序言与财务团队的核心特质
我认为我们将能够吸引并打造出世界上最优秀的财务组织，因为人们最渴望的正是学习的能力，感受到自己被持续投资，并被那些渴望在自身专业领域做到极致的人所包围。
英文原文
I think we're going to be able to attract and build the world's best finance organization because so much what people want the ability to learn and feel like they're being invested in and surrounded by people that want to go and become the best at their craft.我们很想听听，是什么让你团队里的某个人成为一名优秀的雇员、一个出色的人才。
英文原文
>> We'd love to hear a little bit about what [music] made someone a great hireer on your team, a great person.我希望团队成员具备极强的好奇心。你加入进来，试图去解决问题并深入理解业务。你希望将 AI 作为辅助和赋能你的工具，但保持独立思考者的角色依然极其关键。我们现在正处于一个大量“废话内容”（slop）被指出来的时代。我听一位首席财务官说过，使用 AI 导致每天产生无数文档。任何你拿出来要求别人阅读的内容，都会耗费他们的时间、精力和体力。而这其中 95% 的挑战仅仅在于把你的数据梳理到正确的位置。
英文原文
>> I want folks that feel extremely intellectually curious. You're coming in and trying to solve and try to understand the business really well. You want to use AI as a tool to help you and empower you, but it's really, really critical to still be that thinker. We're in an era right now where a lot of slop is being called out. I've heard one CFO talk about how using AI just led to unlimited amount of docs that are being created every day. Anything that you're putting out there that you're asking someone to consume, it's going to take them time, effort, and energy. 95% of the challenge there is just like getting your data in the right spot.我们认为接下来的 12 到 18 个月内会发生什么变化？
英文原文
>> What do we think changes in the next 12 to 18 months?如果我们最大的业务复杂度保持不变，我认为一年后我们可能不需要相同数量的人手。额，
英文原文
>> If our biggest complexity was staying the same, I think we probably wouldn't need the same number of people a year from now. Um,从监督者到业务推动者
欢迎收听 a16z 播客。我是 Seema Amble，与我一起的是我的合伙人 Ivan Maryasin。我们最近撰写了一篇题为《你需要一位新 CFO》的文章。我们的观点是，AI 正在将财务工作提升到一个全新的水平。首席财务官的职责越来越包括构建自动化流程，以及真正设计相应的系统和团队来实现这一目标。今天我们的嘉宾是 Figma 的首席财务官 Praveer Melwani。我们很想探讨你在职业生涯中的见闻，以及你和你的团队目前正在开展的工作。Praveer，感谢你的加入。
英文原文
>> welcome to the A16C podcast. I'm CMA Amble. I'm joined here by my partner Ivon Macarov. We recently wrote a piece called You Need a New CFO. Our argument uh was that AI takes um finance to the next level. And the CF CFO's job increasingly includes building and automation and really designing the systems and the team to do that. Today we're joined by Pier Milwani um who's the CFO at Figma. and we'd love to explore what you have seen over the course of your career and uh what you and your team are doing now. Preier, thanks for joining us.好的，感谢邀请，我很高兴来到这里。
英文原文
>> Yeah, thanks for having me. Excited.或许可以先作为开场，Ivan 和 Praveer，我们想了解一下在你们两人的职业生涯中，自动化是如何改变财务角色以及团队的具体工作的？
英文原文
>> Maybe to kick things off um both [snorts] Ivonne and Preier would love to hear a little bit over the course of both your careers how automation has changed the role in finance and what the team does.你知道，我一直觉得财务人员的角色应当是一位运营层面的领导者。当我加入 Figma 时，所有的自动化组件根本都还没有建立起来。我觉得为了在当时的工作中取得实效，我的大量工作需要沉浸在 SQL 中，构建数据表，真正去理解如何编排财务信息，以推动我希望在全公司展开的对话。我经常回想起自己刚接手那个职位时的日子，思考当时作为“力量倍增器”的举措是什么，正是那项工作让我有能力在整个组织中产生规模化的影响力。如果说有什么不同的话，我认为当年我所构建的东西，借助今天人们可用的工具，变得更加可行、合理且触手可及。因此，我希望越来越多的人能够感到自己有能力以这种运营的视角，去定位自身并与全公司的业务伙伴协同合作。因为我确实认为，这就是将该角色从被视为“交通警察”式的传统财务职能，转变为推动业务发展的使能者以及组织内真正的变革推动者的魅力所在。
英文原文
