6 Best AI Research Agents Compared for Make.com Users (2026)

An ai research agent can save you hours, or waste them, depending on which one you pick. This guide compares six of the best options for 2026 (Perplexity, Elicit, Consensus, ChatGPT Deep Research, Gemini Deep Research, and NotebookLM) built for Make.com users who already automate other parts of their workflow and don't want research findings stuck in a chat window.

Most comparisons stop at "which tool is smartest." This one goes further. You'll see real pricing, real strengths and weaknesses, and a working Make.com setup that automatically files every finding into a Google Sheet or Doc the moment your research agent generates it. No manual copy-pasting, no lost findings buried in old chat threads. If you're already comfortable pairing AI automation tools that pair with your Make.com stack, this comparison picks up right where that leaves off.

If you're weighing speed against academic rigor, or a free tier against a real subscription, you'll have a clear answer by the end.

Head-to-Head Comparison

Here's how the six stack up on what actually matters for a research-heavy workflow: speed, citation quality, pricing, and whether Make.com can talk to it directly.

Feature	Perplexity	Elicit	Consensus	ChatGPT (Deep Research)	Gemini Deep Research	NotebookLM
Best for	Fast, cited web research	Academic literature review	Evidence-backed scientific answers	Long, structured research reports	Research inside Google Docs/Sheets	Synthesizing your own uploaded sources
Free tier	Yes (limited Pro searches/day)	Yes (5,000 one-time credits)	Yes (10 analyses/month)	Yes (capped, falls back to a mini model)	Yes (basic Gemini access)	Yes (fully free, standalone)
Entry paid plan	$20/month (Pro)	~$10-12/month (Plus)	$10/month (Pro)	$20/month (Plus)	$19.99/month (Google AI Pro)	Bundled free, or with Google AI Pro
Deep research runs	20/day on Pro	Credit-based, not run-capped	Credit-based, not run-capped	10/month on Plus	Included, expanded on Pro	N/A (source-synthesis, not open web research)
Native Make.com app	Yes, official	No (webhook/API only)	No (webhook/API only)	Yes, official (OpenAI app)	Yes, official (Google Workspace apps)	No official app or public API yet

Pricing figures above were verified at the time of writing. See the "Pricing & Hidden Costs" section below for the full breakdown and disclosure.

Pros & Cons

Perplexity

Elicit

Consensus

ChatGPT (Deep Research)

Gemini Deep Research

NotebookLM

Who Is This For?

This comparison is for Make.com users who already automate other parts of their business and want their research process to work the same way. If you're a solopreneur tracking competitors, a consultant building client reports, or an agency owner who needs findings to land somewhere specific instead of a chat window, this is written for you. For the broader category picture beyond just research agents, our full AI Agents comparison guide is a good next stop.

It also works if you're brand new to AI research agents but already comfortable with Make.com. You don't need a beginner explanation of what automation is. You need to know which tool to plug into your existing stack, and which one will actually connect to it without a workaround.

If your work leans academic (a formal literature review, a PRISMA-style systematic search), skip straight to the Elicit or Consensus sections below. Neither one is built with Make.com in mind first, but both still cover the research side well if that's genuinely your priority over automation.

Deep Dive Into Features

Perplexity's Deep Research mode. Perplexity's Pro plan includes 20 Deep Research runs a day, each pulling from dozens of sources and returning a cited answer in a couple of minutes. Its model picker lets you route a query through GPT, Claude, or Gemini without leaving the app, and its citation density is genuinely higher than most competitors here. The real advantage for a Make.com user isn't the research itself though. It's that Perplexity already has an official app on Make's integrations page, with actions for chat completions, search, and webhook triggers built in.

Elicit's structured extraction. Elicit doesn't just summarize a paper. It pulls specific fields out of it (population, method, outcome) and builds a comparison table across every paper in your search. That's a genuinely different capability than a general chatbot skimming abstracts. For a systematic review, this alone can save days. The tradeoff is that Elicit has no official Make.com app yet, so getting its output into your automated pipeline means routing through a generic webhook or API call instead of a one-click connector.

ChatGPT's Deep Research reports. ChatGPT's Deep Research mode produces the longest, most structured output of any tool in this comparison, often running to several pages with clear sections and a source list at the end. That makes it a strong fit for a client-facing deliverable. The catch is volume. The standard Plus plan caps you at 10 Deep Research runs a month, which is workable for occasional deep dives but tight if research is a daily part of your job. If SEO research is more your speed than general topics, AI SEO tools that connect to Make.com covers that narrower use case in more depth.

Pricing & Hidden Costs

Sticker price only tells half the story here. What actually matters is cost per research run, and that's where these tools split hard.

