Good research has always been an essential part of good writing.
Primary sources give your work a factual basis and give your audience the opportunity to validate your perspective for themselves; without those sources, you’re essentially just sharing your opinion. And opinions don’t win you any points for credibility.
(Don’t just take our word for it; source credibility has been a topic of research for decades, as far back as one classic 1951 study that found participants were almost forty percent more likely to believe the exact same factual information when it was presented by a source they perceived as credible.)
Credibility is particularly important on the internet, where it not only helps separate high-quality work from poor imitations but also helps that content rank higher on search engines, which prioritize helpful, reliable, people-first content.
And it may be even more relevant now than ever, with the proliferation of AI-generated content that either doesn’t reference sources at all, or does so poorly.
Providing sources for your work matters. But how do you conduct the kind of in-depth research that can differentiate your blog articles from the competition?
This article breaks down the TMM approach to conducting research—an essential step for producing content that resonates with readers.
Step 1: Prepare Yourself
Our research process doesn’t start with conducting interviews or reading primary sources; it starts with stepping back and getting prepared.
Don’t dive straight into research without a plan. With so much content out there, you’re liable to spend more time clicking through subpar results than gaining valuable insights from high-quality resources.
Instead, focus first on preparing and getting into the right mindset before starting the research phase of your article. This will likely improve the quality of your research and can help you execute it more efficiently. The more you do this, the better you’ll understand how specific mindsets may serve certain content, and you should be able to find and enter a relevant mindset quickly.
Budget Enough Time
To begin, make sure you give yourself enough time to perform your research in the first place. At TELL ME MORE®, we allocate about half the time spent on any given article to research and outlining.
Research shouldn’t be a stage you rush through to meet your deadlines; it’s an essential component of the helpful, person-oriented content that internet-users prefer and search engines prioritize.
If this means you need to extend your article turnaround time or make other changes to your process, do so—your clients will notice and appreciate the effect this has on your articles.
Study the Big Picture
Then, before you do anything else, make sure you understand your article’s objectives and how they serve the overall message. This goes beyond the topic you’re writing about, and instead involves examining the “why” of your piece.
Why is someone searching for information on this topic? Why are they choosing to click on your article when it appears in their results? Explore the big-picture psychological motivations behind your target audience’s intent. Are they trying to:
- Overcome some obstacle?
- Contribute to their self-image?
- Learn something new?
- Advance their careers?
Answering these questions can help you understand the kind of information that will be important to visitors for each article you work on, and how you might present that information.
Imagine we were writing an article about scholarship opportunities for a client in the education sector. The topic of the article is clear enough, but what happens when we step back and ask why someone would want to read the article?
Presumably, this article is primarily relevant for people who are thinking about going to school, but aren’t sure if they can afford it. So their actual goal is to remove an obstacle standing between them and their education.
This perspective will help us gather (and present) information that will be specifically relevant to what our audience wants—not just to the words they’re searching for.
As you repeat this process from one article to the next, you’ll probably notice that there are a handful of goals that keep coming up. Keep a list of these concepts; it will streamline this part of the process in the future.
Establish Clear Context
After you’ve clarified the big picture significance of your article, get online and do a little preliminary exploration to develop some context for your topic.
This includes doing keyword research to hone your article concept and make sure that the piece you eventually produce is relevant to what people are searching for. Tools like Semrush or Ahrefs are great for this.
Then, use Google and AI tools like ChatGPT, Claude, or Gemini to quickly get the “lay of the land” on your topic. This isn’t “research” so much as orientation, with the goal of ensuring that you have a sound grasp of the subject matter, and that there won’t be any holes in your treatment of the topic.
As part of this exploratory phase for our scholarships article, we might ask ChatGPT about the factors that might stop someone from applying to college. Getting this context could be useful for creating a more well-rounded article. Based on the results in the image below, we might consider tying in information regarding work-study opportunities, support from the financial aid office, or the number of students who receive scholarships.
From there, we could keep probing ChatGPT (or Google) for more information on the basics of all the topics we mean to cover in our piece.
For some content creators, this might represent the extent of their “research” phase. Don’t let this be you. If you stop here, you’ll only wind up producing carbon copies of existing content, which will often struggle to gain traffic because they don’t stand out to searchers as uniquely valuable; your work may even be factually incorrect, which can also hurt your rankings.
Instead, it’s important to recognize the real purpose of this preliminary research—to get ideas for how you can go beyond what others have already done.
As it exists right now, pure AI content has sharp limitations when it comes to factual accuracy, both in terms of interpreting information and authenticating sources. This is why it’s still so important for humans to do research and validate sources. AI-generated pieces are perfectly content citing other unverified AI-generated pieces; we aren’t.
Step 2: Go Straight to the Source
Once you have the scope and purpose of your article locked down, you’re ready to start your actual, in-depth research.
This means digging into primary sources for genuine insights to support your content and contribute to its trustworthiness.
Text Sources
You’ll likely want to start with text sources, since they’re generally the most accessible.
Start by searching for case studies, industry reports, scientific papers, and government statistics to gather reliable industry insights. All of these sources can be gold mines for data, trends, and other invaluable perspectives backed up by real-world proof.
