Linear GTM 如何做到 1 亿美元 ARR

在 YouTube 上打开视频 ↗
Linear 创立的目标是打造更好的问题追踪工具。七年后,在获得 1.34 亿美元融资之后,其 CEO Karri Saarinen 发布了一篇题为《问题追踪已死》的文章。在本期节目中,我们与 Linear 的 COO Cristina Cordova 交流,聊了公司如何应对这一局面,以及其市场推广策略一路以来的变化。
Cristina 在 Stripe 工作了七年半,曾在 Notion 负责平台与合作伙伴关系,加入 Linear 之前是 First Round 的合伙人,并于 2023 年加入 Linear。Linear 的创始人在 Coinbase、Airbnb 和 Uber 亲历了遗留问题追踪工具的种种问题后将其打造出来,并通过 Vercel、Ramp 等初创企业以自助式方式扩散开来。
她加入时,Linear 大约有 40 人,只有一名销售人员。此后,公司推出了 MCP server、一个用于在 Linear 中构建智能体的平台、与 Codex、Cursor 和 GitHub Copilot 的集成,以及自家的 Linear Agent。一年前,智能体创造了 Linear 工作量的 3%。现在这一比例已超过一半。Linear 今年 ARR 突破 1 亿美元,付费客户超过 40,000 家,净收入留存率达 177%;8 月,员工以 25 亿美元估值出售了 9,900 万美元的股票。
我们聊了她为什么在加入三周后就停止追逐企业客户,为什么 Linear 从未招聘过一名 SDR,为什么它不让 LLM 自行回答支持工单,以及支撑“数月内 ARR 达 1 亿美元”的 AI 预算对其他人意味着什么。
Linear 于 2025 年 5 月推出其 MCP server,Cristina 称它基本上是同类产品中的第一个。一年半后,Linear 中创建的问题有超过一半来自智能体。
但这并不意味着 Linear 想成为一个上面架着 MCP server 的数据库。Cristina 认为,满足于此的 SaaS 公司正在犯一个大错。如果你产品内部的智能体工作流不比外部的更好,那就没人有理由打开你的应用。
所以她关注两条曲线:智能体使用量和应用使用量。如果智能体使用量上升,而应用使用量没有,那你正在变成一种大宗商品。在 Linear,两者都在上升,因为人们仍会回到应用里查看和协调工作。
跳转至 MCP 讨论 →
当 Karri 在 3 月发表《Issue tracking is dead》时,Linear 的大多数客户都认同他的观点。然后他们问了一个显而易见的问题:那我到底在买什么?Linear 的答案是:一个围绕智能体编排构建的产品开发系统,人类不再需要做提交、打标签、分配和关闭 issue 这些手工活。
客户一直在把它往这个方向推。大多数前沿 AI 实验室都在用 Linear,并希望它能走得更快。像 Ramp 和 Coinbase 这样的“前 AI 时代”公司正在构建自己的内部智能体,需要一个 API 把它们接入 Linear。
Cristina 说,Linear 幸运在于它足够小,可以自我蚕食。Google 面临的问题要难得多:在每个搜索结果上方放一个 AI 答案,看广告的人就会变少。
跳转至《Issue tracking is dead》讨论 →
Cristina 说,眼下加快采购速度的是 AI 预算。公司为以前不存在的 AI 工具划出了专门的资金,任何能装进这笔预算的产品都能跳过通常的预算争夺战。在一些公司,这笔预算非常高。
情况已经开始变化。买家在签一年期合同和付费试点,而不是三年期合同,因为没人知道自己两年后需要什么。而一些「一夜之间做到 1 亿美元」的故事,其实只是与国防公司或大型律所签下的几笔 3000 万到 4000 万美元的合同。这和 Linear 的生意完全不同——Linear 拥有 4 万名付费客户,而直到 6 月之前,它的定价完全基于席位。
当这些一年期合同到了续约的时候,我们就会知道谁的 ARR 是真的。
跳转到 AI 预算讨论 →
2023 年加入三周后,Cristina 就把「向企业级市场进军」从 Linear 的公司目标上划掉了。当时公司只有一名销售人员。她转而瞄准那些不得不拆掉现有工具的成长期公司——就像 Stripe 在赢得转换客户之前很久,就靠拿下新业务取胜一样。
Linear 终于在 2025 年宣布进军企业级市场,路线图上排满了面向大型组织的功能。然后智能体浪潮来了。Linear 构建了 MCP server、带一整套全新 API 的智能体平台,以及与 Codex、Cursor 和 GitHub Copilot 的集成。这些都不在计划之内,而很多企业级工作始终没有完成。
Cristina 还是选择抢先往前跑。竞争对手选择观望,而企业开始把 Linear 放进候选名单。
跳转到企业路线图讨论 →
Linear 从未招过一名 SDR。Cristina 认为资格审查电话是在浪费客户的时间:他们想要的是演示,而不是被问预算。AE 自己做外呼,并借助工具让外呼保持个性化。
它也不做前置部署工程。Linear 应该足够容易上手,以至于没人需要访问你的系统。取而代之的是,解决方案工程师与企业 AE 一起销售,有时会在一夜之间做出一个可用的 ServiceNow 集成,向潜在客户证明这能做到。Cristina 补充说,相比每个团队对“bug”的定义都各不相同的工具,一个有明确主张的工具能给智能体提供更干净的数据。
支持仍由内部团队负责,Linear 不会让 LLM 自行回复工单。AI 起草回复,由人决定发出什么内容。客户有时会以为回复是 AI 写的,因为来得太快了。三年半里,Cristina 从未因糟糕的支持体验收到过升级投诉。
跳转至销售团队讨论 →
Cristina 很想要一个像 Intercom 的 Fin 那样的模式,按其 AI 解决的每一次支持对话收费。但对 Linear 来说更难。当一家公司认为自动把 bug 分派给正确的团队是产品经理的活儿,另一家公司却不这么认为时,这项功能该值多少钱?
