks like this:

Round 1: Same prompt for all three tools

Give Gemini, ChatGPT and Claude exactly the same information:

  • Target group
  • Newsletter target
  • list of topics
  • Desired structure
  • tonality
  • CTA target

Round 2: Evaluate results according to four criteria

Evaluate each output according to:

  • Clarity of structure
  • Relevance to the reader
  • Linguistic quality
  • Reusability

Round 3: Define tool roles

Then you define roles instead of beliefs.

For example:

  • Gemini for raw structure in the Google workflow
  • ChatGPT for standardized templates and variants
  • Claude for stylistic refinement and reading flow

This way, you don't use AI arbitrarily, but process-related.

Practical example

A consulting company would like to send out a customer newsletter once a month. This is how it works so far:

Beforehand:

  • Topics are collected spontaneously
  • Someone writes a text "quickly"
  • Subject lines are created on demand
  • Every issue looks different
  • the opening rate fluctuates greatly

Then an AI-supported workflow is introduced.

New:

  1. The team collects four to five topics every month.
  2. Gemini uses these to create three structural proposals.
  3. ChatGPT builds the complete newsletter based on a fixed template.
  4. Claude revises the text for clarity, tonality and flow.
  5. The marketing team selects the best final version and sends it out.

Afterwards:

  • Consistent structure
  • Less coordination effort
  • Faster production
  • clearer communication
  • more professional overall impression

The key difference is not that AI "writes texts".

The difference is that an unclear writing process becomes a standardized content workflow.

Steps that can be implemented immediately

To get started right away, implement these five steps:

1. define a fixed newsletter architecture

Make it binding:

  • Introduction
  • Main topic
  • Short updates
  • CTA
  • Closing

2. create a standardized master briefing

A single document is enough:

  • Who reads?
  • What is to be achieved?
  • What language is appropriate?
  • Which topics occur regularly?

3. test all three tools with identical input

Don't decide based on gut feeling, but compare with the same material.

4. document the best prompts

The best results are almost never achieved by chance. Record working prompt patterns.

5. build a standard process from them

Determine

  • who provides topics
  • who creates the design with AI
  • Who does the final approval

Strategic classification

Many companies underestimate newsletters because they see them purely as a marketing channel.

In reality, they are a strategic communication format:

  • for customer loyalty
  • for trust
  • for visibility
  • for positioning
  • for cross-selling
  • for thought leadership

If you don't structure this process, your newsletter will remain dependent on individuals, daily form and time pressure.

However, if you set it up properly with AI, the result is a scalable communication process that delivers better quality month after month with less effort.

This is precisely where the competitive advantage lies: not in the tool itself, but in its systematic use.

🚀 Next step

If you not only want to understand AI, but also use it in a structured way in your company, then:

👉 Find out more about our AI training:
https://bloo.school

👉 Find out about our Smart Market Fit offers:
https://bloola.com/smf - The Smart Market Fit course
https://bloola.com/smf-system - The Smart Market Fit system for companies

👉 Or find out more about our consulting and automation solutions:
https://bloola.com