Topical Map AI
Topical Map AI (topicalmap.ai) builds a topical map, a plan of related articles that together cover one subject, for search engine optimization (SEO), the work of making web pages appear high in Google's results. The user types a topic, for example "home automation". The tool asks AI models from OpenAI and Anthropic for hundreds of keywords, the words and questions that people type into a search engine, and sorts them into clusters, groups of keywords that one article or one section of a site can answer. SEO workers believe that Google ranks a site higher when it covers a subject completely, and they call this topical authority. The company's page in 2025 said that an SEO worker needs more than 10 hours for this work and the tool needs 60 seconds.
Pete McPherson ran the blog, podcast and YouTube channel Do You Even Blog. After ChatGPT appeared, he moved all his work to AI in early 2023 and started to build apps by having AI write the code, with no experience as a developer. He built Topical Map AI in about three days in March 2024 to solve his own problem and first sold it as a US$49 one-time purchase. From April to November 2024 he took partners who promoted it to their SEO audiences, and he wrote that revenue grew to US$7,000 to US$8,000 a month. In a video of 28 December 2024 he said that it earned US$6,000 to US$7,000 a month, as reported by eBiz Facts. He then left it alone for about ten months and sold it at a price he has not published; his article of 3 October 2025 reports the sale. The buyer, Megan Ragab, writes on the site that she bought it in 2025 and runs it today.
Every money figure on this page is the founder's own statement, grade C, and one of them is known only from a third party's summary of his video. No payment processor or buyer has confirmed a number, and neither the partners' share of the revenue nor the sale price is public. The page also shows what happened to earlier businesses that depended on traffic from a platform they did not control, or that sold a content tool against larger competitors.
The numbers, graded
Each row shows the evidence grade and the source numbers from the list at the end of this page. The grades are explained in How Aiolum verifies.
| Fact | Value | Grade | How we know | Sources |
|---|---|---|---|---|
| Revenue per month, December 2024 | US$6,000 to US$7,000 | C | Founder's video of 28 December 2024, in the summary by eBiz Facts of 20 January 2025: "TopicalMap.ai, which now earns $6-7K per month". The speech in the video could not be transcribed for this page. | [2, 3] |
| Revenue per month, April to November 2024 | grew to US$7,000 to US$8,000 | C | Founder's own article of 3 October 2025, after he took partners who promoted the product. He calls it mostly passive for him. | [1] |
| Run rate from the December 2024 range | about US$72,000 to US$84,000 a year | calc | Calculated: the range times 12. The outcome value on this page is the low end, US$72,000. | [2] |
| Sale | sold at a price that has not been published | C | Founder's article of 3 October 2025: "Sold Topical Map AI after 10 months of ignoring it". His projects page: "Built, launched, and successfully sold the product." The buyer writes on the home page that she bought it in 2025. | [1, 10, 8] |
| Revenue at the time of the sale | not published | C | The latest figure is from December 2024, about ten months before the sale was reported. | [1, 2] |
| Earnings from the two companies he sold | more than US$350,000 | C | Founder's article. The figure covers Topical Map AI and FABB AI together and does not separate revenue from sale proceeds. FABB AI was listed for US$45,000 in December 2024 and sold within 10 days. | [1] |
| Portfolio revenue in 2024 | about US$100,000 from several AI-coded apps | C | Title of the founder's video of 28 December 2024, and eBiz Facts: "$100,000+ Revenue in 1 Year". This is not Topical Map AI alone. | [2, 3] |
| Time to build the first version | about 3 days, March 2024 | C | Founder's article. He built it to solve his own problem with keyword research and content planning. | [1] |
| Launch offer, April 2024 | US$49 one-time for 100 topical maps | C | Archived home page of 13 April 2024, shown as US$50 off a price of US$99, for the launch only. | [6] |
| Plans in 2025 | US$29, US$59 and US$99 a month | C | Archived home page of 20 January 2025: 30, 100 and 250 credits a month, or US$290, US$590 and US$990 a year. | [7] |
| AI models in the product, 2025 | "the latest OpenAI & Anthropic" models | C | Archived home page of 20 January 2025. The founder's projects page says that he integrated OpenAI. | [7, 10] |
| People named as the builders, January 2025 | 4 | C | The founder, the SEO trainer Charles Floate and two people listed with Merged.ca. Charles Floate called the founder "my business partner" in a video of 2 September 2024. | [7, 4] |
| Legal form, April 2024 | a sole proprietorship owned solely by Pete McPherson | C | Archived terms of 13 April 2024, under the law of Michigan. The terms today name the buyer as the sole owner. | [5, 9] |
| Outside funding | none reported | C | A sole proprietorship in April 2024. The buyer's page today shows "$0 VC funding (proudly bootstrapped)". | [5, 8] |
| Use by October 2026 | more than 4,000 accounts and 12,000 maps | C | Buyer's home page, counted since the launch. | [8] |
| Time from launch to the reported sale | 19 months | calc | Calculated: March 2024 to October 2025. | [1] |
How to read this table: a grade C value is a statement by the founder, his partner or the company's own pages, or a third party's summary of the founder's words. A "calc" value was computed on this page from the rows above it. No value is grade A or B.