>> You know, I've always felt like the the role of a finance individual should be this operational leader. Uh and you know you know when I joined Figma none of the automation pieces were were you know in place at all. Uh and I felt like in order for me to be effective at the time that I had joined you know a lot of my work needed to be living in SQL building the data tables really understanding how how I could orchestrate the financial information to drive the conversations I wanted to then have uh across the company. Um, and you know, I continue to I continue to kind of go back to that initial days that I had in that seed and what I was doing as the force multiplier, the piece that actually gave me the ability to have the scaled impact across the organization. And if anything, I think, you know, what I was building way back then is that much more possible and plausible and accessible to individuals by uh by the tools available to them today. So, um I hope that more and more folks feel like they can be empowered to take that operational sort of lens to the way that they actually go and and position and work with their with their counterparts across the business. Um because I do think that's the magic of shifting this role from, you know, what can be viewed as like a traffic cop sort of financey type function to one that's a business enabler and and a true change agent within the organization.进一步探讨你在 AI 出现之前的自动化方式，或许可以向大家解释一下你当时具体在做什么？是在 Excel 中使用宏，还是对文档使用 OCR 技术？
英文原文
>> And to double click on what you were using for automation maybe pre-ai um maybe explain to people what exactly you were doing. Was it macros in Excel? Was it using OCR for documents?数据架构与数据定义的演进
坦白说，当时几乎什么方法都用。最开始我整天泡在 SQL 里，查看我们所有的产品指标和使用数据，并将其与我们在 Stripe 中的财务信息对应起来。我会写出四五百行的 SQL 代码，用它生成核心数据表（golden tables），然后交给其他人，或者同步到我们当时使用的各类 GTM 业务系统中。我觉得自己整天都生活在查询编辑器里。然后我会去告诉——
英文原文
>> Yeah, honestly it was like everything under the sun. So like I uh it started like living in SQL looking at all of our uh product metric and utilization data mapping it to all the financial information we had in Stripe. uh I'd go and create these like four or 500 lines of SQL uh that I'd want to then use to create the golden tables that I'd go and pass off to someone or then be able to pass to different go to market systems that we'd be using. Um so I felt like you know uh I spent my day living in that living in uh that query editor. I'd then go and tell当然还要等待查询运行——
英文原文
>> and of course waiting for the query to run for完全是这样，并且你祈祷自己第一次就写对了。后来我们招聘了数据科学家，我就会去找他们说：“嘿，我写了这个东西，你们能把它加到我们的 ETL 流程中吗？”这样它就能变成定期执行的任务，我就可以直接使用相同的信息同步给 Salesforce、Marketo 或当时使用的任何系统。这对我来说是最初的突破口，让我意识到：如果我们在定义上、在探讨事物的方式上保持一致，我们就再也不必争论什么是真实的、什么是事实，我们可以直接跳过那一步，真正聚焦于我们究竟该如何推动业务发展。
英文原文
>> totally. Yeah. And you hope you got it right the first time. Uh, and then I'd go and I' I'd, you know, eventually I had, you know, we eventually hired data scientists and I'd go and tap them and say, "Hey, I have this thing I went and built. Can you go and add it to our ETL process?" So, it goes and, you know, there's a recurring job that I can then use that same information and pass it to Salesforce or Marquetto or whatever systems that we were using at the time. Um, and that was like the initial sort of unlock for me of like, oh my gosh, if I can get consistent in our in our definitions, in the way that we're talking about things. Um, no longer are we ever debating like what what's true and what's fact. Um, and we can just get past that and really focus on, you know, how are we actually going to go and drive the business.我之前和一个二十岁出头的人聊天，他抱怨一个 prompt 运行需要等那么久。我告诉他，你根本没有体会过那种苦苦等待 SQL 查询运行几个小时，结果跑完发现根本行不通的痛苦。所以现在这完全是一个数量级的差距。
英文原文
>> I was talking to someone who was in their early 20s, um, who was complaining about how long a prompt takes to run. And I was explaining to them, you've not lived the pain of like hours on hours of waiting for SQL a SQL query to run and then only to come back and be like, it didn't really didn't totally work. Um, and uh and so yeah, it's it is just an order of magnitude.完全同意。
英文原文
>> Totally.时间维度确实非常有趣，尤其是我们共同经历的那个时期。我在 Webflow 和 Figma 时，曾看到疫情初期使用量暴增，起初我简直不敢相信。我不知道你是否也有类似经历，当时我甚至以为是数据出错了或者是系统坏了，我们花了好几周才搞清楚。而如今我们所处的时代，面对某些用量或收入的突增，你在一小时内就能查明原因。这是一个截然不同的世界。你也有过这种体会吗？
