Klipy Hit Five-Figure MRR. The Approval Gate Is the Wedge. (2026)

The short version

Klipy reached a publicly disclosed five-figure MRR in July 2026 with three founders and a product that removes the work around selling: logging conversations, rebuilding deal context, preparing follow-ups, and keeping the pipeline current. Founder Jung-Hong Kim says the first version launched in November 2024, is used by around 4,000 companies, and is still bootstrapped by the same three-person team. 1

The replicable part is narrower than the headline. Klipy is not winning by being another chatbot for salespeople. Its product thesis is that every channel and meeting should feed one account timeline, which then produces a reviewable next action. The human still approves every outbound message and record change. That combination gives the product a concrete wedge: speed without autonomous outreach.

The non-replicable part is Kim's starting position. He had already built and sold two machine-vision companies, had enterprise architecture experience, and met his cofounders while working as a consultant. The team understood the sales-operations problem from inside enterprise software. A cold-start founder can copy the workflow thesis and the pricing logic. They cannot copy that domain history or the six years of scar tissue that led Kim to this problem.

Snapshot

Metric	Publicly disclosed	What it means for this case
MRR milestone	Five-figure MRR, disclosed July 15, 2026 1	The exact dollar amount was not published, so do not reverse-engineer one.
Founder team	Three founders: Jung-Hong Kim, Tina Law, and Joey Lee 2	Meets the channel's team-size filter.
First product launch	November 2024 3	The product was more than six months old at the disclosure.
Customer count	Around 4,000 companies in the founder's July disclosure 1	Paying-customer count, retention, and customer mix are not disclosed.
Current public plans	Free; Starter at $39/seat/month; Growth at $89; Professional at $149; Enterprise is custom 4	Monthly, per-seat pricing with usage allowances and feature gates.
Public operating signals	2,300+ sellers, 53+ countries, 2.3M+ emails processed monthly, and 1,000+ meeting hours monthly 2	These are current site claims, not a revenue breakdown.

The public record contains two different scale measures: Kim says around 4,000 companies, while Klipy's official information page says 2,300+ sellers. They may use different definitions or snapshots. They should not be added together or treated as the same customer metric. Churn, ARR, trial conversion, revenue by plan, gross margin, CAC, and LTV have not been disclosed.

Origin: the CRM problem was a sales problem

Klipy began with a failure, not a generic AI brainstorm. Kim says he had built and sold two machine-vision companies for retail analytics, then watched a later retail-focused startup collapse when COVID erased the market it depended on. He went into serious debt and spent about six years as a management consultant focused on infrastructure and enterprise architecture. During that period, he met the two people who became his cofounders. 1

The relevant pain was specific. Kim had used HubSpot for years, but found it difficult to make sales teams keep the system updated. Salespeople did not want to perform data entry, so the CRM became a delayed and incomplete record of what had happened. Sellers then spent time reconstructing the context before every follow-up, switching among email, meetings, messages, and the CRM itself.

Kim described the problem in a LinkedIn post after the Indie Hackers story: good salespeople were losing deals because they were buried in data entry, context chasing, and rebuilding deal history before each follow-up. He said Klipy was built to remove that weight so one person could operate like a larger sales team. 3

That origin matters because the product is built around a repeated operational failure, not an abstract category. Kim was both a seller and a developer, so he could dogfood the workflow and hand-code the first MVP in about three months. The initial product launched as an automatic CRM in November 2024 and has since expanded toward a supervised sales agent. 1

Wedge: faster follow-up without giving up control

Klipy's public positioning is a supervised AI sales agent with a built-in CRM. It captures conversations across email, calls, LinkedIn, WhatsApp, and Telegram, then drafts the follow-up, CRM update, meeting brief, or next task. The seller reviews and approves the action. Klipy's homepage states that nothing is sent without that approval. 5

The product wedge has four connected parts.

Conversation-native context. Instead of asking a seller to paste a transcript into a chatbot or update a CRM after a call, Klipy keeps a timeline across the channels where the deal actually happened. The next draft can refer to the prior conversation, not just the last email.
Execution after the meeting. Meeting transcription is only one component. The product turns a call into follow-up drafts, to-dos, pipeline changes, and pre-meeting briefs. The value unit is the next sales action, not the recording.
Approval as a product boundary. Autonomous SDRs promise volume by sending without the seller. Klipy makes the opposite trade: it prepares the work quickly but leaves the final send and record change with the human. That is a meaningful choice for founders and consultants whose relationships are hard-won.
A CRM that can be filled automatically. Klipy says it can act as the system of record or sync with HubSpot, Salesforce, Attio, and Pipedrive. The buyer does not have to abandon an existing CRM to remove manual logging. 2

This is more specific than "better AI sales automation," but it is still a positioning claim, not an independently audited win-rate comparison. The public material does not show Klipy's win rate against Gong, HubSpot, Fathom, or autonomous SDR products. What it does show is a coherent reason for a buyer to switch: a seller wants the speed of automation while keeping control of the words that reach a prospect.