在有人解决这个问题之前,Linear 只对让它产生成本的部分收费,其余保持免费。席位费不变。Linear Agent 聊天是免费的,尽管它跑在前沿模型上。在编程会话中,用户自行选择 OpenAI 或 Anthropic 的模型与运行框架,Linear 按公布的 API 价格加一小笔加价转售。Loops 则消耗 AI 额度运行。
Linear 不会对过去免费的东西开始收费。新的 AI 功能以免费测试版上线,并提示即将收费,而且价格是从高起步再往下走,而不是反过来。
跳转至定价讨论 →
Cristina 说,Linear 的人员规模大约只有同阶段、同收入公司的一半。如果销售团队比公司其他所有人加起来还大,代价就是工程团队不会再像现在这样信任销售。
她从 Stripe 带来了一个 40% 规则:公司至少 40% 的人应该是工程、产品、设计和数据岗。如果销售团队想增加人手,这有助于先招工程师。Linear 没有硬性规定,但她最早做的几件事之一,就是按职能做了一张人员编制表,而且她至今仍在关注这一比例。
招聘是刻意放慢的。候选人要针对一个真实问题做工作试炼。她的题目是撰写 Linear 未来六到十二个月的市场推广计划,后来这成了她的工作。公司约 170 人,唯一的 C 级高管就是创始人团队和她。
跳转到招聘讨论 →
Linear——Cristina 担任 COO 的产品开发系统。
与打造 Linear 的人分享公司增长——8 月以 25 亿美元估值进行的要约收购,付费客户超过 40,000 家,净收入留存率为 177%。
问题跟踪已死——Karri 3 月发文谈 Linear 为何面向智能体构建产品。
Linear MCP server——Cristina 称其 2025 年 5 月发布基本是该类别的首创。
推出 Linear Agent——这款内置智能体的聊天功能 Linear 保持免费。
我们如何打造 Triage Intelligence —— 这个能在新进 issue 中建议标签、识别重复项的功能,也是 Cristina 举出的「AI 不必是智能体」的例子
Linear 的 Coding Sessions —— 6 月发布,终结了 Linear 的全席位定价模式,用 AI 积分运行 Claude Code 和 Codex。
推出 Loops —— 可周期性执行的智能体任务,同样以 AI 积分计费。
Intercom Fin 的定价 —— 每次成功解决的对话收费 0.99 美元,Cristina 希望这种按结果计费的模式也能适用于 Linear。
0:51 认识 Linear 的 COO Cristina Cordova
1:39 为什么 Linear 在只有 40 人时就招了一位 COO
5:01 她的入职试用任务:为 Linear 制定一份市场进入计划
12:11 把企业级市场从公司目标中划掉
20:00 为什么只做 MCP 服务器背后的数据库还不够
23:39 在“问题追踪已死”之后如何做销售
34:40 更小、更扁平的团队,且不设 SDR
36:14 由 AI 起草客服回复,人类负责把关
42:23 为什么 Linear 不做前向部署工程
47:53 为什么一些 AI 创业公司能这么快做到 1 亿美元 ARR
50:06 从按席位收费转向按用量计费
57:07 放下 2025 年的企业路线图,转向智能体
1:03:28 Linear 如今的护城河是什么
1:05:54 为什么不招 100 名销售人员?
1:08:59 Stripe 关于工程人员编制的 40% 规则
由修订版剪辑自动转录;人名与措辞可能存在错误。
原文较长,译文为节选;译文由英文原文机器翻译,术语按行业惯例统一,可在文末展开原文核对。
查看英文原文(Scaling Linear’s GTM to $100M ARR — Cristina Cordova, COO)
Linear was founded to build a better issue tracker. Seven years and $134 million in funding later, its CEO Karri Saarinen published a post titled "Issue tracking is dead." In this episode, we chat with Cristina Cordova, COO of Linear, about how the company navigated that and how its go-to-market changed along the way.
Cristina spent seven and a half years at Stripe, led platform and partnerships at Notion, and was a partner at First Round before joining Linear in 2023. Linear's founders built it after living with legacy issue trackers at Coinbase, Airbnb and Uber, and it spread self-serve through startups like Vercel and Ramp.
When she joined, Linear had about 40 people and one salesperson. Since then it has shipped an MCP server, a platform for building agents into Linear, integrations with Codex, Cursor and GitHub Copilot, and its own Linear Agent. Agents created 3% of Linear's work a year ago. Now it's more than half. Linear passed $100 million in ARR this year with more than 40,000 paying customers and 177% net revenue retention, and in August employees sold $99 million in stock at a $2.5 billion valuation.
We talk about why she stopped chasing enterprise deals three weeks after joining, why Linear has never hired a single SDR, why it won't let an LLM answer support tickets on its own, and what the AI budgets behind $100M-in-months ARR mean for everyone else.
Half the work on Linear is now created by agents
Linear shipped its MCP server in May 2025, and Cristina says it was basically the first in its category. A year and a half later, more than half the issues created in Linear come from agents.
That doesn't mean Linear wants to be a database with an MCP server on top. Cristina thinks SaaS companies settling for that are making a big mistake. If the agent workflows inside your product aren't better than the ones outside it, nobody has a reason to open your app.
So she watches two lines: agent usage and app usage. If agent usage climbs and app usage doesn't, you're becoming a commodity. At Linear, both are up and to the right, because people still come back to the app to see and coordinate the work.
"Issue tracking is dead"
When Karri published "Issue tracking is dead" in March, most of Linear's customers agreed with him. Then they asked the obvious question: so what am I buying? Linear's answer is a product development system built around agent orchestration, where humans stop doing the manual work of filing, labeling, assigning and closing issues.
Customers had been pushing it there. Most of the frontier AI labs run on Linear and wanted it to move faster. Pre-AI companies like Ramp and Coinbase were building their own internal agents and needed an API to plug them into Linear.
Cristina says Linear is lucky it's small enough to cannibalize itself. Google has a much harder problem: put an AI answer above every search result, and fewer people look at the ads.
The AI budget behind $100M in months
The thing speeding up purchases right now, Cristina says, is AI budget. Companies carved out money for AI tools that didn't exist before, and any product that fits inside it skips the usual fight for spend. At some companies that budget is very high.
It's already starting to change. Buyers are signing one-year deals and paid pilots, not three-year contracts, because nobody knows what they'll want in two years. And some overnight-$100M stories are a handful of $30 to $40 million contracts with defense companies or giant law firms. That's a very different business from Linear's 40,000 paying customers on what was, until June, entirely seat-based pricing.
When those one-year deals come up for renewal, we'll find out whose ARR was real.
