Skip to content
MakerGain
← Back to project

Nine Assistants, One Subscription: What CustomGPT.ai Sells Is “Answers That Point Back to the Page”

Edited by MakerGain

In CustomGPT.ai's anonymous pharmacist association case, nine assistants divided by department content share one entry-level subscription; in their busiest month they handled 1,006 questions in total, with zero support tickets over three years. The product retrieves only from published materials and attaches a source link to each answer, turning “adding an assistant” from procurement into configuration. Standard costs 99 United States dollars per month and Premium costs 499 United States dollars per month, tiered by number of agents and query quota, similar to DocsBot's tiering logic.

CustomGPT.ai's official website has an anonymous customer case: a nationwide professional association of pharmacists in the United States runs nine AI assistants on an entry-level subscription. In its busiest month, those nine assistants together handled 1,006 questions; over three years, the association filed zero support tickets with CustomGPT.ai. [1]

Zero tickets can easily be read as “the AI is smart.” The case's own explanation is the opposite: these assistants only report content the association has already published, each answer gives a source, and they do not make clinical judgments for pharmacists. The retrieval scope is limited to published materials, and this boundary is structural—it relies on retrieving only from that set of published materials when answering, not on additionally telling the model “don't talk nonsense.”

This is probably the easiest entry point for understanding this company. What it sells is a Q&A bot that fixes “where the answer comes from” in the interface, not a model that talks better.

Which step it takes over

The concrete action is this: feed in a website or sitemap, help desk, knowledge base, PDF/Office/Google Docs, YouTube/Vimeo videos, audio, and podcasts; the official site says it supports more than 1,400 document formats and more than 100 integrations, and can mix multiple sources into one “source of truth.” After processing, the content becomes a chat window that can be embedded in a website, connected to LiveChat, or integrated into an existing system via API; each answer includes a link that points directly to the source. [2] Integrations page[3]

In the product, this line is design; in the business, it is a division of responsibility: members get not just a passage, but also where that passage comes from; when they have doubts, they can follow the link back and check which material it is based on.

Image: source CustomGPT.ai actual image source[4], retrieved 2026-11-10, original interface preserved.

Technically this is called retrieval-augmented generation. The term does not matter here; what matters is that it determines which step of the workflow the product is placed in—it mainly finds things that have already been written for people, rather than writing new things for them. It is not surprising that organizations like associations, which “face liability if they answer wrong,” appear in its customer cases.

One assistant per department

In that pharmacist association case, there is one detail that explains why the product looks this way better than the numbers do: the nine assistants are built separately by department content, owned by each department, not by IT. Adding an assistant does not require a new contract, a new budget line, or a new procurement process. The case says the ninth assistant costs the same as the second.

At the same time, the case describes “using one assistant to cover all content” as an approach that mixes different content and reduces accuracy. [1]

This trade-off is not flashy, but it is two sides of the same thing as plans being tiered by the number of “agents.” On the pricing page read on November 14, 2026, tiers are divided by number of agents, with query and document quotas scaling accordingly: Standard is $99 per month ($89/month billed annually), with 10 agents, 1,000 queries per month, and 5,000 documents per agent; Premium is $499 per month ($449/month billed annually), with 25 agents, 5,000 queries per month, and 20,000 documents per agent, plus automatic website content sync and removal of CustomGPT.ai branding. [5]

The case does not say who paid for this subscription, nor does it name which plan. What it says is: for organizations like this, what often blocks the second assistant is not technology, but another contract, another budget line, and another person responsible for procurement—the case's exact words are that at that point “the idea quietly dies.” The product makes “adding an assistant” a configuration action rather than a procurement action, and that is probably what it wants to sell to departments.

The cheaper tier costs more per Q&A

There is a mechanism in the official documentation worth pausing on. Query quotas are not calculated at the same cost for every question: credit consumption varies with context depth—Compact and Balanced are 1 credit each, Extended 2, Rich 3, and Max 4. At the same depth, Standard users pay 1 more credit than Premium/Enterprise, so Balanced on Standard is 2. Having the assistant additionally verify an answer or analyze an attachment increases consumption further; the pricing page lists the model selector separately on the Enterprise tier. [6]

In other words, how many “deep” questions you can actually ask with the 1,000 queries in the $99 tier depends on how you ask. This is not a hidden charge, but it is the kind of clause that is only read carefully when the bill arrives.