| Plan | Per month, billed yearly (US$) | Per month, billed monthly (US$) | Limit |
|---|---|---|---|
| Basic | 56 | 25 maps a month, 20 on yearly billing | |
| Pro | 104 | 46 maps a month, 5 maps at once | |
| Enterprise | 179 | 80 maps a month, 3 seats, API with 100 requests a day | |
| Agency | 374 | 139 maps a month, 5 seats, white-label exports, API with 200 requests a day |
Prices are per month with monthly billing. The page has a switch for yearly billing that says "SAVE 17%", but the yearly prices load by script and could not be read; on yearly billing the Basic plan gives 20 maps a month instead of 25. There is a 30-day refund if fewer than 10 credits were used, and users can add their own OpenAI key to cut costs. Under the founder the prices were lower: a US$49 one-time offer for 100 maps in April 2024, then US$29, US$59 and US$99 a month for 30, 100 and 250 credits in 2025, or US$290, US$590 and US$990 a year.
The timeline
From a blogger's switch to AI in 2023 to a tool built in three days and sold about 19 months after its launch. Each entry carries its grade.
- January 2023CPete McPherson, who runs the blog, podcast and YouTube channel Do You Even Blog, decides after the release of ChatGPT to move all his work to AI. [1]
- March 2023CHe sees a YouTube video in which ChatGPT writes code and starts to build his first app, with no experience as a developer. [1]
- August 2023CHe tries again: an app that takes three weeks to build grows at once to US$1,000 a month in recurring revenue. [1]
- October 2023CFABB AI, a Fully Autonomous Blog Bot, makes about US$3,000 a month. A health emergency makes him pause all his work. [1]
- March 2024CHe builds Topical Map AI in about three days to solve his own problem, makes "a few grand" and then leaves it to run. [1]
- 13 April 2024CThe first archived copy of the site: "1,000 keywords in 1 click", US$49 one-time for 100 topical maps. The terms name him as the sole owner. [5, 6]
- April to November 2024CHe takes partners who promote the product. Revenue grows to US$7,000 to US$8,000 a month. [1]
- 2 September 2024CThe SEO trainer Charles Floate publishes an interview with him, "Meet The Dev - Topical Map AI", and calls him "my business partner". [4]
- December 2024CHe lists FABB AI for sale at US$45,000, and it sells within 10 days. [1]
- 28 December 2024CHis video "How I made $100k last year from AI-coded apps". eBiz Facts reports from it that Topical Map AI earns US$6,000 to US$7,000 a month. [2, 3]
- 20 January 2025CThe home page names four builders and sells plans at US$29, US$59 and US$99 a month, with OpenAI and Anthropic models. [7]
- 3 October 2025CHis guest article on Making Sense of Cents reports that he sold Topical Map AI after ten months of ignoring it. [1]
- October 2026CThe buyer, Megan Ragab, runs a rebuilt version at US$56 to US$374 a month, with more than 4,000 accounts created and a 40% recurring commission for affiliates. [8, 18]
Why it worked
Four causes, in the order they mattered. Each one is a reusable lesson.
Distribution through partners
The founder says he hates marketing. He took partners who already had the audience he needed, SEO people with their own followers, and revenue grew to US$7,000 to US$8,000 a month about eight months after the launch. He describes his usual offer to a partner like this: he builds the app, gives the support and runs everything, about 97% of the work; the partner promotes it; they split the monthly profits and the sale 50/50.
His own problem, built in days
He started from the part of blogging that he hated, keyword research and content planning, and built the first version in about three days with AI writing the code. A product this small costs little time to test. His rule is to decide in the third month whether to stop a project or keep it.
Charge from the first version
The first offer was a US$49 one-time deal for 100 maps. By 2025 it was a subscription with a fixed number of credits a month, from US$29 to US$99, which limits how much model use each customer can cause.
Built to sell
Before each project he asks, "Can I sell this in 12 months?" or "Can I make this 90% passive?" He sold FABB AI ten days after he listed it, and sold Topical Map AI after ten months without work on it. He writes that buyers look for stable revenue, an easy transition and an interesting brand, and that he would have earned more if he had not paused both companies.
How it was built
These details come from the founder's article, the page of his December 2024 video, his projects page and archived copies of the product's site. They are grade C. [1, 3, 10, 7, 8]
- Built by a founder with no experience as a developer, by having AI write the code, which he calls vibe-coding. He recommends the AI code editor Cursor, or Windsurf.