Linear threw out its 2025 enterprise roadmap
Three weeks after joining in 2023, Cristina crossed "move upmarket to enterprise" off Linear's company goals. It had one salesperson. She went after growth-stage companies that would have to rip out an existing tool instead, the same way Stripe won new businesses long before it won switchers.
Linear finally declared an enterprise push in 2025, with a roadmap full of features for large organizations. Then agents happened. Linear built an MCP server, an agents platform with a whole new set of APIs, and integrations with Codex, Cursor and GitHub Copilot. None of it was planned, and a lot of the enterprise work never got done.
Cristina chose to race ahead anyway. Competitors waited, and enterprises started putting Linear on their shortlists.
No SDRs, no forward deployed engineers
Linear has never hired a single SDR. Cristina thinks the qualification call wastes the customer's time: they want a demo, not questions about their budget. AEs do their own outbound, with tooling to keep it personal.
It doesn't do forward deployed engineering either. Linear should be easy enough to adopt that nobody needs access to your systems. Solutions engineers sell alongside enterprise AEs instead, sometimes building a working ServiceNow integration overnight to show a prospect it can be done. Cristina adds that an opinionated tool gives agents cleaner data than one where every team defines "bug" differently.
Support stays in-house, and Linear won't let an LLM answer tickets on its own. AI drafts replies and a human decides what goes out. Customers sometimes assume the replies are AI because they arrive so fast. In three and a half years, Cristina has never had an escalation over a bad support experience.
Everyone wants outcome-based pricing
Cristina would love a model like Intercom's Fin, which charges for each support conversation its AI resolves. It's harder for Linear. What is auto-triaging a bug to the right team worth when one company thinks that's the PM's job and another doesn't?
Until someone cracks that, Linear charges for what costs it money and keeps the rest free. Seats stay. Linear Agent chat is free even though it runs on frontier models. In coding sessions, users pick OpenAI or Anthropic models and harnesses, and Linear passes through the published API price plus a small markup. Loops run on AI credits.
Linear doesn't start charging for things that used to be free. New AI features launch as free betas with notice that pricing is coming, and prices start high and come down rather than the other way around.
Why not hire 100 salespeople?
Cristina says Linear runs on about half the headcount of companies at its stage and revenue. A sales team bigger than the rest of the company would have cost it the trust engineering now has in sales.
From Stripe she brought a 40% rule: at least 40% of the company should be engineering, product, design and data. If sales wanted more people, it helped recruit engineers first. Linear has no hard rule, but one of the first things she built was a headcount spreadsheet by function, and she still watches the ratio.
Hiring is slow on purpose. Candidates do a work trial on a real problem. Hers was writing Linear's go-to-market plan for the next six to twelve months, which then became her job. At about 170 people, the only C-suite is the founders and her.
Show notes
Linear — the product development system where Cristina is COO.
Sharing Linear's growth with the people building it — the August tender at a $2.5 billion valuation, with more than 40,000 paying customers and 177% net revenue retention.
Issue tracking is dead — Karri's March post on why Linear is building for agents.
Linear MCP server — the May 2025 launch Cristina says was basically a first for the category.
Introducing Linear Agent — the built-in agent whose chat Linear keeps free.
How we built Triage Intelligence — the feature that suggests labels and catches duplicates in incoming issues, Cristina's example of AI that doesn't need to be an agent
Coding sessions in Linear — the June launch that ended Linear's all-seat pricing, running Claude Code and Codex on AI credits.
Introducing Loops — recurring agent jobs, also billed with AI credits.
Intercom Fin pricing — $0.99 per resolved conversation, the outcome-based model Cristina wishes fit Linear.
Timestamps
0:00 Intro
0:51 Meet Cristina Cordova, COO of Linear
1:39 Why Linear hired a COO at 40 people
5:01 Her work trial: a go-to-market plan for Linear
12:11 Crossing enterprise off the company goals
20:00 Why being a database behind an MCP server isn't enough
23:39 Selling after "issue tracking is dead"
34:40 Smaller, flatter teams and no SDRs
36:14 AI-drafted support replies with humans in charge
42:23 Why Linear doesn't do forward deployed engineering
47:53 Why some AI startups hit $100M ARR so fast
50:06 Moving from seats to usage-based pricing
57:07 Trading the 2025 enterprise roadmap for agents
1:03:28 What Linear's moat is now
1:05:54 Why not hire 100 salespeople?
1:08:59 Stripe's 40% rule for engineering headcount
Transcript
Automatically transcribed from the revised edit; names and wording may contain errors.
0:00 Karri writing this blog post, like issue tracking is dead. So like how do you explain to your customers that like, hey, you know, our entire business was issue tracking before and now our CEO just wrote this blog post? The great thing about our business is that we're not like so big to like kill ourselves. Right? Yeah. Like, it's probably really hard if you're Google to, like, kill the search business. Yeah. Right? And put an AI, you know, box at the top of every search result, and maybe no one's gonna look at the ads. Right? So your entire business is upended. And, like, yes, the unit of the issue still exists, but we probably shouldn't have a world where people are, like, manually tracking work. Yeah. So, like, that's what's dead. Right? So in a lot of cases, when we were talking to various customers as this, you know, blog post came out, most of our customers were like, yeah. Tracking is dead. And by the way, like, what are you guys? What am I buying? Right? Cristina Cordova, you're the COO of Linear. Could you introduce yourself? Yeah. Yeah. So I'm Cristina. I've been at Linear for about three and a half years now. Joined as COO, and I'm still COO today. Prior to Linear, I spent time at Notion leading our platform, so the build out of our APIs, and then our partnerships and self serve growth initiatives as well. And then before that, I was at Stripe for seven and a half years. And there, I led a lot of our early partnerships and go to market, and then moved over to the product side and built out some new products for early stage companies. So I've been probably anywhere from like the product side of a company and like managing engineers all the way to, managing, you know, the business side of a company as well, which has been really fun.