At roughly 100 deep-research-style queries a month (a realistic pace for someone using this daily), Perplexity Pro's $20/month covers you comfortably since its cap is 600 runs a month (20/day), working out to well under $0.05 per query at that volume. ChatGPT's standard Plus plan caps out at just 10 Deep Research runs a month for the same $20, so hitting 100 runs means jumping to the $100/month Pro tier, a very different cost per query. Consensus and Elicit both use credit systems rather than hard run caps, so their entry tiers ($10/month and roughly $10 to $12/month respectively) can stretch further for lighter, more targeted queries but drain faster on long, complex ones. Gemini's research capability comes bundled inside Google AI Pro at $19.99/month, though Google doesn't publish an exact Deep Research run cap the way Perplexity does. NotebookLM sits outside this comparison entirely since it works from documents you upload rather than the open web, and it's free either way.

One more cost most comparisons skip: the automation layer itself. Make.com's free plan includes 1,000 operations a month, which comfortably covers a research-to-sheet pipeline unless you're running dozens of queries a day. Paid Make.com plans start around $9 to $12/month for 10,000 operations if you outgrow that.

Pricing above was verified at the time of writing. Since pricing changes frequently, please confirm current rates and plan details directly on each provider's official website before making a decision.

Automate AI Research Agent Findings With Make.com

This is the part none of the top-ranking comparisons for this keyword actually show you. Every one of them stops at "here's the best tool." None of them show you what to do with what that tool hands you.

Perplexity has an official, Verified app on Perplexity AI's official Make.com integration page, with real actions for chat completions, search, and webhook triggers already built, confirmed against Perplexity's API documentation. This isn't a workaround. It's a native connection. Elicit and Consensus don't have official Make.com apps yet, so if you want to automate either of those, you're routing through a generic HTTP/webhook module calling their API instead of a one-click connector. ChatGPT and Gemini also have official Make.com apps for their underlying models, though "Deep Research" as a packaged feature isn't itself a distinct Make action. You're calling the same models through their standard completion actions.

Here's the actual setup, using Perplexity since it's the cleanest native path:

• ☐ Perplexity AI account (Pro plan recommended for Deep Research access) — runs the research
• ☐ Make.com account (free plan works for light use) — connects everything
• ☐ Google Sheets or Google Docs account — stores every finding automatically

The workflow itself is three modules:

In plain terms: Perplexity's chat completion action runs your research query and returns a cited answer. Make.com's built-in text parser splits that answer into the fields you want (summary, key sources, date). The final module, Google Sheets' "Add a Row" action, appends that finding as a new row automatically, no copy-pasting, no lost chat threads. If you're already tracking our AI market research tool setup, this pairs naturally alongside it.

Once those findings are landing in a sheet automatically, the next natural step is making sure they actually get acted on instead of sitting there. If you're managing that follow-through inside a task system, your project workflow is worth automating the same way.

Tier gating to flag: Deep Research access on Perplexity requires the $20/month Pro plan; the free tier's limited Pro searches won't reliably trigger the same depth of answer this workflow depends on. If you're on Elicit or Consensus instead, factor in the extra setup time for a generic webhook module since there's no native app shortcut yet.

This is exactly the kind of thorough breakdown of Make.com automation workflows you'll find from SmartSoloFlow: not just which tool wins, but how to make it actually save you time once you've picked it.

Final Verdict & Use-Case Scenarios

Ideal for solopreneurs and consultants automating client research: pick Perplexity, because it's the fastest, has the cleanest citations, and already plugs straight into Make.com with no workaround needed.

Ideal for a formal academic literature review: pick Elicit, because its structured extraction across dozens of papers does work a general chatbot simply can't replicate.

Ideal for quick, evidence-backed answers to a specific claim: pick Consensus, because it strictly sources peer-reviewed papers and shows you the scientific agreement at a glance.

Ideal for long, client-facing deliverables: pick ChatGPT's Deep Research mode, because its reports are the most structured and polished of the group, straight out of the box. This same principle of matching the tool to the deliverable shows up in our AI sales agent workflow, where qualification quality mattered more than raw speed too.

Ideal if your whole workflow already lives in Gmail, Docs, and Drive: pick Gemini, because the research and the destination for it sit in the same ecosystem.

Ideal for synthesizing documents you already have: pick NotebookLM, because it's free, and it stays grounded in your own source set instead of wandering the open web.

Conclusion & Rating

Six real tools, six real jobs. Here's where each one actually lands, based on documented features, pricing, and researched user feedback rather than a single hands-on test:

⭐⭐⭐⭐☆ Perplexity: 4/5
⭐⭐⭐⭐☆ Elicit: 4/5
⭐⭐⭐☆☆ Consensus: 3/5
⭐⭐⭐⭐☆ ChatGPT (Deep Research): 4/5
⭐⭐⭐☆☆ Gemini Deep Research: 3/5
⭐⭐⭐☆☆ NotebookLM: 3/5

For a Make.com user specifically, Perplexity earns the overall recommendation. Not because it's the smartest tool here (Elicit and ChatGPT both out-depth it in their own lanes), but because it's the one that turns research into a working automation without extra plumbing.

Last Updated: September 18, 2026