Using Google AI Studio, I built an alternative to SEMrush’s keyword research tool for free. The tool is lightweight, fast, and uses real results from Google to generate a response. These are pulled in real time, giving it a major advantage over static keyword databases that can feel behind the curve.

If you are a solo builder, content creator, or small business owner, that matters. Before investing hours into a new article or landing page, you want a quick way to gauge whether a topic has potential. This tool helps do that without the recurring cost of a full SEO suite.

What This Tool Can Do

  • Estimated cost: The build itself is free, though API tokens still apply. In many cases, you can get pretty far using Google’s free tier.
  • Total tokens: Lets you see exactly how many tokens each query uses.
  • SERP analysis: Reviews the top 20 results and estimates domain strength, backlink pressure, and overall competition.
  • Difficulty score: Assigns a keyword difficulty rating on a scale of 1 to 10.
  • Monthly volume: Provides an estimate of search demand.
  • Keyword intent: Helps identify whether the query is informational, transactional, or somewhere in between.
  • Lower-competition ideas: Suggests related terms that may be easier to target.
  • Content strategy: Gives direction on what kind of page or article to create.
  • Cluster pages: Recommends supporting content around the main keyword.

Who Is This For?

This app is for founders, solo developers, bloggers, and business owners looking to cut costs without losing their competitive edge in SEO. If you are already comfortable with ChatGPT, Perplexity, or similar AI tools, you will probably feel right at home inside Google AI Studio.

Unlike some tools that require command line setup, local servers, or extra configuration just to get started, this SEMrush alternative can live right in your browser. That makes it especially useful for people building niche sites, testing SaaS ideas, or trying to grow traffic before committing to expensive software.

If you are just getting traction and want to validate ideas before dumping money into traditional SEO tooling, this is a very practical place to start.

How To Build It

Here’s the prompt I used to create the shell of the application:

Please build a SEMrush Keyword Difficulty and Volume app. Allow users to enter keywords, use live search data and rank them in difficulty on a scale of 1 to 10, difficulty is determined by the top 20 results, their DA, backlinks, and other factors Google deems relevant. Also provide recommendations for cluster pages, collaboration ideas, and content.

Once you launch it, check the output against real-world results from your SEO tool of choice. AI can still hallucinate, especially when estimating things like authority and backlinks. Also keep in mind that Google AI Studio writes in React, so if you plan to host this outside the browser, you should be ready to work with JavaScript, Node, and the rest of the stack.

Tech Stack

Frontend

  • React 18
  • Tailwind
  • Gemini 3 Flash
  • @google/genai SDK for real-time SERP analysis

Backend & Web Server

  • Nginx
  • Node.js

How It Works

The tool accepts your keywords in an input field and gives you a toggle between Fast and Deep mode. Deep mode takes longer, but the output is better. Fast mode is ideal for brainstorming and topic validation when you want answers quickly.

In Deep mode, the app calls the googleSearch tool. Gemini performs a live Google search for your keyword, ingests the results page, identifies the ranking domains, and analyzes the content within those pages. From there, it forms a better understanding of the competitive landscape.

Fast mode is less involved. Instead of leaning on live SERP analysis, it relies more heavily on Gemini’s own knowledge to estimate the metrics. It is not as thorough, but it is much better for quickly narrowing down ideas before going deeper.

Limitations

The app is useful, but it is not perfect. There are still areas where the model misses the mark, and that is just part of working with AI right now.

  • Domain Authority: This is an AI estimate, not a true Moz or Ahrefs metric. It is more of a directional signal than a precise score.
  • Deep mode speed: Live SERP analysis takes longer, especially compared to a lightweight prompt-only request.
  • Collaboration ideas: These can be generic and may need your own judgment to become useful.

For me, the Domain Authority estimate is the biggest compromise. Gemini has a decent feel for major sites like Wikipedia or X, but it has a much harder time judging smaller niche sites accurately.

Where To Host It

If you are fine using the app in your browser, you can simply leave it inside Google AI Studio. That is the easiest option by far. You do not have to worry about setting up a VM, configuring DNS, adding authentication, or maintaining another service.

Host It Locally

If this is a personal tool and you do not need anyone else accessing it, local hosting is a great option. Download the code from AI Studio, unzip it into your project folder, run npm install, add your Gemini API key to a .env file, and launch it with npm run dev.

Just keep in mind that running it locally can eat up battery life faster than the browser version.

Host It With Google

Google AI Studio includes a publish option that lets you deploy the app. You can connect it to a Google Cloud project and follow the prompts to push it live. Once deployed, it will sit on a unique .run.app domain.

My Recommendations

  • If you are working solo, use the browser-based option.
  • If you are working with a small team, publish it.
  • If you are working solo and handling sensitive info, keep it local.

How This Compares to SEMrush

This is not a full replacement for SEMrush, but it is a very strong alternative for brainstorming, content planning, and keyword validation. Before you spend hours writing a 2,000-word article, you want to know whether there is real interest and whether the competition is realistic. This tool does that part very well.

Where it falls short is in backlinks and authority data. Because those numbers are inferred through Gemini rather than pulled from a large proprietary SEO database, they are not as dependable as the numbers you would get from Moz, Ahrefs, or SEMrush.

Use This AI Tool If

  • You are brainstorming content ideas
  • You run niche sites
  • You are a solo founder or developer
  • You want to validate keywords before writing
  • You want a lightweight research tool
  • You want to save money

Use SEMrush If

  • You run an agency
  • You need backlink data
  • You track rankings over time
  • You run paid ads
  • You manage multiple large websites
  • You need enterprise-level SEO data

The Future of SEO Tools

The future of SEO tools is personalized AI tools, not giant platforms.

AI has progressed to the point where almost anyone with internet access and a modest monthly budget can build tools that handle keyword research, content planning, and validation. Full-scale platforms like Ahrefs and SEMrush still have their place, but they are also expensive and often overkill for independent site owners.

Before committing to another pricey SaaS subscription, it may be worth taking the no-code route with Google Workspace Studio and building a personalized productivity tool that fits the way you actually work.