- Stack: SvelteKit for the web app, Firebase as the database and backend, Stripe for payments. The description of his December 2024 video, which lists topicalmap.ai among his builds, adds Resend for emails, Zapier for automation and Cursor.
- His projects page lists OpenAI, SvelteKit and Firebase for Topical Map AI. The 2025 site offered "the latest OpenAI & Anthropic" models.
- Each plan had a monthly number of credits: 30, 100 or 250 in 2025. Paying customers also got a private Discord community.
- Customers came through partners with SEO audiences. By his account his own Do You Even Blog audience is a large asset for all his products. He was disappointed by every Product Hunt launch that he tried.
- The buyer has rebuilt it since 2025, with four map types, exports, team seats, an API and an option to use your own OpenAI key.
Tools named in the sources
Each one opens its page in the Tools tab, with every story that uses it.
The dead that came before
Five businesses in the Loot Drop database lived on traffic or attention that another company controlled, or sold a content tool against larger competitors. The values below are from that database and are not independently verified.
Waterproof Digital Camera
The promiseAn affiliate blog of camera reviews, built with aggressive SEO methods to earn commissions from Amazon through Google traffic.
Why it diedGoogle's algorithm updates removed the search traffic that the site was built on.
Upworthy Insider
The promisePositive stories written to spread through Facebook.
Why it diedFacebook changed the algorithm of its feed, and the traffic fell. There was no second source of readers.
Room 77
The promiseDetailed data on hotel rooms and their prices.
Why it diedLarge travel companies competed with it, and Google reduced its role, according to the database.
Hyperink
The promiseE-books on niche topics, produced quickly for small audiences.
Why it diedSelf-published books and free content online filled the same niches, and Amazon controlled the store.
Pressly
The promiseA simple platform for brands to collect and publish content with less work.
Why it diedCompetitors sold full suites that included the same function, and Pressly did not grow beyond it.
What changed
Each of the five earned its income from attention that another company controlled: Google's search results, Facebook's feed or Amazon's store. When that company changed its ranking or competed directly, the traffic and the income went away. Topical Map AI sells to people who are in that position with Google: bloggers, freelancers and agencies whose sites depend on search traffic. Google's helpful content system, updated on 14 September 2023 and made part of its core ranking in March 2024, is designed to lower content "made primarily to gain search engine traffic", and a Pew Research Center study found that people click a normal result about half as often when Google shows an AI summary.
Two things differ. Topical Map AI is a tool, not a content site, so it did not need its own pages to rank: its customers carried that risk, and its partners brought customers to it directly. And it took about three days to build, with a small stack, so the founder could sell it as soon as he stopped working on it. The buyer now carries the risk that fewer people will pay to plan content for Google while Google answers more questions itself.
Can one person copy it now?
The method, yes. The product, only with a partner who has the audience.
A small tool that removes one tedious step for people who already pay for tools, built in days with AI writing the code, is still easy to copy, and the founder's own account says that the hard part was finding customers, which he solved with partners. The weak point is the market: SEO tools are many, a larger tool such as Surfer already has a topical map feature, and AI summaries in Google now take clicks from the customers' own sites.
Where it is still open: Small tools for other groups whose work is still manual and repeated: newsletter writers, podcast producers and YouTube creators, and tools for visibility in AI assistants instead of in Google.
- 1Pick a task that you do yourself every week and dislike, and build the smallest tool that does it, in days, with an AI code editor.
- 2Charge from the first version: a low one-time price for early users, then monthly plans with a fixed number of credits so that model costs stay inside the price.
- 3Find a partner who already has your customers' attention, and offer a share of the profit and of a later sale for promoting it.
- 4From the first day, keep the product easy to hand over: stable revenue, little support work and a clear brand, which the founder says is what buyers look for.
What to watch
The risks are the mirror image of the causes above.
Self-reported, partly second-hand, and old
Both revenue figures come from the founder, and the December 2024 figure is known only from the summary of his video by eBiz Facts. The latest figure is about ten months older than the reported sale, and he says that he ignored the product in that time, so the revenue at the sale is unknown. The sale price is not public.
The money was shared
The home page named four people as the builders in 2025. The founder describes his usual partner deal as a 50/50 split of the monthly profits and of the sale, and says that he has made it twice, but he does not say what the split was for Topical Map AI. His figure of more than US$350,000 covers two companies together.
Google decides whether the customers' sites get traffic
A topical map is a method to rank in Google. Google's helpful content update of 14 September 2023 aimed at content made mainly to gain search traffic. In March 2024 that system became part of Google's core ranking, and a new policy against scaled content abuse covers many pages made to manipulate rankings, "no matter how it's created". Core updates followed in August, November and December 2024 and in March, June and December 2025. A customer who publishes AI articles from a map in large numbers is in the group that these rules aim at.
