Building Your Cyborg Content Strategy: Research and Planning

Dave Snyder


March 25, 2024

Our first installment gave an overview of what the cyborg approach to content development was all about. In this next part, our journey into this methodology will focus on the planning phase — the first step in the process. This is the most critical phase to ensure quality content outputs. Without proper cyborg content strategy research and planning, your end product runs the risk of lacking consistent voice, tone, factual accuracy, and formatting.

By understanding how to properly develop a Human + AI planning model, you build a strong foundation for your long-term Cyborg content system. There are several steps you will want to prioritize during the planning phase, which I break down in this series so you can build a cyborg content strategy.

If you haven’t yet, be sure to review our intro to the Cyborg Content Strategy.

Jump ahead:

Research in the Cyborg Content Strategy

The most important lesson to learn early when working with AI models is that if you rely only on their training knowledge you are unlikely to get high-quality responses.

In other words: garbage in, garbage out.

The AI models don’t understand the difference between factual information and disinformation. They do not prioritize based on expertise. This makes the creation of quality research to guide the model imperative.

So, our goal here in the research phase is to build a document that can then be utilized to generate an outline within our brief. There are three key areas to look at during the research process:


Your research will not get far without the proper tools, and the right tools will ultimately guide you in how to format your prompts for eliciting the best possible output from the AI model.

Utilizing APIs will be the best bet for pulling relevant data to guide briefing and content planning, no matter the topic or vertical.

12 Things To Look for in a Content Strategy Analysis Tool

Data-Rich APIs

Data-centric content works really well with the current LLM models on the market. By tapping into data-rich APIs you can give your content accuracy and authority. Here are some examples of a few of the data-rich APIs out there:


Automotive APIs, such as Car API, give you access to a wide range of data, including vehicle specifications, vehicle identification number (VIN) decoding, vehicle history reports, fuel efficiency data, maintenance schedules, diagnostic information, and a lot more.

By utilizing these industry-specific APIs, you can bridge a gap between human-generated and AI research, further streamlining content planning and development processes such as in the case of creating and scaling authoritative and accurate content for automotive clients.

Car API provides this kind of information through its platform. Built for developers, it’s free to use and pulls data on key information such as make and model, engine and mileage, design features, and mechanical specifications.

The API also performs VIN lookups for vehicles from 1990 and newer. Here are some more of the key highlights of this automotive API:

  • REST + JSON API fully documented in OpenAPI.
  • Available HAL+JSON and JSON-LD formats.
  • 80,000+ vehicles continuously updated.
  • Human-readable text and RGB values for 1990 vehicles and newer.
  • Three paid subscription tiers to fit diverse business needs.

Stock Market

For anyone who is looking for accurate financial information, is an excellent API. It aggregates data not just from the biggest financial players on Wall Street but goes further to gather information in other small markets.

There are four primary sources of data this API pulls from: stocks from all U.S. exchanges and dark pools, indices from the S&P, Dow Jones, and more, all U.S. options trades and quotes, and currencies, including Forex and crypto.

This is essentially giving access to the entire ecosystem of U.S. equities data, and it’s accessible from this single platform. If you are working with clients and creatives in the finance sector, a tool like Polygon can be extremely valuable for pulling accurate information to utilize within your content. Here are a few more key features of this API:

  • Multiple access methods, including flat files, Restful and WebSocket APIs, and SQL query options.
  • Tutorials and examples to use alongside your projects.
  • Free and paid subscription options.


Sports writers, experts, and reporters must rely on up-to-date and real-time data to bring sports stories to target audiences. API-Sports is one such platform that delivers this information to users from several years of available data.

Tools like this make it ideal for gathering sports stats, pre-match and live-game odds, and scores that are updated every 15 seconds.

API-Sports also does a good job at making this information easily accessible for anyone and not only developers and programmers. The range of sports data coverage is wide as well and easily pulls stats for football, baseball, basketball, hockey, and a lot more.