There is also a useful target customer hidden in the architecture. Klipy's official information names owner-sellers and lean B2B sales teams as its main audience. Those buyers feel the cost of every missed follow-up but may not have a sales-operations hire to maintain a complex stack. For them, a product that connects inbox, meetings, pipeline, and playbooks can replace coordination work rather than merely add another AI feature. 2

Pricing teardown: charge for approved work, not for previews

Klipy's current public pricing is simple at the top level and more interesting underneath.

Plan	Price	Allowance and operating boundary
Free	$0/month	200 tokens/month, one inbox, one seat
Starter	$39/seat/month	400 tokens/seat/month, unlimited inboxes, up to five seats
Growth	$89/seat/month	1,500 tokens/seat/month, unlimited inboxes and seats; API, webhooks, and selected CRM integrations
Professional	$149/seat/month	4,000 tokens/seat/month, unlimited inboxes and seats; custom playbooks, MCP, Salesforce sync, PII masking, and audit logs
Enterprise	Contact sales	Custom commercial and security terms

Klipy's pricing page documents this plan structure. 4

The first decision is free review. Klipy says it generates drafts, briefs, follow-ups, and pipeline updates without consuming tokens during review. A token is spent when the user approves an action. The page lists different costs for different actions, including two tokens for a draft and ten tokens per hour for a meeting recording. Paid plans can buy extra tokens at $0.10 each, with a cap that the customer controls. Rejected work costs nothing. 4

That is a clever answer to a common AI SaaS problem. If the product charges for every generated draft, users hesitate to explore it. If it gives unlimited execution, the company absorbs an unpredictable model bill. Klipy lets a seller inspect the value before the billable event, then ties revenue to the action the user accepted. The model is not free usage; it is free evaluation followed by paid execution.

The second decision is the anchor. Starter is $39, Growth is $89, and Professional is $149 per seat. Professional is about 3.8 times Starter's price and is paired with security, audit, custom-playbook, and agent-integration features. Growth is marked as the most popular plan and carries the broadest expansion jump: 1,500 tokens versus 400 on Starter, unlimited seats, API and webhooks, and CRM sync. The upsell is therefore a mix of usage capacity, team scope, and operational integration rather than more words from a language model.

The free tier is not a content teaser. It lets one person connect one email-and-calendar inbox, see the core CRM and meeting workflow, and try the product on real deals. That is enough to expose the wedge before a buyer commits. Paid plans remove the one-person and one-inbox ceiling, then add the integrations and controls a team needs.

The founder's Indie Hackers post says the team tested channel add-ons, token-based pricing, lifetime deals, and other models before settling on result-based pricing. The company also uses bonus tokens for testimonials, referrals, and social posts. 1 Those incentives can make acquisition cheaper, but they are not proof of referral volume or payback. Klipy has not disclosed paid conversion, plan mix, or whether the token model improves retention.

Acquisition: launch communities first, direct sales second

Klipy's founder describes two practical acquisition motions rather than one magic channel.

The first was launch-community distribution. The team studied the previous month's top performers on Product Hunt, Microlaunch, Reddit, and AppSumo, then created offers tailored to each platform. Lifetime deals were used to recruit an initial 100 core users. Those users were valuable as testers and potential referrers, even though Kim notes that lifetime deals are difficult to make profitable for a product with ongoing LLM costs. 1

The specific tactic is worth separating from the platform list. Klipy did not treat a launch directory as a one-day traffic event. It used the offer to recruit people who already lived in the target market, watched what they complained about, and converted some launch-site members into affiliates. The deal created a feedback cohort and a distribution layer at the same time.

The second motion was cold direct sales. Kim calls it crucial in the early period. His public playbook starts with ad pixels and funnel-event tracking, then recommends building a well-matched audience on LinkedIn Ads, collecting testimonials, publishing explainer videos and lead magnets, and scraping competitor LinkedIn page followers for targeted outreach. The pitch is to tell a prospect what Klipy does differently, not to send a generic AI-sales message. 1

Klipy's later support layer is SEO and AEO through help articles. That can compound discovery and reduce activation friction, but the public disclosure does not name ranked keywords, traffic, conversion rates, ad spend, CAC, or the percentage of MRR from any channel. The honest conclusion is that the tactics are concrete while the channel attribution is not.

