Research and data collation, providing direction, quick thinking assistance, rephrasing and presentation, and meta details generation are five ways LLMs can elevate your content production process.
In the rest of the article, let’s understand each of them and learn how to do them with ChatGPT (or any of its competitors).
1. Research and data collation
This is arguably the most essential LLMs bring to the table as far as content creation is concerned. Finding context-specific information has always been a challenge because Google has killed blogging with its dumb E-E-A-T guidelines which reward generic but keyword-rich content.
For instance, if you are writing about “how mechanical keyboards help increase efficiency”, you will have to search for broad topics. First, you will learn about how mechanical keyboards are different from membrane keyboards and what it does to the overall typing experience.
Then, you will tie it all together in your mind to explain how it boosts efficiency in a day-to-day work setting. This methodology, although works, takes a long time, especially for niche topics that don’t have a lot of information on the internet to begin with.
(I know there are plenty of articles that talk about how mechanical keyboards make you more productive. It’s just an example.)
However, with AI tools, you can simply ask your question and get a personalized answer. The question can be framed in a natural language which streamlines the research process even further. You don’t have to collect information, churn them in your head, and spit out the sentences.
There were plenty of articles that cost me hours to gather the information. The writing process also becomes disjointed as you may have to revisit some of these resources mid-sentence.
ChatGPT, thanks to its huge database, can get you context-specific information, in the right scope in one tab. The biggest benefit here is the reduced cognitive load. It will make your salary worth it as you won’t be pulling your hair (I can’t relate cause I am bald) to find the right resources and cycling through multiple tabs.
Plus, you can quickly understand any complex or novel concept and terminology at your level of comprehension by asking ChatGPT to “dumb it down.”
2. Providing direction
You must have encountered writer’s block at least a couple of times in the past week, especially when writing short-form content and introductions of articles. The frequency of these moments will drastically reduce with AI-powered chatbots.
The easiest approach here is to tell your intentions with the content piece and ask for ideas with a blurb explaining it. You can provide more details in your follow-up messages to get more refined ideas.
Other ways include asking ChatGPT for a multi-perspective analysis of a particular opinion or statement. This is quite helpful for diving into the minds of the target audience and figuring out their expectations from your content.
There are a few ways you can use LLMs to get unstuck:
Share a screenshot of the section where you are lost and ask the chatbot what you can write there
Generate 5-6 variations of the paragraph or sentence you may need
Seek suggestions for the right angle for your content based on your core message
Even if the output is total trash, it will help you think out of the box and the right way forward will pop up in your mind.
3. Quick-thinking assistant
Have you ever struggled to come up with an example that a health-conscious busy corporate professional can relate to? What about finding a better phrase or term for “something that makes you go faster?”
There are plenty of such quick problems that bamboozle authors in the middle of a sentence which affects productivity. Yes, the golden rule of efficient writing is “write first, edit later” but sometimes you are just stumped, especially with minor issues.
Tell you what, there were plenty of instances when I wondered if a sentence in front of me was grammatically correct. Just yesterday, I was confused about whether it was “by-product” or “byproduct.”
ChatGPT takes care of it all.
However, there is a big error — huge responses.
You just want a short and concise context-specific answer to your silly question. It is the equivalent of checking what 25+75 is in a scientific calculator. And ChatGPT’s default response is longer than a typical health insurance claim proceeding. That’s anything but brief.
I have created a simple yet effective Quick Answers GPT which answers all your burning questions within 50 words or less. I frequently use it to get synonyms, quotes that I can’t quite remember, quotes I remember but forgot who said it, and check grammatical correctness.
It is also coded to check the internet before answering time-sensitive questions related to topics like sports, news events, and weather.
4. Rephrasing and presentation
Let’s say you have written an article titled “top 10 mechanical keyboards for writers” and want to include a table at the beginning. The table would contain the name of the keyboard and a bunch of other columns like price, rating, customizability, connectivity, and features.
Manually, it will take you hours. ChatGPT can do it within minutes.
There are many other applications where you can extract specific information from your existing content pieces and present it in a different way. For instance, you may want to convert a couple of paragraphs into a set of short bullets.
Be mindful that these tasks require manual prompting which requires a bit of testing at first. Also, the content produced by GPT, even if your article was the resource, still needs to be proofread and refined.
5. Meta details generator
Meta details include the meta description, excerpt, subtitle, alternate title suggestions, alt texts for images, and URL slug. This information educates the search engines about the nature of your article and plays a pivotal role in improving rankings.
However, writing all of these meta details is tedious and can take about 30 minutes. Since it is only relevant to the search engines, you can leverage LLMs for it. I’ve created the Article Meta GPT for this purpose.
You can provide whatever data you have, the title, an outline, or even the complete draft, and it will return all of the meta details based on that. Plus, if you are revamping an older article, you can drop the published link and it will read the article to produce the details.
Conclusion: Use AI the right way
You can’t tighten a screw with a hammer. Even if you try, you will do more harm than good.
AI is quite powerful and can deliver all the promised benefits like improved content quality and reduced turnaround time. But only when used the right way. It disappoints me immensely when I see other authors or professionals use LLMs for tasks it’s not equipped to do.
The best way forward is to educate yourself about how these technologies work, play with the relevant tools, and continually iterate your approach.
—
Next Chapter: ChatGPT Is Best for Content Writing (4 Reasons)
Previous Chapter: Generative AI in Content Writing: Pros and Cons
Index (with Prologue): Content Writing With AI: Principles, Strategies, and Prompts
Commenti