1:30 So why did Linear hire a COO at, like, 40 employees? Is that, like, odd or is that pretty normal? I mean, I I generally don't believe every company needs a COO. Yeah. Right? Like, when I talk to founders and they're like, hey, like, I think I found product market fit. Like, how do we start to scale, like, the business side of the company? I am often, I feel like, cautioning against maybe hiring someone like me. So I think for every company, it's a little bit different. I think you have to really understand as, like, a founder, like, what are the things you need and what are the things you're looking for and the people that you bring on to the team. Often early on, it's just like I need someone to, like, sell something. Right? We have expanded beyond the capability of the founder CEO to sell every net new customer. Right? So there, I generally talk about, well, you should hire folks in sales. Maybe hire a sales leader to build out a team. Right? But if you're not exactly sure what you're looking for in, like, marketing or operations or other parts of the company, then it may very well not make sense to hire someone like me. You know, I think with the Linear founders, you know, they were incredibly focused on, like, design and engineering and product. And I think they had a really great, you know, kind of combination of skills between the three of them. But as we were starting to scale the business side of the company, they didn't necessarily look at, like, the next salesperson that we hired and said, I know exactly what we're looking for here, and I know whether or not sales is performing really well. Right? So if you don't look at different parts of your company and say like, okay, I know what great is, and I feel like I have a good sense of, like, how to get to great, it can sometimes be helpful to hire someone like me or, like, a functional leader, like, when it comes time to do that.
3:00 Yeah. So, like, what did the business look like at that point? Like, can you talk about, like, what revenue they were at? Or Yeah. We were probably... I wanna say, you know, we had some large customers, but no, like, public company customers. Probably no customer above, I wanna say, like, maybe 500 seats or something like that. So I would say, like, we mostly tapped out at, like, mid market. And we were probably best well known for being a great product for startups, but not necessarily a great product for other companies Yeah. At that time. You know, and I think at that point in time, we were we're doing well with like a product led growth model where most people were using our product because they signed up self serve. Mhmm. And then grew with us over time. So companies like Vercel and Ramp and many other customers like that, you know, really never talked to a human being at Linear. Signed up for the product, started using it, and then started scaling, which was great. Like, you know the product can do a lot of work on its own. But it was also start... Like, time to start selling to, you know, the company that is not even gonna put, like, a lick of data into your service without, you know, signing a contract Yeah.
4:30 And doing a pilot and surveying their team on, you know, what tool they're using before and what... You know, what tool is this and what's better. Right? So that was a very different kind of motion for the company to start scaling to larger companies. And that was, like, pretty much my objective. And we've at least 10 x ed, I think, since I joined. So... And you joined, like, four years ago? Three... Almost three and a years ago. Yeah. Oh, okay. So I think the Linear founders have talked about this. I think you have as well that, like, before you join Linear, you always do a, like, a trial period? Yes. So did you also have to do a trial period? I did. Okay. Yeah. Yeah. So, yeah. Like, what was that like? Yeah. I mean, I think my project was quite simple. It was something like, you know, come up with the go to market plan for Linear, you know, with a bounded time period, like, for the next six to twelve months or something like that. Right? And I think the good thing about that is that, you you know, that is the job. Like... Right? So you're not giving someone some, like, fictional project that's not really, like, representative of, like, what they would do when they join the company. It's like a real, project. And that would, you know, effectively be my job in joining the company. So Yeah. You know, if they didn't like the outcomes that I produced as part of that project, they probably, you know, shouldn't hire me. Yeah. So I think it was a really good way of saying, like, I have some ideas of what the company can do better, but also I wanna, like, get feedback from the people who are on the ground. Like, you know, we had a salesperson on our team. We had someone doing customer success. We had some people doing support and, you know, one or two folks doing marketing. Right? So kind of getting their perspectives on different parts of the business and, like, where I should focus and where where to prioritize.
6:00 And then making sure that, like, the company was also bought in on, like, that vision for the world, was I think really helpful for me. And then when I actually started at the company, it was almost like I have this little blueprint Yeah. For, you know, what my job should be and what I should accomplish. Can you talk about, like, what was in that plan? Think a lot of it was like brand and marketing oriented. So like, you know, things around like making Linear inevitable in a lot of ways. So, you know, either you're already on Linear or you should be like planning your move to Linear effectively. And most companies that we wanna acquire should be, you know, in, you know, one world or the other. So a lot of that was around, you know, driving the brand, building storytelling, I think increasing awareness for the company because I think we were we were well known, but I would say in like particular niches, not really outside of Yeah. Okay. So it's like, you know, we were doing marketing, but like the only marketing that we were doing was like sending out emails if we, had a change log of something new that we released. And it turns out that, like, less than 10% of users, like, subscribed to emails. So most people didn't necessarily know even just, the new features we were shipping, which was interesting. And then we would, like, send a tweet.
7:30 Okay. Like, from from the founders' accounts or, like, from the company account? And then and then, you know, the founders would reshare and, you know, they have their own, like, pretty strong, like, social presence as well. But, you know, that was like the extent of it, really. Yeah. Like, so, you know, a lot of it was just thinking about, okay, like, how are we telling a more holistic story in a better way? And then how are we actually reaching the people that we wanna be reaching? It turns out that, like, actually a lot of our users are, like, in the app, you know, day to day. Yeah. And so, like, why are we not just, like, giving them notifications about, like, what's changing within the product, like, within the place they're already at every day. Right? So, you know, just thinking about things that could be great ways to market to existing users or market to, like, net new customers that also didn't feel like off with the brand. Right? Yeah. So I spent a lot of time on that. A lot of it was also just like growing the sales team. You know, I think a lot of early stage companies, especially that are like engineering driven, like product led growth oriented are like, you know, the product should sell itself. And then everyone realizes like, well, that gets you to a certain point. Yeah. And then when you want to, like, grow faster, right, you also need to start thinking about, reaching people in different ways. And not everyone buys products self serve. And especially when you're choosing a system that's like so core to like your product development process, which we are. Yeah. A lot of companies are not going to make that decision without like building trust with human beings. And, you know, when I was doing my work trial, we had, one person on sales. Right? That was not you. So it was just like, you know, there's there's only so much you could do with, like, one person.