There is a blunter line on the pricing page: when a plan limit is reached, the system stops indexing new data (word-count cap) or stops answering queries (query-count cap). Pricing page[5] An assistant installed on a member services page suddenly stops answering one day—this failure mode is not hard to imagine, but who discovers it and who handles it, the official page does not say.

The trial and cancellation terms are also worth knowing: the 7-day free trial requires a credit card and automatically charges the selected plan after it ends; refunds apply only to the current subscription fee, not to last month's overage or add-on charges. [5]

Compared with DocsBot, how the quotas are sliced

Pick one common basis for comparison: both tier by “number of bots + query/credit quota.” DocsBot's Standard is $149 per month, giving 3 DocsBots and 15,000 AI credits per month; its credits also start at 1x per message, and may consume more with a larger model or when an AI action is triggered. [7]

In pricing structure, both do the same thing: price “adding a bot” and “asking a question” separately. The difference is in how the quotas are sliced. Which to choose depends more on where the material lives and who maintains synchronization. If content is in Google Drive, CustomGPT.ai's integration is available on Standard and above, but automatic sync in the Google Drive integration is only on Enterprise—sync runs once a day, and if a file is deleted while auto-sync is on, the copy in the platform will also be deleted. [3]

What remains unclear

Company-level revenue, number of paying customers, churn rate, and cost structure are not given in the official pages and cases read, so whether this business makes money cannot be judged in this article.

Metrics also need to be read separately by entity. The associations industry page lists GEMA as a customer success case, saying it supports 100,000+ members, automates 88% of member questions, and that 95%+ of members are satisfied with answers that include citations; a typical branded pilot goes live in about 2 weeks. Industry page[8] This is a marketing page's own account of another customer, and it is not the same as the anonymous pharmacist association above.

What can be confirmed is the shape of the product: it limits the answer scope to approved materials, leaves each answer with a link that can trace back to the source, and turns “adding an assistant” from a procurement action into a configuration action. This case of three years with no support tickets represents only one association and cannot be generalized; but it at least shows that the most concrete part of this business is that it turns “where this sentence comes from” into a traceable action—follow the link back and see which page the sentence actually comes from.

Continue reading

AI search
He Treated Reddit as a Survey. 135,000 People Used It for Research. In 2025, It Stopped at a License

Fed built GummySearch into a tool for audience research using public Reddit discussions: it searched and tracked community conversations by categories such as pain points and people seeking solutions. After commercial licensing talks with Reddit failed, he chose to stop sales on November 30, 2025, and gradually shut down the site rather than continue operating without a license; paying users kept transitional access, and the free tier retained only login, export, and account deletion.

AI supportAI searchAI commerce
Customer Support Bots Start Looking Up Orders for You, and Chatwith Bills “Talking” and “Acting” Separately

Chatwith bills customer-support answers and external actions separately: messages use credits, while actions successfully executed through MCP are billed separately; plans start at 9 US dollars, and an extra 1,000 MCP actions cost 6 US dollars. The founder says he put about 21,000 US dollars of AppSumo revenue into Google Ads in 2024 and at one point a Google Ads bill went unpaid; TrustMRR shows MRR of 3,622 US dollars and 42 active subscriptions in November 2026.

AI audioAI productivityAI search
You Ask Claude What the Client Said Last Week, and It Digs Through Your Voice Notes

Speechyou connects voice notes and meeting transcripts into Claude, letting users search, quote, and jump back to the original audio; the connection uses OAuth 2.1 and is read-only. The annual prices on the two pages are $67 and $100, while monthly is $15 on both. The founder self-reports MRR 1800$; the TrustMRR record shows $1,351 and 104 subscriptions, and the one-person team has listed the product for sale at $35,000.

AI supportAI productivity
Card first, then entry: ChatDash has turned payment collection for AI voice agents into an annual-fee business

ChatDash turns a white-label AI agent dashboard for agencies into a business selling customer slots annually: customers must first add a card before entering the dashboard, and Stripe automatically handles subscriptions and usage-based billing; SaaS Mode is limited to a Growth add-on and supports only Retell voice and OpenAI text. As of 2026-11-14, the TrustMRR page reports cumulative revenue of $603,545 and MRR of $17,167.