英文原文
>> Y the time is really interesting, right? Uh especially the era that we've lived through. Um when I was at web flow at Figma, we saw for example a huge spike in usage in coit that I could not believe in the beginning. I don't know if you've you've experienced it this way too where I thought our data was broken or something else was broken but it took us weeks to figure it out and I feel like today in this day and age we live in a day where you could figure it out to an hour with some of the some of the spikes some of the some of the usage revenue as well like this is very different world we live in. Um did you experience that as well?是的，我们的业务在疫情初期实现了腾飞。我认为那也正是我们开始与 Andreessen 展开合作的早期阶段。但确实，我每天醒来都能够重新梳理全盘背景并了解发生了什么。在白天的任何时候，团队都知道我总是在查看每小时的数据面板或每日指标，我不知道他们是否喜欢这样，但我会确保自己能发现可能看起来像异常的情况，或者提出正确的问题。我认为他们看待我就像他们看待自己在组织中的角色一样：一个带着想法的搭档，希望与他们一起全力打造尽可能最优秀的业务，而这一切都建立在彼此共享相同数据的基础之上。
英文原文
>> Uh I mean yeah I mean our business took off uh during the early days of the pandemic. I think that was actually the the the time where where we started to partner with the folks at Indry was like right at right at the early stages over there. Um, but yeah, I mean, I wake up every day. I mean, I I have the ability to to to go and, you know, rebuild my context on what's happening and then at any moment during the day, you know, the team team loves it when I'm I don't know if they love it, but they but they know that I'm constantly in there going and taking a look at, you know, hourly dashboards or or sort of daily metrics on things to make sure that I'm also um sourcing things that I think maybe look like anomalies or, you know, asking the right questions. And I think they view me similarly to how they view how they view their role within the organization is like uh you know [snorts] a counterpart who has ideas who wants to come to them and and truly build the you know the best business possible all using and having access to the same data available to each other.很有意思。在我们读大学的那个年代，监控节奏更多是按月或按季度的；后来变成了每周的节奏。我记得随着数据更多向云端迁移，大家倾向于看一个综合的周报数据面板；而到了现在，每日监控已经成为现实。这是否改变了你的工作习惯，甚至改变了你审视和思考数据的方式，以及思考数据的频率？
英文原文
>> Yeah. What's interesting like when you and I went to college around that time that was more of a monthly or quarterly cadence um and then it became a weekly cadence. I remember maybe especially with data moving more into the cloud it became like let's have a big weekly dashboard to where now what's possible is is daily um has that changed your habits the way you even just look at think about data how often you think about it all说实话，我把数据及其访问权限视为自己高效开展每日工作的关键钥匙。在我在 Figma 的整个时间里，我每天醒来都会看。我记得刚加入的前几天，大概会有 30 个人注册付费方案，我会逐行查看：他们是谁？是个人邮箱还是企业邮箱？是我们可以跟进拓展的人吗？我能从中吸取什么经验，来帮助指导我们最终构建出的面向组织的团队产品？在早期阶段这极其重要。而现在，我深刻理解任何一个方向产生波动的连锁反应。如果你在一件事演变成大问题之前及早发现，能为公司节省成百上千美元；如果发现的是积极趋势，你就可以投入更多资源去顺势放大。因此，我认为这种审视是每日的、甚至是更高频的节奏。我的目标是如何赋能团队乃至整个公司的其他人，让他们同样对这些指标产生责任感，真正以业务所有者的姿态去排查问题，而不是说“我可以等到下周”——你确实需要尽早解决，因为当前的节奏变化非常迅速。
英文原文
>> honestly I view data and having access to it as like the key unlock to how I can run my day effectively um you know and so throughout my time at Figma was the thing that I would wake up every day and I you know I remember we there would be like 30 people that would sign up for a paid plan um in those first few days when I joined and I'd go through line by line who are they are they personal addresses are they business addresses are they people that we can reach out to what can I learn from this um to help us inform what we eventually built which was our organization's product um and I think like I think early days it's extremely important uh and then now