9:00 Right? And were they, like, an experienced salesperson? Yeah. Yeah. Yeah. Yeah. Yeah. He's actually our sales leader today. Okay. Cool. So leads our entire sales team. So he's great. But, you know, at the time, I would say this wasn't really on him. It was more on the company. It's just like, how do we actually think we wanna grow this team? Right? You can grow a sales team in lots of ways. You can have like lots of SDRs. Mhmm. We don't have any SDRs. You can have forward deployed engineers now. That's very popular. Right? We actually don't do, forward deployed engineering. We have solutions engineers, which I, you know, I think is distinct and somewhat different. But we added those later. We really... And that's part of the sales org? Yeah. Cool. We really optimized our team at the beginning for hiring AEs who had experience, like, selling technical products. Because we knew, like, if you had sold to a designer, if you had sold to an engineer before, we wouldn't necessarily have to teach you how to do that. And I think selling to someone who's like very discerning, doesn't wanna see your fancy slides, just wants to get into a demo Yeah. You know, is very different from selling to like a person in, I don't know, people ops or, you know, something else. So, looking for people with that, like, very explicit experience and then making sure that we could, like, onboard them and train them and get them, like, really competent in the product itself. Yeah. Because that's required to build trust with our customer. Right? So just figuring out, like, what's different about this company and what's going to be different about hiring or scaling, you know. And also, like, what's the way that the founders wanna do it and, like, can we do that? So support is an interesting example. When I was interviewing, the founders were very clear that, like, they did not wanna outsource support. And I've never worked at a company that didn't outsource support. So that was like...
10:30 Yeah. Thought Stripe does it internally. Right? No. They outsource. Oh, okay. Yeah. Interesting. Unless something's changed since I was there. But, you know, they have internal folks who work on support too. Yeah. But they do outsource. So, you know, I think for us, that was something that we really wanted to maintain because we really cared about, like, delivering a high quality experience, delivering a fast experience, and knowing that for a long time, we didn't have a sales team. Right? Yeah. So because we didn't have that, we knew that your first touch point, often, if you wanna talk to a human being, was going to be support. So the support better be good. Yeah. Right? And so we've still been able to figure out how to scale the team today into, like, a global team that does not outsource at all. So, you know, we'll see what the future brings. Yeah. You know, never say never on any of these things. But you know, are certain ways that we wanna do things and we probably wanna do them for like as long as possible, as long as long as it's like feasible for the business and makes sense.
12:00 Yeah. So when you first joined, like, I guess I've never been in a sales role, but like do you have to say, okay, we're starting with startups and we can't just immediately go to, like, big enterprises and start selling them? We have to, like, kinda go step by step. So, like, did you have to go, oh, we have to first get, like, medium sized businesses, and then at some point, we'll go enterprise? Like, how does that work? Yeah. Yeah. So when I joined Linear, I think we had, like, three goals, like, three company level goals. And I think one of them was, like, move up market to enterprise. And then we had, like, a company off-site three weeks after I joined. And I, like, got up in front of the whole company, which was, like, I don't know, 35, 40 people at the time. And, like, I showed the slide of the three goals, and I, like, crossed the, like, one that says we're gonna go after enterprises out. Okay. Because I was like, we have one person. We have literally one person on sales. Like, you know, you're not going upmarket to enterprise with one person. Yeah. Like, you know, so you have to think about, like, what are you capable of doing? What do you need to, like, learn in order to to get there? And then do actually have the resources to, like, accomplish that goal? Right? And, you know, I I don't think we had any of those things at that period of time. Yeah. Right? We had companies that we had primarily won because they had scaled into being like a growth stage business or like a mid market business effectively. But we hadn't done a lot of like acquiring net new, like mid market companies at that point. So I changed that goal to say like, no, we're gonna go after like companies in the growth segment. And that's gonna be our goal. And if we can win a lot of companies who are in the growth segment, who already have, like, a tool, that does in some ways what we do, and we're... We have to go in and we have to convince them to, like, rip that out and replace it with Linear, then if we're able to do that, that gives me the confidence to say that, you know, in a year or in two years, we're gonna be able to go after enterprises.
13:30 But like, I don't have confidence in that now, so why would I set that as a goal for the team? So completely agree. It's like really about setting those baby steps and making sure that, like, you're doing well with companies that you can serve today incredibly well. So that those companies, when they're doing like, you know, they're getting, like, the reference calls from the larger companies being, hey. Like, do you like Linear, and has your experience been good? They're saying, like, really positive things. And that helps us land, like, the net new customer. And, like, it worked very similarly when I was at Stripe. Like, we did not win a lot of, like, net new companies that were switching over from a competitor Yeah. For a long time. We did a great job, like, winning every net new business that was created. So, like, a startup Yeah. That's implementing payments for the first time. And then we kind of started to win companies who were switching over. And some of that stuff just takes time. It's very... Like, it's very different to sell a product where you have to rip technology out that already exists and replace it than just to sell something that's, like, nice and shiny and new on top of, like, an existing set of systems.
15:00 Yeah. So knowing that I was in, like, a very similar position in this role, it was important that, like, I just oriented the team around, like, what I felt was like a solvable problem. And then only in twenty twenty five... So that was 2023. Only in 2025 did I, you know, go up in front of a company and say like, we're going after enterprise this year. Right? Oh, wow. And that was when we started our enterprise sales segment and built out that team. Yeah. So... Okay. You guys are going like medium sized businesses. And so you are getting everyone to basically like... Because most peep... Most companies have an issue tracker. Right? Yep. Or some kind of... So you have to try to like rip them away from them. So yeah. Like how do you convince them to say, oh yeah, we're using x y z product. It works for us. But, yeah, like, you know, pitch me, basically. Yeah. And Why should I switch it? The early days, it was about being, like, significantly better Yeah. Than what the alternative solutions were. Right? I don't think you can be, like, 20% better, 30% better, and get people to switch because there's a switching cost involved. Right? So I think you have to figure out how you articulate how your product is better than what is already, like, serving the market reasonably well. Reasonably well enough to... That they have like a lot of companies that are using their product. Right? But the good thing about the space that we were entering, and I think one of the reasons why like the Linear founders chose this space were because they were all users of those legacy tools Yeah. Right, when they were working at, like, Coinbase and Airbnb and Uber. And so they knew, like, no one liked these tools.