For our example article, a resource like the National Center for Education Statistics is a must-visit; dedicated sources like this often contain a wealth of reliable information, as you can see in the image below.
As you start wading through information, make sure that any content you pull comes from a reliable source. Industry players often produce their own research, which may or may not be valid. When in doubt, check your findings against additional sources to ensure you steer clear of bias.
Case Study: Transforming Data
When incorporating data into your articles, look for opportunities to go beyond simply repeating findings from the original source. From performing basic calculations or creating graphic visualizations to conducting more advanced statistical analysis, when you transform data you create unique insights that readers aren’t able to get anywhere else.
One piece we delivered for a client in the higher-education sector involved gathering data from several different governmental and academic sources. Rather than simply presenting the data – which would have been interesting and informative, but not necessarily insightful – we instead explored our findings for opportunities to expand on them.
We wound up broadening our scope to conduct additional research, weave together disparate data sets to produce novel metrics, adjust some statistics to compensate for variations in the raw data, and combine the results into a weighted ranking. This enabled us to make novel observations that simply wouldn’t have been possible without transforming the data – and resulted in a valuable article for the client.
For some articles, ChatGPT’s Deep Research feature can be particularly useful for developing a clearly sourced, comprehensive report on a topic in a single pass. But this tool can be a double-edged sword. These Deep Research reports can be great in that they can quickly surface relevant primary research, and can make sophisticated connections between key kernels of information and the broader topic of the piece.
But there’s a flipside. For one thing, these reports are only as good as their inputs; while the platform’s reasoning logic does attempt to prioritize high-quality sources, unreliable material often filters through and can easily derail an entire report. Also, there’s a risk that your thinking may become anchored to these reports, since the finished product reads as “complete” and “polished.” This can lead you to miss important information that didn’t make it into the report, misinterpret the information that did make it in—and miss an opportunity to deliver a high-quality article for your client.
Your best bet? Be realistic—like all AI resources, AI research tools are no panacea. They excel at surfacing research that might be relevant to your questions, but it still falls to you to investigate the results to ensure they are trustworthy and pertinent.
Confirmation bias is always a concern with [AI research tools]; ask for scholarly research indicating that cigarettes are good for you, and you’ll get just that. At this point it still takes a human to ask the right questions and then go deeper, doing things like investigating the authors of a paper—even one published in a scientific journal—to discover potential conflicts of interest that could introduce bias.
No matter what kind of sources you’re dealing with, always be sure to confirm their reliability before using them in your work.
Is this a reliable source?
Always verify the source’s reliability by performing the CRAAP test (Currency, Relevance, Authority, Accuracy, and Purpose). Ask yourself:
- Is this source current? How recently was it published? Is it possible the information has become outdated since then? (This is less of a concern for scholarly research, although you should still try to validate older claims; but for industry reports, currency is a must.)
- Is this source relevant? Is it actually responsive to the topic of your article, or is the connection dubious?
- Was the source produced by a real authority? Is the author credible? Are they trustworthy? Do they have the standing to make the claims contained in their work?
- Is the source accurate? Is the information contained in it true? (Check facts against other sources whenever possible, to be sure.)
- What is the purpose of the source? Why was it published? Is there any ulterior motive (like pushing a product, service, or agenda) embedded in the content?
Also, always make sure you’re getting information from the primary source. Don’t settle for citations in secondary sources, or unsourced claims of any kind. Track down the original citation for any data point you want to include—if you can’t find that, don’t use it.
Lastly, try to link to full text and/or unpaywalled content whenever possible. A link to an article abstract is better than no link at all, but giving readers the opportunity to see the source for themselves is always best.
SME Interviews
In addition to gathering insights from already published works, you can (and, whenever possible, should) make an effort to obtain original perspectives by conducting interviews with subject matter experts (SMEs). This enables you to give your readers insights they couldn’t possibly obtain anywhere else, which boosts your content’s all-important E-E-A-T—Google’s framework for evaluating the quality of content.
Start by asking your client for SME contacts—they’ll likely be able to connect you directly to relevant individuals either within or associated with their organization who are positioned to offer unique input that’s tailored to your client’s needs.
For our scholarships article, reaching out to our client to get connected with someone in the financial aid office would be a natural choice.
You may also want to consider cultivating your own network of experts relevant to the field or fields you often work in. This can be advantageous in that you’re not bound by or dependent on individual clients to gather material, meaning you may be able to reach farther to gain a wider breadth of insight.
Original Research
SME interviews are great if you’re looking to present specific domain expertise or share personal stories—which do help to boost your content’s credibility. But at the same time, interview quotes can only present one person’s (expert) opinion.
Depending on the type of article you’re writing, you might also consider conducting your own original survey research as a means of drawing insights from the perspectives or habits of a wider group.
This can be a great strategy if you notice that there’s a lack of recent, compelling, and/or trustworthy data on the topic you’re writing about. Often, you’ll see sites parroting the same questionable statistics simply because they’re available; this could be a great opportunity for you to differentiate yourself by presenting novel research that readers won’t be able to find anywhere else.