Start with the free plan and access the data you need straight from a single dashboard. You will also have access to the platform’s other APIs with your free registration. These are also included with the free plan, but you can upgrade at any time:

  • Premade widgets for customizing update frequency.
  • Access to code libraries in multiple languages.
  • Seamless integration across all platforms.
  • Free and paid plans are available.

Real Estate

Real estate APIs like the Zillow Group API, provide a wealth of valuable data that allow users to aggregate mortgage, MLS, “Zestimate’s”, sales and rental transactions, and public data sets.

Not only that, but Zillow Group offers access to almost 20 available APIs through its platform, enabling you to access tons of up-to-date information.

This is extremely useful for content development, as it promises accurate, detailed information on hundreds of thousands of real estate types. Aside from this, you can also pull additional insights like:

  • Real estate performance data.
  • Agent review management.
  • Neighborhood and real estate metrics.
  • Access to 15 years of public records across the entire U.S.


You can find and use data-rich APIs for literally almost anything, including entertainment topics.

With the IMDb API, for example, you can access IMDb data through the AWS Data Exchange to find information on practically any movie, TV show, or video game — including ratings, essential metadata, box office performance metrics, viewer safety ratings, and more.

Here are several more examples of what this powerful tool gives you:

  • Access to over 10 million movies, TV shows, and video games and 13 million cast and crew members.
  • Insights into in-development projects and upcoming features.
  • Structured canonical data sets.
  • Parents’ guide.
  • Trivia and facts guides.


If you are a developer, then ScraperAPI is a great solution for data and website scraping. If you aren’t a developer, though, don’t worry. Its DataPipeline tools make it easy to schedule and download crawls without code.

The downside on the DataPipeline is that you’ll get unparsed HTML for web pages, and that can eat up context tokens when you’re working with your model.


Agenty is, in my opinion, a better selection for no-code scraping. It offers great tools in its system, but it also has a Google Chrome plugin that can make free scraping with parsed results into JSON a breeze.

The Campaign Manager’s Guide to Solving Data Issues


Claude 3 models, such as what Anthropic uses, are currently my go-to for creating research documents. These allow for 200,000 token context windows in the prompting, which means you can put about 150,000 words of information in your prompt.

This keeps you from having to multi-shot prompt your research document. The results are on par with GPT4 models in terms of ability to follow directions and the quality of output.

Custom GPT

If your resource material for your research is something you have internally or in document formats, csv, or PDF then building a custom GPT is a great way to generate quality research documents.

When building and using a custom GPT bot for research, like in OpenAI’s ChatGPT, you can simply upload the files you need to generate your research documents into the GPT’s knowledge base.

When uploading files, you should keep in mind the following limitations:

✅ The maximum file size for any individual file is 100MB.

✅ The total limit for the size of files that you can upload in a single batch is 1 GB, with a maximum of 50 files.

✅ If you have larger files to upload, we recommend splitting them into smaller parts or considering compression to ensure they fit within the specified limits.

A note of transparency: I am currently heading up growth for the company behind this product, but that’s because I truly love it.

Brief clusters thousands of news stories per day into singular topics then summarizes them and delivers the key points you need to know. If you are generating content for blogs or news sections, this is an amazing utility.

Not only does it offer the key points, but you can view each source’s summary, clearly see each source URL, and navigate to the source for more information.

Joe Mifsud, co-founder of Brief, on the service:


Below is my current prompting approach for generating solid research documents from Anthropic Claude 3, which you can use with any model:

Take the following HTML and data from various web pages with research about [TOPIC] and create a detailed, bullet pointed research document focused on the following categories:


  • List all [DATA FOR CATEGORY 1].


  • List all [DATA FOR CATEGORY 1].


Do not add any information not found in the research.

Human Interaction

The most important aspect of a cyborg content approach is that humans should be guiding the research data collection process. Letting a system run on its own is bound to retrieve bad data.

As the system retrieves the data, humans need to interact with the resulting research pages to add a quality assurance level. LLM models will hallucinate and go off task, no matter how solid the prompting or data is.

Rogue information in the data can also lead to issues. This is why human interaction should point the research creation process in the right direction and then check that it delivered the right results.

Download the free guide: Why Should You Have a Style Guide?