16:30 Yeah. Right? And when they were getting rolled out, like, at their companies, they were like, oh, no. Like, you know, the fun time is going away. We have to start using these tools. Right? So, in a lot of ways, it really helped that they had this, like, knowledge and existing awareness of, like, the feeling that came with the competitors in our space Yeah. And could figure out how to build a product that got people really excited, like, to manage their work. Right? Which is presumably not something that you would expect. So a lot of the emphasis in what we built early on was in like design and UX. Right? Yeah. Make it a product that people want to be in every day and have to use. The hard part about that is that people have to like try it before they get it. Right? Yeah. Versus, like, seeing a couple slides in a sales deck and being like, okay. Like, I get it. Right? So one of the things that we did, I think, very well strategically was just having a product that anyone could start using. So, you know, a free product initially, that once you hit certain limits in terms of usage, you would have to start paying for. So a lot of people who started to use Linear for their company, actually were just playing around with it for their own purposes for free and got a sense of like how usable it was and how much easier it would be to utilize the product compared to competitors. So we started very much there. And I think that worked well for a period of time. Now our approach in terms of like how we sell Linear is very different. We're typically talking to companies about the sea change in the space that has happened over the last year and a half, two years in terms of like, you know, your entire, like, product development process has been upended. Mhmm. The way that you build products has completely changed.
18:00 And it doesn't make sense anymore to live in a world where you are... Like, you have, like, a manual way of saying, like, oh, like, I found a bug. Oh, I'm going to, like, report that bug into my issue tracker. Yeah. Oh, I'm going to label it and assign it to this other person. And then that person is going to fix it and, you know, add a status update and then close it and then, like, notify the customer. And every step in that path is, like, completely manual. That's just... It doesn't make any sense for the world that we live in now. And you talk to most companies and they're like, yes. I completely agree. And so we've really oriented the company in the last year and a half around agents and agents performing a lot of the parts of the workflow that are effectively not product development. Like everything I described in that whole chain is like, that's not you building products. And so if you can make Linear a really great tool for agentic workflows, then the individual engineer or the individual PM or designer who's using Linear can ultimately be spending more of their time building products and less time like manually moving around work in a system to prove that they are adding value to a company.
19:30 Yeah. That's really interesting because it seems like the pitch early on was the user experience. Right? And now now it seems like the the humans are not going to be interacting with the with the UI. I actually think that a lot of companies right now are feeling like it's gonna be okay to just be like a database Mhmm. And, you know, have an MCP server and, you know, give all of their data away Yeah. To agents and have the agents like interoperate with them and think that like that's going to be a great business model. And and I don't think that's going to be a great business model ultimately. Yeah. I think you have to bring the agentic workflows into your tools and you have to make them better than the tools that people would be using externally to interoperate with your system. Otherwise, there's no reason for people to go into your tool. Right? And so for us, you know, the metric that we're typically looking at and and we've seen like a lot of change here where, you know, we launched our MCP server in 2025. I think we were basically the first company in our space to do it. And now, over 50% of issues created on Linear are created by agents in just like a year and a half.
21:00 Yeah. The work is created in an entirely different way than it was before. You know? So that kind of illustrates how much has changed in a pretty short period of time. And then when we look at how many users are coming to the app and integrating or creating issues within the app itself, that's also going up significantly. Okay. Right? So sometimes people believe that like, oh, I'm just I'm just using MCP server for everything now. But like you are also going back to the app and you're interacting with it because you need to visualize things. And often, like, a chat interface isn't the best way to visualize all the work that a team is doing. It's not great for collaboration because it's typically, like, single player only. Mhmm. And so I think for us, as long as, like, you know, the MCP usage line and the app usage line are kind of going up into the right together Yeah. It proves that, you know, we have both a great UX that's great for humans and also great for agents. Right? And that's like our goal, I think, as a product, that you can be great at both of those things. And, you know, it doesn't mean that like now that agents exist, you, like, don't have to care about the interface. Mhmm. And app usage is measured by just, like, number of people using it? Because, like, I'm I'm assuming also, like, your customer base is just growing. Right? So how do you Work created
22:30 in the application. So, you know, our, like, unit of work, like, smallest unit of work is the issue. Right? So if you just look at, like, work created, like, that's like a human going into the app, like our mobile app, our desktop app, you know, our web app, all of those places, and creating work. Right? If you look at how the work is like moving through the pipeline in a way, like getting assigned to an agent, sometimes that's done manually by a human. You know, sometimes an agent does it. Mhmm. But assign... Like delegates to another agent. Right? Yeah. Often, we're now part of the workflow. Right? You can use a feature we have called loops that will, like, automatically, assign an agent to a given issue and have the agent start, like, coding against it, for so in a lot of cases, it's like an agent is doing the work, but it's a Linear agent. Yeah. Right? Which we also wanna see. Right? So we have to look at, like, multiple different factors. But it's generally like app sessions that we're looking at in terms of trying to understand activity. Yeah. So I guess that leads me to like Karri writing this blog post like issue tracking is dead. Yeah. So I'm assuming you were in the middle of selling some contracts when he wrote that. So how do you explain to your customers that like, hey, our entire business was issue tracking before and now our CEO just wrote this blog post. Yeah. Yeah. Like how do you explain that to them and how did that transition happen?
24:00 Yeah. Well, I think the great thing about our business is that we're not, like, so big to, like, kill ourselves. Right? Yeah. Like, it's probably really hard if you're Google to, like, kill the search business. Yeah. Right? And put, an AI, you know, box at the top of every search result, and maybe no one's gonna look at the ads. Right? So your entire business is upended. And so, you know, I think that's a little bit of the like innovator's dilemma challenge that a lot of tools that were, like, pre AI have, and we were certainly one of those companies that was pre AI. Yeah. And so for us, like, the great thing is that it's like, okay. You know, we see that the world is going in this direction, and you can either, like, lean into it with the plan that you build the best product for the new era Mhmm. Rather than trying to, like, hold on to an era that, like, is probably not going to exist in a few years. Right? So, you know, if the workflow that I talked about with, like, you know, this like manual intervention by humans at every step of the of the process is not going to be the work that we live in. Like, you look at Linear now and like, yes, the the unit of the issue still exists. But we probably shouldn't have a world where people are, like, manually tracking work. Yeah. So, like, that's what's dead. Right? So in a lot of cases, when we were talking to various customers as this, you know, blog post came out, Most of our customers were like, yeah. Tracking is dead. And by the way, like, what are you guys? What am I buying? Right?
