- Why most people skip this step (and pay for it later)
- Start with what people are already typing (Google Search Console)
- Google Trends: the free tool people use wrong
- Reddit and Quora: unfiltered, uncomfortable, and free
- YouTube comments: the most underused free research tool there is
- Your own platform analytics (free, and most people barely open them)
- Hashtags and search terms as audience research, not just reach tactics
- Look at who's already following you (and who isn't)
- Polls and direct questions: the free tool that beats all the others
- Where free content itself becomes free research
- The B2B version of this problem
- A five-step free audit you can do this week
- Frequently asked questions
The short version: Google Search Console, Google Trends, Reddit search, YouTube comments, and your own platform's native analytics will tell you more about what your audience wants than any paid tool under 500 pounds a month. The trick isn't finding the tools, it's reading what they show you instead of skimming for a number that flatters your plan.
Why most people skip this step (and pay for it later)
I spent the best part of a decade building content calendars around what I thought my audience wanted. Guesswork dressed up as strategy. It's the fastest way to burn six months on a content plan that gets 40 views a post.
Here's the bit nobody likes admitting: the free tools I'm about to list are not the hard part. Reading them is. I've watched clients pull up Google Search Console, see 200 people typed "AI consultant cost UK" into Google, and then go and write a post about "10 AI trends for 2026" anyway, because that's the post they wanted to write. The tools give you the truth. Most people ignore it because it's less exciting than their own idea.
Start with what people are already typing (Google Search Console)
This is the tool I open before anything else, and it's free the moment you verify your website. Not a "freemium" free, not a "free trial" free., permanently free.
Go to the Performance report and filter by "queries." You'll see the exact phrases people typed into Google before landing on your site, or before your site showed up and got ignored. That second group matters more than people think.
Real example from my own site: I once found 340 impressions in a month for a phrase I'd never targeted, "how much does an AI consultant cost." I didn't have a single page answering it directly. I wrote one. It's now one of my best-performing pages for that exact question, and it's why I now point people toward a proper breakdown on how much AI consultants charge whenever the question comes up in client calls.
Step by step:
- Search Console, then Performance, then Search results
- Sort by impressions (not clicks) so you catch demand you haven't served yet
- Look for queries where you rank position 11 to 30, that's Google telling you the content exists but isn't good enough yet
- Group similar phrases into themes, that's your next month of content sorted
Google Trends: the free tool people use wrong
Google Trends is free and most people use it to check if a topic is "trending," then stop. That's the least useful thing it does.
The useful bit is the "Related queries" and "Related topics" panel at the bottom of any search. Type in your core topic, scroll down, and you'll see what else people searching that term are also searching. It's a two-minute way to find angles you hadn't considered.
Also worth doing: compare two or three terms against each other. When I compared "social media marketing" against "AI marketing" over the past two years, the crossover point was clear, mid-2024. That single chart changed how I pitched myself to clients for the next twelve months.
Reddit and Quora: unfiltered, uncomfortable, and free
If you want to know what your audience complains about, not what they say on LinkedIn where everyone's performing competence, go to Reddit. Search your topic plus "reddit" in Google (Reddit's own search is poor) and read the top threads. People are ruder, more specific, and more honest there than anywhere else.
I found an entire subreddit thread complaining that AI consultants "just sell ChatGPT training and call it a strategy." Uncomfortable to read. Also exactly the objection I now address in the first three minutes of every discovery call, because ignoring it wouldn't have made it less true.
Quora works the same way but skews more B2B and more "explain this to me like I'm new." Good for finding beginner-level content gaps most competitors skip because they assume everyone already knows the basics.
YouTube comments: the most underused free research tool there is
Nobody talks about this one enough. Pick three or four videos in your niche with decent view counts, and read the comments, not the top ones (those are often just praise), sort by "Newest first" instead. That's where people ask the questions the video didn't answer.
I did this for a client in the coaching space. Watched four videos, read maybe 200 comments, and found the same unanswered question eleven times: "how do I know when I'm ready to raise my prices." Nobody in that niche had built a proper piece of content around it. We did. It became their best lead magnet.
Your own platform analytics (free, and most people barely open them)
This is the one that annoys me most, because it costs nothing and people still skip it. Every platform gives you free audience data:
- Instagram and Facebook: native Insights show you audience age, location, active hours, and which posts drove saves versus likes (saves matter more, that's intent to come back)
- LinkedIn: your Analytics tab shows follower demographics by job title and industry, free, no export needed for basic reading
- YouTube Studio: Audience tab shows exactly when your viewers are online, down to the hour
- TikTok: the Creative Center is free and shows trending sounds, hashtags, and content by industry, even if you've never posted a single video there
Since Facebook shut down its standalone Audience Insights tool a few years back, a lot of marketers assumed the free option disappeared with it. It didn't, it just moved into each platform's native business suite. If you want the fuller comparison of what replaced it and what's worth your time, I've laid out the current options in this rundown of free and paid social audience insights tools, because the landscape shifts often enough that it needs its own space.
