The role of a technical writer has evolved in leaps and bounds over the last few years, especially with the arrival of AI. While AI has impacted some day-to-day work of a technical writer, it hasn't replaced humans. If you're worried about AI taking over, this article explains why human technical writers are still very much in demand.
The basic skills needed to be a good technical writer have not changed. Even in the AI era, human skills are needed – to write prompts and evaluate the output.
How do you know that technical writing is for you? For starters, do you love to write? What's more, do you like objective writing that’s based on facts? And finally, do you like to organize information in a logical way that’s easy to digest? Exercising your writing muscles to document what high-tech products do is what technical writing is all about.
What is technical writing?
If companies were once looking for a “technical writer”, today they might call it a “content designer” or “technical communicator”. They all come under the general umbrella of technical writing, with different nuances:
- Technical writing or communication – This encompasses information that communicates everything a user needs to know to use a product or solution. No longer limited to text, technical communication also includes how-to videos and infographics.
- Document writing – Originally, technical writers wrote documents that were printed and bound, or burned to a CD and delivered to customers with the software. These documents included user guides, developers’ guides, API documents, and release notes. Today, these documents have morphed into online help.
- Online help – The instructions for using almost any product can be found on the internet. Sometimes online help is only available to paying customers. Users often visit online help sites when they need to figure something out, so it has to be easy to search and find answers.
- Content strategy – With so much information out there, technical writers need to consider content design or content strategy: How is information structured, and what content best meets an audience’s needs? Arranging information logically is also known as information architecture.
- UX writing or microcopy – Any text that appears in a product's interface is crafted by a writer. The interface text guides the user on how to use the product and enhances the user experience. App buttons, menu options, and error messages all contain text that was written by a technical or UX writer.
What technical writing is not
Technical writing is not marketing or copywriting.
Writing for marketing aims to generate sales, usually by convincing someone to buy the product. It grabs people’s attention, like a fishing hook, and then reels them in to buy something.
But once customers buy the product, marketing is no longer relevant. It doesn’t help users figure out how things work, or what to do if they get stuck.
Technical writing doesn’t try to sell, but rather helps people use the product. Technical writing walks users through step-by-step procedures, explains features, and solves problems. In fact, good technical writing can even double as a presales tool – potential customers might ask to see the documentation to understand how a product works before buying it.
Skills you need in your toolbox
What skills do you need in your technical writing toolbox to succeed? Here are some essential skills of the trade, from being organized and curious, to customer-focused and AI-savvy:
Be organized
A technical writer needs to make sense of things, like categorizing information or chunking it into clear topics. This makes it easy for customers to find what they’re looking for. The volume of digital information we produce continues to grow exponentially from year to year. For customers consuming technical documentation, this information explosion poses a potential hazard: Can they find and access the information they need? The question of “Where is the information?” can become more important than the content itself. When you gather and organize information, you:
- Structure information logically to make it easy to find. The rule-of-thumb for organizing a lot of information topics is to start with the broadest or most general category, and drill down to more specific categories.
For example, if a product has five major features, there should be a separate category for each one. Within those general categories, you can have more specific categories, such as prerequisites, setup procedures, workflows, and troubleshooting. - Give context to show customers where they are. No one wants to get lost, especially in the middle of a complex installation or configuration procedure! Navigation bars, breadcrumbs, and starting every section with a topic sentence help to provide context.
- Keep track of what you document and where, for example, with a task-tracking tool like Jira or Monday. This might seem like meta-work (work about work), but it’s important to have a reference record for the future, know the current status of your work, and keep tabs on what you’ve accomplished.
Be curious, ask questions
Documenting a new feature or enhancement is like mapping a 3D cube: There are at least six aspects you need to cover. The more questions you ask, the more complete and useful your documentation will be.
If you’re using the feature for the first time, what would you want to know? Examples of questions you might ask your development or product team include:
- What does this feature do?
- Do I need a license or special permissions, like admin privileges, to use it?
- What are the prerequisites for using it?
- How do I turn it on or activate it?
- What are the default settings? Can I adjust or customize the settings to fit my needs?
- Are there any limitations in using this feature?
But don’t stop there. Zoom out and get an overview of the big picture too:
- What problem is this feature solving, or what use case does it address?