25:30 And so it's much more oriented around, you know, now being like a product development system. Right? We want you to focus on building better products, and we are going to build a system that enables you to do so with very little manual work. So it's also about, you know, saying, like, we're gonna create a new category. And like, yes, we will, you know, in a lot of ways still play in the old category, but we are also driving the vast majority of our resources as a team into this new category of things like agent orchestration, managing work in a way where you don't necessarily have to be doing a lot of manual, you know, taking tasks through a system, and that you're operating far more agentically, both, like, within Linear and also outside of Linear. So I guess, like, that transition or, I guess, that blog post, were you blindsided by that? Like, Karri writing Or like he was No. No. We were all were all on the same So were you seeing like were you seeing your customers were asking for that? Or was it that you wanted to push your customers that way? I think some of our customers were asking for I mean, Linear is used by the vast majority of all the Frontier AI labs, as an example. You know, the vast majority of top AI companies already use Linear. So if you look at, like, that customer set, so, you know, this kind of, like, net new generation of, like, companies Mhmm. That started after Linear. Many of those companies were saying, you know, this is the way that we're going. Right? And we want Linear to move in that direction. Otherwise, it's not gonna be the right product for us. Right? And then at the same time, we also saw with companies that were also like pre AI. So examples of this might be companies like, Ramp or Coinbase where, you know, they existed before all of this Yeah. And decided that they were going to invest their own internal teams to build like internal agents to do work with them. Yeah. Right? And so we were like, hey, we actually need an API that enables you to create agents within Linear.
27:00 So if you're one of these companies that, you know, it's almost pre AI but on the bleeding edge of AI still, and you're creating your own like bespoke agent to help manage various different types of work or code internally and you wanna own the proprietary data that's associated with that, we wanna build the API so that you can deeply integrate Linear with those tools. Right? And like there wasn't a system or a set of APIs that any of our competitors had at the time for that. So again, just like being first to build a lot of these new ways of interacting with Linear, you know, formats of agent workflows, and then building agents directly into Linear as well. I think we're all very new for us, and some customers were asking us for them. And I would say a lot of customers weren't. But I can tell you that now a lot of them are very happy Yeah. That we have a lot of these features. Because not every company is on like the bleeding edge of the way that product development is changing.
28:30 I guess that's what I was trying to get at is like, would you... I... I'm assuming because you were trying to go more like enterprise, a lot of them were those like legacy type companies. Right? Yes. And and not the kinds of companies that are really pushing us Yeah. In this direction. But now those are the companies that are reaching out to us and saying like, hey, a lot of the way that we're doing things has changed and the systems that we use to build products should be changing just as rapidly as like the landscape of coding agents as an example. Mhmm. And if you feel like your underlying toolset hasn't really changed, hasn't really evolved in this new world, then you should probably think about changing your toolset. Right? When you start like building these new features, how do you teach your customers how to use those features? How did you guys do that? Yes. Yes. So... Yeah. One, I would say, you know, again, we really emphasize with, you know, functions like sales and customer success that they understand the product very deeply. And we really try to get our sales and success teams to like share learnings with each other. So, you know, a lot of our calls are recorded with Gong as an example. And often, our own customers will be like, you know, screen sharing their workflows with us. And then we get to take that as like a learning to say, you know, here's how this company is doing it. Really interesting. Maybe I can take that to like another customer. Right? So there's both the, like, learning from your customers, which... I mean, when you have customers who are builders, you should always think about, like, how you're going to learn from them versus purely just how you're going to teach them. Right?
30:00 It's nice to have, like, AI Labs as customers. Exactly. Exactly. And then you're taking those learnings and you're saying, how do I package this up for a user that is maybe not at that point yet? Right? And then, you know, for example, we had customers who were like, oh, yeah. Like, wouldn't it be great if, you know, there was like an agent that existed that would like triage everything for me? Because like, no. Nobody wants to like groom their backlog. Right? Like, what a task that feels so outdated. Right? And so, you know, we built a a feature called triage intelligence, which effectively enables, anything that comes into your triage workflow, bug reports, feature requests, etcetera, to be, like, automatically labeled, assigned, put on the right team's queue. And like, you, as a human, do not have to do anything for that to take place. Right? So it's like, you don't need to, like, build an agent for this. We can just do it for you. Yeah. Right? So not everything necessarily needs to be an agent. It can just be part of the product. Right? So it's both taught us, you know, certainly how to teach the next customer, like, how to do things better. But it's also helped us incorporate that into the product itself.
31:30 Yeah. And then it becomes far more, you know, easy for a person on our team to sell. So how does that work if... Like, how do you teach salespeople how to use, like, engineering tools? Oh, well, the great thing about Linear is that, like, anyone can use it. Like, you know, if I think about, like, GitHub issues or something, when I go into GitHub, I'm like, oh. Like Yep. As, as, you know, someone who's not an engineer, like, oh, I'm gonna mess something up. Like, I don't belong here. Right? But I think the great thing about what we do is that that's not true. Like, we try to create a product that's, like, incredibly usable and, like, well designed, not just for engineers, but really for anyone. Mhmm. Right? And so now increasingly, we're getting rolled out across, like, you know, wall to wall within organizations. So, you know, teams like sales and ops and, like, anyone who wants to, like, know what's on the product road map or, like, be an input into what's on the product road map is using our product. And it's very usable for them too. Right? So I think that makes things far easier just because it's just... It's not scary Yeah. As, like, a system or a tool. And our salespeople, you know, have learned that. It just... In being at the company, we do a lot, as you can imagine, in Linear. So Yeah. You need to know how to use it just to be at the company generally. And then again, a a lot of it is in great, like, training of your people. And I think we've invested a lot in enablement. You effectively have to go through, like, various coursework effectively in order to be certified to even sell Linear Yeah. And get on a call solo with the customer. So we really require that, like, our sales reps understand a lot about the product. And, you know, maybe the most, like, complex engineering workflow would probably be, like, our code reviews product and or, like, our GitHub integration. Like, those can be somewhat complicated often because you have to set GitHub and and have like a demo GitHub workspace and Yeah. A demo Linear workspace to make sure that they all work together. That can be a little bit more complex.