I I I just understand the the um the repercussions of you know movement in any one direction uh and if you catch something on before it becomes um uh you know a big deal, it can actually save the business like hundreds, thousands of dollars. Um, uh, but if you catch in the other direction, you can really put more juice behind it and and really seek to to to unlock in the other way. So, you know, if anything, like I I view this as like a daily, if not like, you know, more than that rhythm. Um, and I think the the the goal for me is how do I empower other folks on on my team and across the business uh to similarly feel the accountability um to those same metrics and truly act as business owners and and want to go and triage things and not say like hey I can wait till next week like you do need to solve it sooner rather than later because I do think the rhythms here um are moving really rapidly.AI 时代的人才画像与招聘考核
在深入探讨你们实现了哪些自动化之前，我们想听听在 AI 普及之前，是什么特质让你团队中的员工成为优秀人才，或者胜任领导角色？
英文原文
>> And before we get into what you've automated, would love to hear a little bit about uh what made someone great, a great hireer on your team, a great person um in a leadership role maybe preai.坦白讲，许多特质今天依然相似，只是运用方式略有不同。我希望团队成员极具求知欲。你加入团队，努力去深刻理解业务，并以推动正确结果的方式去解决问题。我认为这一点在今天并没有改变，我依然非常看重。我认为现在的改变在于，你可以让大量乏味重复的工作耗费越来越少的时间，从而把更多精力投入到寻找业务中具有突破性的亮点并将其放大。因此，求知欲、数据敏锐度、提出扎实问题的能力，以及构建思考框架以理解所获得的数据并进行清晰周全的沟通，这些都是核心素养。
英文原文
>> I think I mean honestly a lot of similar things today, but I think that you'd apply them slightly differently. Um I want folks that that feel extremely intellectually curious. you're coming in and trying to solve um and try to understand the business really well and then seek to problem solve in a way that um you know is trying to drive the right outcome. Um I don't think that necessarily has changed today. Like I still very much value that. Um I do think that you can make a lot of the tedious tasks or the repetitive nature of things um uh feel like they take less and less of your time so that you can spend more and more of your time. um really trying to seek how you can um uh you know find the places of magic within the business that you can go and supercharge. Um so you know intellectual curiosity, fluency with data, ability to ask really solid questions um and then structure your thoughts in ways where you can actually u make sense of the data that's being given back to you um and then thoughtfully communicate it on the other side of it.你会如何面试考察这些能力？考核方式发生了哪些演变？
英文原文
>> How would you interview for that? you how has it developed or some of the way挺有意思的，我现在正在招聘一名财务副总裁，对我来说案例研究环节是非常神圣的。我既想测试你如何利用手头的工具——我完全预期大家会使用工具，在案例中我不会给他们太多信息，因为我们是一家上市公司，公开数据很丰富——但关键在于你如何将那些碎片线索合乎逻辑地拼凑在一起。模型的输出是否合理？你不仅能做出那个结果，而且是否真正理解其底层运转机制？因此，我会将案例研究的对话聚焦于：你是否真正理解了自己搭建的结构？在搭建过程中，你是否测试了该模型中对任何变动最敏感的部分？通过你建立的工具，你是否对业务的实际运作形成了强烈的直觉？接着就是我们刚才讨论的，你是否能进行有效沟通？因为任何人都可以坐在那里，使用他们喜欢的任何大语言模型，非常轻松地搭出一个模型。但关键在于，你是否把它当成一个真正能赋能你思考的有效工具？在当今时代，你绝不能把思考外包出去。你应该把 AI 作为辅助和赋能的工具，但成为那个独立思考者依然至关重要。
英文原文
>> uh you know it's interesting today I'm uh or rather right now I'm I'm uh hiring for a VP finance uh and the case study part of the process for me is so so sacred um I want to both test for like how are you using the tools available to you um so yes I you know I fully expect folks to I don't give them a lot of information as part of this case study we're a public company so I get the benefit of you know a lot of data out there um but how you actually piece that that those puzzle pieces together in a way that um actually makes sense. The model output makes sense. You have the ability not just to go and create that output, but you like truly understand how it works. Um