For a one-person shop, this is a more useful acquisition sequence than "post every day." Start with a concentrated cohort that will use the product, gather proof from that use, then target a narrow buyer list with a specific operational claim. If the product is meant for sales teams, the founder should be willing to do sales. Klipy's current scale cannot be attributed to launch sites alone, and the founder does not claim that it can.

What a cold-start founder can copy

Pick a workflow you can observe in detail. Count the handoffs between inbox, meeting tool, CRM, calendar, and follow-up instead of starting with a category label like "AI for sales."
Define one irreversible product boundary. Klipy's is simple: it can draft and prepare, but it cannot send or change a record without approval. A clear boundary makes the product safer to trust and easier to explain.
Capture context before generating output. A follow-up assistant with only the last email is easy to replace. A timeline that unifies calls, messages, email, and CRM state has a stronger reason to stay connected.
Charge around accepted work. Let buyers inspect output for free, then charge for the action they approve. Make the cost visible before the action and give them a hard monthly cap.
Build the free tier around the first proof moment. One inbox and one seat are enough if the customer can see a real follow-up draft quickly. Do not give away unlimited team operation before the buyer reaches that moment.
Make upgrades follow operating complexity. Add seats, inboxes, integrations, audit logs, and playbooks as the customer's process grows. Do not make the main upsell an arbitrary word or message quota.
Use launch communities as a research cohort. Tailor the offer to each community, recruit a small group of committed users, and turn their objections into product and sales material.
Track the numbers the public story lacks. Trial-to-paid conversion, approved-action rate, token cost per account, churn by tier, channel CAC, and gross margin are the difference between a useful playbook and a compelling anecdote.

Honest assessment: the advantage is accumulated sales context

Klipy's three-person structure is replicable in principle. The specific team is not. Kim is a serial founder who had built and sold two companies, worked in enterprise architecture, and combined engineering and sales ability. Tina Law and Joey Lee joined him after working together on enterprise projects. The product came from years of exposure to how software fails inside sales organizations, not from a weekend survey of a new market. 1 2

The other advantage is access to enterprise buyers and the credibility to sell into workflows involving sensitive conversations. Klipy's official information lists security controls, CRM integrations, and an approval-gated architecture. Those choices may help procurement, but the public sources do not disclose sales-cycle length, enterprise win rate, or how much revenue comes from larger accounts. Treat the enterprise positioning as an operating direction, not a proven moat.

There is also a technical head start in the product's integration surface. Supporting Gmail, Outlook, meetings, LinkedIn, WhatsApp, Telegram, several CRMs, APIs, webhooks, and MCP is a substantial amount of product work for a solo founder. It is not impossible to reproduce, but it is not a weekend feature either. A smaller entrant should choose one buyer and two channels rather than copying the entire surface area.

What remains genuinely portable is the reasoning: remove a recurring coordination cost, connect the systems that create the context, and make automation safe enough that the buyer will use it on consequential work. That is a product thesis. It is not a promise that the same launch path will produce the same MRR.

Three lessons that generalize

The wedge is a workflow boundary, not an AI label. Klipy's useful distinction is drafted versus sent. The product earns trust by preparing the next move from a connected deal history while keeping the human as the final step.
Usage pricing works better when the billable event is legible. Free review, approval-based token consumption, a visible cap, and feature-based plan gates give the customer a clear answer to "what am I paying for?" They do not remove the need to monitor margins.
A three-person team is not the same as a cold start. Klipy has three founders, but those founders brought enterprise experience, technical ability, sales ability, and an existing network. Copy the operating constraints and product choices, not the implied starting line.

Sources

Source	Used for
Indie Hackers: A 5-figure-MRR success after a failed product buried him in debt	July 15, 2026 disclosure; founder background; November 2024 launch; three-founder status; around 4,000 companies; origin; pricing experiments; launch communities; direct-sales tactics; disclosed gaps.
Jung-Hong Kim's LinkedIn post	Founder-authored confirmation of the launch date, five-figure MRR, around 4,000 companies, and three-person bootstrapped team.
Klipy pricing	Current plan prices, allowances, approval-based token mechanics, pay-per-use terms, integrations, and plan gates.
Klipy homepage	Current product positioning, supervised approval workflow, channels, CRM integrations, customer quotes, and public operating signals.
Klipy official information for AI assistants	Named key persons, 2024 launch year, product modules, integrations, target customers, limitations, pricing summary, and scale claims.

All revenue, customer, and operating figures are public founder or company disclosures. MRR, ARR, customer mix, retention, channel contribution, CAC, LTV, and margins have not been independently audited or publicly broken out.

Klipy Hit Five-Figure MRR. The Approval Gate Is the Wedge.