33:00 But, you know, most of that is, you know, in relying on another tool and an integration workflow with it. But I think the great thing is like now we have like salespeople on our team shipping like copy edits to the website. And we have marketers on my team who are, you know, deciding like, oh, we wanna change like how we store this data so that we have like a better sense of like what this revenue operation system does. And, you know, things like that Yeah. Are like now accessible to everyone. And an engineer on our team just has to like review that change and Yeah. You know, approve it. And I think it's also making, like, just Linear the product and what we do far more accessible to them too. But, yeah, it takes time
34:30 Yeah. Yeah. Like with anything. So are your... Like like, is your team composition changing? Because now you're having, like, different teams, like, push copy updates to the website that maybe normally wouldn't have done that. And maybe, like, are you seeing, like, customer team compositions are changing because of maybe, like, how they're using Linear or just, like Yeah. Just agentic development, like, in general? Yeah. I mean, I think one of the things that people are talking about a lot today are... Is effectively how teams are changing in that they're remaining smaller, flatter, etcetera. I feel like I've been a lot of companies that, like, aim to work that way for a period of time, and then they just like grow out of it. And Yeah. And it changes. I think luckily, like, as a lot of these developments in our space were happening, like, Linear was already operating that way and I think has continued to operate that way. A lot of which, you know, can be due to AI. But I think as an example, you know, other teams do this differently. I think we've never had SDRs as an example. And I think that that comes from a belief that like a salesperson should be able to like write an outbound email. Like, I don't see why. Like... So what do SDRs normally do? Typically, they're, like, either doing outbound to new prospective customers and trying to get them in the pipeline or they're qualifying those customers as they come in. So they'll do, like, the first call. But unfortunately, that first call is very, like, not useful to the customer because, You're the second is just being like, so how big are you? And, like How much money do think might be able to pay us? Right? And again, like, our customer would hate to be on that call. They're like, why are you asking me questions? You are supposed to give me a demo.
36:00 Yeah. Right? Like, why am I wasting thirty minutes of my time on this call just to get to a person who can give me a demo? Right? So, you know, we don't have that function because I don't think it works well for our customers. But also because you can create systems and tooling and automations that make it easier for an AE to do outbound to a customer. Right? That feels like custom and useful. Right? And is able to like ultimately get to some point of engagement. Right? So, you know, that's one maybe difference between us and and other companies. I think we've I think we've also opted to think about areas of the business where we can have like fewer employees. Right? And what can you optimize with like tooling? So our customer experience team is like a great example. So, you know, what are the tools that we can use to automatically draft replies to customers Yeah. That write into us. Do you not see that like having LLMs write stuff to customers like produce slop? Like Yes. Like AI slop? Like, how do you how do you fix that? Yeah. I mean, the CX team, like, humans are the quality check I see. On that. Right? So we do not have AI send mass. Like, they... Like, a lot of companies. And other companies can do this, and and that's totally fine. Right? No no shame to other people. But a lot of other companies will say, hey, like, a 100% of questions that come into our support team are going to be, like, answered by an LLM first.
37:30 Yeah. And only if that customer writes back angrily, right, will a human come in and be like, oh, sorry about that. Yeah. Like, let me give you a real answer. Right? And we just don't believe in that model. Right? So I would much rather have AI like draft an answer, to the specific question based on the knowledge that, you know, it has coming in. And then a human can decide like, is that a good answer or not? But very often it's like, gosh, this would have taken, three sentences. And instead, I have, like, three paragraphs of, you know, a reply that ultimately is, not the most helpful. Right? And we get so much feedback on support where people were like... Are almost like either was this AI because your response was so fast. Yeah. And we just... You know, our our team's really fast. Yeah. So people think it's AI because it came to them so quickly. And then the second being like, wow. This is like the first good interaction I've had with like a support team in a long time. And that's kinda sad, you know, for like other tools and products out there. But, you know, for us, it's one of the things that like we give a lot of kudos to our team for that because it like sets a brand impression. Sometimes early in a customer's life cycle and sometimes, you know, they've been with us for years and are riding in for the first time and it gets them really, you know, feeling confident they they made the right decision. Right? So I think you can use these tools. They're tools. I don't believe in, like, blindly using these tools in a way that makes your... Like like, compromises your brand Yeah. And ultimately makes your team feel like it's... Like like almost less useful or less knowledgeable Yeah. Because it's relying on a tool to have all the knowledge that, like, a human should have.
39:00 Yeah. So you... Did you mention, like, downsizing the customer experience team? So... Not downsizing. But I I don't think we've ramped it as as we would have without Right? So I think some of these tools are are very helpful for enabling us to, like, not always have to grow teams linearly with, like Yeah. How how fast the business is growing. Yeah. Right? And, you know, I think on, like, other parts of the business, like like marketing and and other areas where it's like, oh, okay. We can do interviews with customers for customer stories. And, you know, we can have an AI tool, like, transcribe this interview, pull out like some of the most compelling pieces, but like we still have a human write the story. Right? Yeah. So just deciding like where you're going to decide to use it, like we don't like the way that a lot of AI writing feels. Yeah. So, you know, we've been like... Well, like, use it to help us, you know, find the interesting parts of, like, two hours that we spent with a customer. Yeah. But in a lot of ways for it to feel like how we want Linear to feel, which is like, you know, custom and handcrafted. Like, you know, we shouldn't use those tools for certain kinds of things.
40:30 Yeah. At least until they get better. So... Astra is pretty good at writing, I've noticed recently. Yeah. But anyway Yeah. Yeah. How does... Like, how does sales engineers fit into your team... Into the sales team? Yeah. Yeah. So we have what we call solutions engineers Oh, sorry. That are effectively the the same thing. They are kind of like partners to our enterprise account execs and probably like the upper end of our growth segment. And so they're spending a lot of time with sales as a customer has like really deep workflow questions. Or they're thinking about, you know, hey, I integrate with like this tool over here with it like a very custom API integration. How would I move that to Linear? Right? So you're asking someone who has like a deeper technical question. Yeah. Right? Or how would I perform a migration of 30,000 people from, like, in a legacy tool to Linear? Yeah. Right? And so, you know, a solutions engineer can come in with a technical mindset. Like, all of, you know, these folks can And the person build they're talking to is an engineer presumably? Yes. Yes. Almost always. And so, you know, that person has very detailed questions. And sometimes the people on our team are building like full end to end demos of like, oh, you wanted to see how we would integrate with ServiceNow? Like, I actually built this ServiceNow integration overnight. And like, let me show it to you. So it also makes a lot of these like transitions that customers are going through like, you know, very easy because they start to see like, wow. Like, it is possible to do these things like very quickly. Like migrating isn't that hard. Right? B
正文保留英文原文,标题与摘要由机器翻译。
原文发布于 Latent Space(Scaling Linear’s GTM to $100M ARR — Cristina Cordova, COO),标题、摘要与正文由机器翻译,术语按行业惯例统一,未经人工逐句校订。