- How does this feature connect to other parts of the product?
When you ask these questions, you’re not grilling your colleagues – you’re making sure you have enough context to explain things clearly. A lot of times, developers will assume certain details are “obvious,” but what’s obvious to them isn’t always obvious to the end user. You can bridge that gap just by being curious.
Be customer-focused
When you’re writing, you need to constantly keep in mind: What’s in it for the customer? Being customer-focused means stepping into the customer’s shoes and looking at the product from their perspective – not through the lens of a developer or sales executive, or even the CEO.
Customers aren’t interested in a long list of technical details just for the sake of it. They don’t care how many months it took to build a feature, who built it, or how clever the solution is. What they do care about is how this feature helps them solve a critical problem, save time and money, and make their work easier.
Instead of listing a feature's “bells and whistles,” ask yourself: How does this new feature make the customer’s life better?
- Focus on outcomes, not the technology – Don’t just explain what the feature or procedure is – make it clear why it matters, and what the user gains from it.
- Talk the way customers think – Avoid developer jargon unless necessary and explain every procedure in the context of customer goals.
- Anticipate pain points – Writing a troubleshooting section is helpful because it gives solutions or workarounds to common frustrations users might experience.
When your documentation speaks directly to the user’s needs, it becomes more valuable and user-friendly.
Write in simple English
If you want customers to understand, you need to use clear, simple English with proper grammar that gets straight to the point. Writing in second person (“you”) helps to make content sound user-friendly.
Avoid long, complicated sentences that take someone up the garden path, or overly sophisticated words that could confuse the user. Technical jargon is fine, as long as you explain it the first time.
Many companies use style guides to make sure content is presented consistently and easy to scan. Even without a style guide, you can structure your writing so it’s easy to read, easy to follow, and avoids ambiguities.
To help customers follow complex information, you can use structural elements to break it down into smaller parts that are easy to digest. These elements include headings, bulleted lists, numbered steps, and tables. This is the art of visual chunking. It’s easier to scan information when it’s presented in distinct sections or small chunks, rather than a single unbroken block of text.
Balance the big picture with details
Technical writing aims to juggle two perspectives: seeing the big picture (macro) while paying attention to the fine details (micro). The big picture is the context that helps customers understand why a feature matters, how it fits into their workflow, and why they need to use it.
Once the big picture is clear, you can zoom in and explain the granular details. That could mean describing a procedure, presenting reference data like parameters in a table, or mentioning limitations.
If you only provide detailed steps without context, users won’t understand why they’re doing it in the first place. Being effective as a writer means striking a balance between general and specific, usually by giving a bird’s-eye view or contextual explanation before diving into the nuts and bolts.
Be a team player
Technical writing isn’t done in a vacuum. A big part of the job involves collaborating with product managers and developers to get information, have drafts reviewed, and stay up-to-date on which features make the final cut and what has been postponed or descoped. Having regular conversations with your teammates helps to build trust.
The review process sometimes feels like a ping-pong match. You send a draft, your colleagues comment, you revise, and it goes back again – until the product manager approves it. This back-and-forth happens a lot in UX writing, where even a single word can spark debate. It takes patience and flexibility to find the wording that makes sense for the product, customers, and the business.
When you’re documenting an evolving product, you need to keep your eye on what actually ends up in the product, not what was originally envisioned or planned. This requires honest communication and a readiness to adjust course as needed. Being proactive and asking what’s actually being released keeps the documentation accurate.
AI as a tool: Capabilities and limitations
Think of AI as a tool in your writer’s toolbox. It can be leveraged for various tasks, provided it can access the relevant information and you give it proper instructions. Poor input = poor output. AI cannot create something out of nothing.
How does AI help with new features?
AI is trained on Large Language Models (LLMs) or existing data sets. So, when you give AI a prompt, it gives you an answer based on the data it was trained on.
If the data doesn’t exist or isn’t accessible on the internet, AI will either admit that it doesn’t know or it will hallucinate. There's still a considerable knowledge gap between what AI knows and what field engineers or subject matter experts (SMEs) know. Humans working on a product are generally more knowledgeable about it than AI.
Remember that AI is a tool that requires human prompting. So, as a human technical writer, your skills remain highly valuable.