AI answers take clicks from websites
Pew Research Center tracked the browsing of 900 U.S. adults in March 2025. When Google showed an AI summary, users clicked a normal result in 8% of visits, against 15% without one, and clicked a link inside the summary in 1% of visits. If websites get fewer clicks, their owners have less reason to pay for a tool that plans content for search.
Larger SEO tools have the same feature
Surfer, a larger SEO content tool, documents its own Topical Map tool in its help center. A small stand-alone tool has to compete with features that customers may already have in a tool they pay for.
Model providers set the costs and the rules
The product depends on models from OpenAI and Anthropic, so their prices and limits set its costs; the buyer's page now lets users add their own OpenAI key to cut costs. The building method depends on providers too: in March 2026 the founder wrote that usage limits made Claude Code "unusable" for him, and three days later that he had cancelled his Anthropic subscription.
A new owner rebuilt it
The buyer writes that she did not build the first version and has been fixing and rebuilding it since she bought it. She raised the monthly prices from US$29 to US$99 to between US$56 and US$374. The figures on this page describe the founder's version, not the business today.
Sources
The grade next to each source is the best grade that the source can support. "Read in full" means that the text of the page was downloaded and the numbers were checked in it.
- C Pete McPherson, guest article on Making Sense of Cents, 3 October 2025: How I've Made $350,000 From AI-Built AppsRead in full. The founder's own account. Source for the dated timeline, the build in three days, the US$7,000 to US$8,000 a month, the partners, the sale and the stack.
- C eBiz Facts, Niall Doherty, 20 January 2025, updated 9 March 2025: He Made $100,000 Last Year From AI-Coded AppsRead in full. A short profile that summarizes the founder's video. Source for the US$6,000 to US$7,000 a month. Also read in the copy syndicated on Goodreads: https://www.goodreads.com/author_blog_posts/25441508-he-made-100-000-last-year-from-ai-coded-apps
- C Pete McPherson, YouTube channel Code Playbook, 28 December 2024: How I made $100k last year from AI-coded appsRead the video page and its description, which list his stack and his builds. The speech could not be transcribed, so the revenue figure is cited from eBiz Facts.
- C Charles Floate, YouTube, 2 September 2024: Meet The Dev - Topical Map AI - Pete McPhersonRead the video page and its description, in which Charles Floate calls the founder his business partner.
- C Topical Map AI terms and conditions, archived 13 April 2024Read in full. Source for the sole proprietorship owned by Pete McPherson and the law of Michigan.
- C Topical Map AI home page, archived 13 April 2024Read in full. The first archived copy. Source for the launch offer and the founder's own description of the problem.
- C Topical Map AI home page, archived 20 January 2025Read in full, and compared with the archived copies of 22 April, 15 July, 12 August and 6 September 2025, which show the same builders and plans, and of 16 December 2025 and 10 February 2026, which show the buyer's new site. Source for the builders, the plans and the AI models.
- C Topical Map AI home page, read 14 October 2026Read in full. Source for the buyer's statement, the current prices and limits, the account and map counts and the own OpenAI key option.
- C Topical Map AI terms and conditions, read 14 October 2026Read in full. The terms now name the buyer as the sole owner.
- C Pete McPherson, projects page, read 14 October 2026Read in full. Source for the sale statement and the stack of Topical Map AI. It dates the product 2023 to 2025, while his article dates the build to March 2024.
- C Pete McPherson on X (@doyouevenblog), posts of 3 October 2025 and 24 and 27 March 2026Read the timeline through api.fxtwitter.com. Source for the link to his article and for his posts on Claude Code limits and the cancelled Anthropic subscription.
- B Google Search Central: A guide to Google Search ranking systemsRead the section on the helpful content system, which says it became part of the core ranking systems in March 2024.
- B Google Search Central blog, 5 March 2024: What web creators should know about our March 2024 core update and new spam policiesRead the section on scaled content abuse.
- B Google Search Status Dashboard: History for RankingRead the list of ranking updates from 2023 to 2025. The address leads to the Google Search Status Dashboard.
- B Pew Research Center, 22 July 2025: Google users are less likely to click on links when an AI summary appears in the resultsRead in full. Source for the click rates with and without an AI summary.
- C Surfer help center: Topical MapRead the collection page, which documents a competitor's Topical Map tool.
- C Loot Drop database export, 1,749 failed startupsLocal export. AI-written summaries, not independently verified. Source for the failed predecessors.
- C Topical Map AI affiliate program page, read 14 October 2026Read in full. The buyer's affiliate program with a 40% recurring commission.