This is also great for collecting data on exactly the subject you want. For our scholarships article, for instance, we might decide to survey students and alumni of our client’s institution to gather data on their experiences with financial aid—producing a body of knowledge that isn’t just timely and reliable, but also highly specific to our target audience.
While there are various methods you might use to solicit input from audiences—like focus groups, online panels, or customer feedback programs—surveys are the most reliable way to collect data from a wide respondent pool. Surveys can be designed to capture both qualitative and quantitative data from your survey population.
Tools like SurveyMonkey or Google Forms make it easy to solicit and collect responses. Keep your data collection process organized and systematic to maintain integrity and reliability, and to make it easier to analyze your results and uncover patterns, trends, and insights to inform your writing.
Step 3: Put Your Research into Action
Doing the research is only half the story—it’s equally important to make sure you’re using it well in order to get the most benefit out of the effort you’ve put in.
Build a Knowledge Base
What happens to the research you perform for each article you work on? Do you ever revisit your findings, or are they out-of-mind once you submit a piece for publication?
If it’s the latter, then you’re wasting much of the effort that goes into conducting research in the first place.
Instead of thinking of research as a means to an end—that is, a means of producing individual articles—step back and think of the broader context of your work. If you frequently produce content in one or a few niches, then your work is really an opportunity for you to become a kind of SME in those spaces.
Your research, then, is not merely a task you perform with each new article, but is an ongoing process through which you can genuinely develop a unique perspective on whatever it is you write about.
Sure, AI makes some things easier. But when something is easy, the differentiator becomes, who is going to do the hard work that others aren’t doing? When it’s this easy to create content, there will inevitably be consequences; losing trust is a big one with AI. That’s what we have to work to build and maintain.
As you conduct your research, consider building a knowledge library that encompasses both textual sources and interview transcripts. You can use this library to track and organize your research. Not only will this enable you to create individual well-sourced articles more quickly and efficiently, it will also make your content more authoritative as you develop unique perspectives through this sustained effort.
Using a shared Google Drive can be a simple way to build this knowledge base and make resources accessible across your marketing team. While a well-organized information library has value on its own, tools like ChatGPT and Claude can easily integrate with Drive (and other platforms), offering a simple way to draw on your team’s knowledge while using AI chat.
You might take your experiments with AI resources even further, to better leverage your organizational knowledge and get more value from the work you’ve already done. Perplexity is a promising tool here. For example, you might compile your content into a structured format (like a well-formatted text document or spreadsheet) and provide those resources to Perplexity in a dedicated Space. Perplexity’s powerful search abilities can then allow you to derive highly nuanced outputs from your existing knowledge base, making your team’s work work for you.
Case Study: Building Expertise
Your ability to generate novel insights will grow as you continue to build up your knowledge base – particularly if you’re also transforming data with original analysis, since this practice can enable you to combine data-backed observations and cultivate an increasingly unique (and uniquely trustworthy) perspective.
One article we produced for a higher-education client involved analyzing data on consumer behavior in a category relevant to one of the client’s specializations. Among other things, this analysis enabled us to draw some fundamental conclusions about consumer purchasing habits. These findings shaped the argument of the article, which performed well when it was published.
Later, while conducting research as part of another campaign for the same client, we surfaced additional information that was closely related to our earlier analysis. We were able to apply our earlier findings to our new research in a way that seriously elevated the new campaign.
We could not have derived that insight from our research on the second campaign; neither could we have foreseen the later relevance of that insight when working on the first article. It was only by performing the analysis in the first place, then storing it in our knowledge base, that we were able to enhance our subsequent content.
Create an Outline
Once you’ve pulled together a body of research, don’t just dive into the writing; use that research to develop your thesis, rather than as a means of simply slapping citations onto preconceived ideas.
Use an outline to help organize your research and provide a structure for your thoughts before jumping into the writing phase. This is great for ensuring that your topic is fully researched and cogent before you start hammering out the text, and it can give you a head start if you need to follow up with SMEs for clarification or elaboration with a deadline looming.
ChatGPT and Claude are excellent resources here, capable of instantaneously pulling your research and notes into a coherent, well-structured outline. You probably won’t wind up using the exact structure they provide you—but turning to these tools to create an outline can be a highly valuable starting point.
Outlines are also great for maintaining alignment with editors and clients before you get too far into the writing. Just because you’ve done great research doesn’t mean your work will always conform to other stakeholders’ expectations; an outline helps you avoid any problems in this department.
Once your team has signed off on your outline, all that’s left is to write the article—which, thanks to your thorough research and well-designed outline, should be a walk in the park.
In-Depth Research Builds Credibility—and Earns Trust
In-depth research is key for creating trustworthy, authoritative content, and should be part of any mature content production process. But the real secret ingredient isn’t any one of the research methods we just covered—instead, it’s the mindset behind your research.
Stay focused on your audience’s goal. Keeping this perspective will bring structure to your research and help ensure you produce work that resonates with what your audience actually wants. Do this, and you’ll produce content that isn’t just well-researched but will be well-received, too—building up your domain authority and giving readers a compelling reason to visit your site again and again.