AI assistants for technical writing
For AI to be effective, it requires a human to guide it and verify its output. Here are some productive ways you can leverage AI in your technical writing:
- Writing assistant – Ask AI to summarize, simplify, and clarify. Use it to write an overview of a topic, explain concepts, summarize a long block of text into one paragraph, and explain complex ideas or processes in a simple way.
- Editing assistant – Get AI to review your documentation for consistency and accuracy, or check that your documentation follows style guidelines, uses accurate terminology, and a standardized format.
- Brainstorming assistant – AI can be a great icebreaker to start writing if you don’t know how to begin. Get the ball rolling by asking AI to suggest opening sentences for a specific topic or outline ideas.
- UX writing assistant – Generate ideas for UX writing, especially ones that are a bit tricky, like error messages, and choose or combine the best ones.
- Research assistant – Use AI as a learning tool by asking questions about industry trends, technical concepts, and use cases. Make sure you ask AI to give you real-life sources and not to fabricate incidents.
- Problem-solving assistant – Leverage AI to help you write troubleshooting sections by asking how it would solve specific issues.
- Glossary assistant – Ask AI to give definitions for glossary terms and arrange them alphabetically. Usually accurate and concise, these definitions still need to be verified by a human writer to check that they’re relevant to your product.
- Insight assistant – Give AI multiple data sources and ask it to extrapolate insights. For example, if you get a lot of customer feedback, you can ask AI to summarize the positive and negative feedback and suggest action items for improvement.
- Code assistant – Leverage AI to understand the meaning of code snippets that you have been asked to include in the documentation.
All the above assistants are actually AI agents, which you can define with a set of instructions. Once you have defined an agent and its task, you can re-use it as needed.
Prompting is a skill in itself
AI delivers impressive results quickly. But remember that AI is only as good as the instructions you give it. No clear input = no useful output. For an AI agent to be helpful, you need to give it precise instructions about what you expect to get in the output. It takes skill to write prompts that are specific, concise, and tailored to what you need.
In a way, prompting is both a science and an art. If it’s too vague, you’ll get generic, unhelpful responses. If it’s too detailed, you might spend more time on the prompt than it would take to write the content yourself.
What AI cannot do
AI won’t think for you. It’s there to help clarify, expand, or polish your writing. The creative thinking, product expertise, and customer-first perspective still need to come from you as the writer.
Here are some current limitations of AI:
- AI is not a prophet – AI cannot write about a future product that has not yet been developed. It only knows the past and whatever exists in current LLMs. It does not know the future or specialized knowledge that exists outside an LLM.
- AI cannot deal with complex or ambiguous situations – Some scenarios require critical and creative thinking to come up with solutions that are understandable and acceptable to customers. AI falls short in this area, which is still the domain of humans.
- AI cannot validate itself for accuracy – Any response that AI gives always has to be checked and validated by a human expert, since it might not be accurate.
- AI cannot sign off on content for publication – At the end of the day, it’s only a human writer who can approve AI-generated content for publishing.
AI is a super-productive and efficient tool, but it’s not a replacement for technical writers, nor is it a substitute for human critical thinking and creativity. However, if you can harness its power, you will have a super tool or a super assistant at your side.
Where are the best places to work?
Technical writers are in demand across a huge variety of industries. Almost every company that builds complex products or solutions needs a technical writer. The hottest opportunities are in software – especially cloud platforms, cybersecurity, and SaaS (software as a service). These companies create products that need online help, release notes, and quick updates.
You’ll also find technical writing roles in manufacturing, healthcare, aerospace, defense, telecommunications, and energy companies. They all rely on accurate and reliable documentation to meet customer expectations and comply with regulations. In short, if people need to use, install, troubleshoot, or integrate a product, they need documentation and someone to write it.
Yes, I want to be a technical writer
If you think technical writing is the career for you, then consider having an online presence on LinkedIn. This is the tool of choice for many HR departments. It’s highly recommended to set up your LinkedIn profile before you even start applying for a job. Sites like Glassdoor and Dun’s can give you information on salaries, company culture, and employee reviews. Whether you aim to work at a leading tech giant or a growing startup, there’s a company out there that needs your skills.
Are you ready to take the plunge? As we live in a tech-saturated world, the demand for technical writers is constantly growing. So, buckle up and good luck!