Developing Your Style Guides

Style guides are a key component of transferring knowledge about the company, voice, and purpose of the content to a writer. It becomes critical when working with generative AI. That being said, you will need to take a closer look at the following:

Human Foundation

We understand how critical a style guide is to create high-quality content with consistency. However, this is one step that you should not be leaving to LLMs.

There is utility in LLMs to help you expedite the process of style guide creation, but the overall task needs to be human controlled. The nuances with style guide development all expose the limitations of LLMs.

This is where you need to dial in concepts like emotion, human connection, and word choice and tone.

Custom GPT Use

One tool that is useful in crafting your style guides is a custom GPT. You can load documents you already have defining style and tone into the knowledge base.

You can also load previous example content, edit requests, feedback, and other important items to help you parse through that information more quickly.

Once you have a solid style guide, you can load that document and test it by uploading raw content. This is a great way to test how human writers will interact with the style guide.

If the LLM can nail down the use of the style guide, then you’ve hit a level where humans will be able to digest it easily as well.

Creating Your Briefs

There are a lot of great writer’s brief creation tools out there, and with the rise of LLM APIs, the options are only growing. The issue I find with most of these is that they seem to focus inherently on only one type of writing: SEO-focused content.

Your content needs are most likely to flow beyond these limitations. Briefs can include something as simple as an outline (which we’ll cover in-depth), but you should also be looking to include other components like research, internal links, source links, SEO, demographic information, and a variety of datasets.

Building the Outlines

The creation of an outline should be the process of taking your research and getting it to where a writer can easily follow the info and know how to approach turning it into written content.

LLMs are well equipped to utilize structured outlines and turn them into usable copy. This is especially true for data-rich content but becomes less useful the more the content requires human expertise and insight.

Sample Prompt

As we did before, take a look at the following prompt to generate a detailed outline based on a style guide:

Generate a detailed outline for [TYPE OF CONTENT], focusing on [DETAILS THAT ARE CRITICAL FOR THE CONTENT]. Structure the outline with headings for each section: [LIST YOUR HEADER STRUCTURE].

Provide concise and informative descriptions under each heading, highlighting differences across trim levels to aid in understanding the vehicle’s offerings comprehensively.

Follow the below style guide and utilize the below research:

#### Objective:

Create structured and detailed outlines for[CONTENT TYPE], focusing on providing comprehensive information on [DETAILS THAT ARE CRITICAL FOR THE CONTENT].

These outlines aim to assist [WHO THE CONTENT IS FOR AND WHAT THE PURPOSE IS].

#### Components of the Outline:

  1. Introductory Paragraph:
    • Brief introduction [EXPLAIN OVERVIEW].
  2. [HEADER].


  1. [HEADER].


  1. [HEADER].



#### Formatting Guidelines:

  • Use bullet points for clarity and ease of reading.
  • Keep descriptions concise yet informative, ideally no more than two sentences per bullet point.
  • Maintain consistent terminology and units of measurement.


Generating the Final Brief

Your final brief should be a document that includes your outline, research, style guide, and any additional data you want to give to the writer. This process sets your human writers up for success, but it also gives you the option to write rough drafts through LLMs to pass to humans later in the process.

As we noted previously, LLMs are a tool. Tools are wielded by humans to craft things. Approaching an LLM as a full content team is going to lead to major issues, especially at scale.

However, ignoring generative AI as a useful tool altogether is just as big of a mistake. AI-based tools can make you and your team more efficient and allow you to focus on maximizing your creativity by taking over the mundane tasks.

And the planning stage is just the beginning of this symbiotic relationship. In future entries in this series, we will discuss further how LLMs can benefit certain types of content creation, quality assurance, and editing. Throughout, we will continue to focus on how to build a cyborg content strategy with a balance between humans and AI that’s necessary to achieve quality results.

Be among the first to receive Part 3 in the Cyborg Content Series when you sign up for the CopyPress Content Marketing Newsletter.

Author Image - Dave Snyder
Dave Snyder

CopyPress writer

More from the author:

Read More About Agency Solutions