Hashtags and search terms as audience research, not just reach tactics
Most people treat hashtags as a reach play. They're also a free research tool, because the volume and pairing of hashtags tells you what your audience groups your topic with in their own head, not in yours.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
If you're working in Reels specifically, I put together a free approach to finding the right hashtags for Facebook Reels that doubles as a way to see which subtopics are getting engagement versus which just have volume. Volume without engagement is a warning sign your "popular" topic is saturated.
Look at who's already following you (and who isn't)
This sounds obvious and it's still skipped constantly. Before you go hunting for new audience data, audit the audience you've already got. On Instagram specifically, there's a free way to see exactly who follows you but you don't follow back, and who follows you that you do, which tells you who's interested versus who followed once and vanished.
I've written up the free method for this in detail, including how to see who doesn't follow you back on Instagram, and the pattern I keep finding with clients is that 15 to 25 percent of their "followers" have gone completely silent, no likes, no views, nothing in six months. That's not audience insight, that's audience noise, and separating the two changes what your engagement numbers mean.
Polls and direct questions: the free tool that beats all the others
Here's the uncomfortable truth I mentioned earlier, in full. All of the tools above give you inferred interest. What people search, what they comment, who they follow. None of it is them telling you directly. The single most accurate free tool for understanding your audience is asking them a direct question and reading the answers, not just counting them.
I ran a LinkedIn poll to roughly 12,000 followers asking what stopped them hiring outside help for AI in their business. Four options, one of which was "cost." Cost won, at 41 percent. But the comments underneath told a different story, most people weren't worried about the fee, they were worried about hiring the wrong person and wasting it. That distinction came from reading 60-odd comments, not the poll bar chart. If I'd stopped at the numbers I'd have built a pricing page. Instead I built trust content first, and it's part of why I now talk clients through what happens before they commit, which I've written out plainly on my page about what working with an AI consultant for a small business looks like.
Polls are free on LinkedIn, Instagram Stories, X, and Facebook. Use them:
- Ask a binary or four-option question, never open-ended, people won't vote on open-ended
- Follow up in the comments or a reply asking "why" to whoever votes
- Run the same question three times over three months, interests shift faster than most content calendars account for
Where free content itself becomes free research
If you're building content anyway, the process of publishing it for free is also a research tool. Every free platform you publish to gives you a small, honest signal back, what got read, what got shared, what got nothing.
If you're early on and still deciding where to even put content before you've got budget for anything paid, I've mapped out where you can start blogging for free today and what that costs you down the line once you outgrow it, because the free tier you start on shapes the audience data you're able to collect from day one.
The B2B version of this problem
Everything above skews consumer and social. If you're selling B2B, the audience interest signal often lives in different places entirely, LinkedIn Sales Navigator's free tier, company job postings (a company hiring for "AI ethics officer" is telling you something about its priorities), and the questions prospects ask on discovery calls, which almost nobody writes down and reuses.
I talked through this in a webinar on digital selling, where the whole point was that most B2B sales teams already have the audience research sitting in their CRM notes and just never look at it as research. If you want the full transcript, it's here: the Simplilearn digital selling webinar breakdown. And if outbound is part of how you're finding these prospects in the first place, the sheer range of free and low-cost options is worth seeing, laid out in this list of over 300 sales prospecting tools.
A five-step free audit you can do this week
- Day one: pull your top 20 queries by impressions from Search Console, group into three themes
- Day two: check Google Trends related queries for each theme
- Day three: read 100 comments on three competitor YouTube videos or Reddit threads in your niche
- Day four: post one direct poll on your most active platform, follow up on comments
- Day five: pull your native platform analytics, compare who's engaging against who's just following
That's five days, no spend, and by the end you'll have more specific audience data than most businesses collect in a year of guessing.
Frequently asked questions
Is there still a free version of Facebook Audience Insights?
The standalone Facebook Audience Insights tool was retired in 2021. The audience data now lives inside Meta Business Suite and each Page's native Insights tab, still free, just moved rather than removed.
What's the single best free tool for understanding audience interests?
Google Search Console, because it shows you exact phrases real people typed before finding (or nearly finding) your content, which is closer to genuine demand than any inferred signal from likes or follows.
How often should I redo this kind of audience research?
Every three months at minimum. I've seen client audiences shift their top concern within a single quarter, particularly around anything AI-related, where the conversation moves faster than most content calendars are built to handle.
Do I need to combine free tools with paid ones eventually?
Only once you've outgrown what free data can tell you, usually once you're managing multiple channels or need historical trend data beyond what native analytics store. Most small businesses can run on free tools alone for a year or more before that becomes a real constraint.