so then I focused the case study conversation on like have you really really understood the the structure that you've created. Have you molded it in a way where you've tested what are the highest um uh what what are the pieces of that model that are uh the the most sensitive to any change? Do you have a strong intuition on uh how the business is actually working using the tool that you've created? Um and then like we were just talking about can you effectively communicate that um because yes I think like you know anyone can go and create like a you know a model sitting in and you know pick your favorite take pick your favorite LM you can go and do that really really easily. Um, but like are you using it as an effective tool that like actually is enabling you to to think? And I think like um you don't want to outsource your thinking today. Like I I think you want to use AI as a tool to help you and empower you. Um but it's really really critical to still be that thinker.是的，你描述的大多数特质无论有没有 AI，都适用于衡量一个优秀的候选人。如果有任何不同，那就是面对交付的工作成果时，需要更强的责任心和严谨性，以确保其背后确实存在深刻的思考。
英文原文
>> Yeah. A lot of what you're describing applies to a good hire with or without AI. I think if anything AI just you need more of the um the accountability and thoroughess to know that the thinking is there behind it um as as you receive a work product.Figma 内部的 AI 自动化实战
让我们深入了解一下你具体自动化了哪些内容，我们也可以顺便聊聊我们在投资组合公司中看到的情况。
英文原文
>> Let's drive in a little bit on or to understand what you've automated and we can talk a little bit about what we've been seeing with our portfolio companies. Um能否向我们介绍几个 Figma 内部已自动化的工作流？你们是如何决定重点攻克哪些环节的，又具体做了什么来实现自动化？
英文原文
>> but maybe walk us through a couple of workflows at Figma that you've automated and how did you decide what to focus on and what did you what did you do to automate them?这些尝试涉及面很广，我可以挑选几个我觉得非常有意思的例子。我认为过去一年我们发展非常迅速，从最初探讨如何指导大家拥抱新工具和新工作方式，到现在我感觉自己每天都被大量非常酷的想法所冲击。我们在全员大会（全员会叫 Fig Nation）上设置了一个环节，展示员工如何利用 AI 进行构建，而我们的财务团队总能拿出亮眼的成果。比如信贷与催收团队的 Kajal，他重构了整个工作流程，涵盖从发票跟踪、现金催收，到确保与 NetSuite 系统的完整对账。他是以一种自建系统的方式完成的，将其与我们使用的各个系统打通。这让他得以解放出来，去专注于真正重要的事情，继续缩短应收账款周转天数；接着再去思考还有哪些繁琐任务可以自动化。随后他又去重构了信用审查的执行方式，把数据集以有趣的方式拼接起来，无需再进行人工手动处理。现在他可以向任何人展示并推广这套工具，真正推动了那里的变革。在我们的战略投资与并购（Corp Dev）团队中，观察他们的实践也非常有趣。他们现在在尽职调查中增加了一项检查标准：“你能通过氛围编程（vibe code）复刻出这个产品吗？”这非常有趣，能看出技术壁垒究竟有多深、技术本身到底有多大吸引力。我认为每个人都应该能够——
英文原文
>> Yeah, I you know they're so broad and so I you know maybe I could pick a few that um to me have been fun to see. I think um I think we've like moved really quickly even over the last over the last year to say you know how do I go and uh try to coach people to embrace new tools and new ways of working to now one where I feel like I'm you know constantly being um uh constant constantly being like hit in the face with a lot of these really cool ideas. Um we do a segment uh at the at the company all hands we all hands is called fig nation where u we show folks how others have built with AI and consistently our team shows up as as a team that's that's there um so it's things with you know Kajal and our our credit and collections team he's gone and rebuilt uh his entire workflow everything from invoice tracking cash collections making sure that we can reconcile everything to to Netswuite and he's done it in a way where it's like a homegrown system where you know it's it's tied to the various systems that we use um it then is freed him up to say okay well I can go and continue to drive down my days receivable by focus f focusing on the on the things that really matter um and then focusing on what are the other um sort of tedious tasks I can go and build automation for and so he's then gone and say well I can go and like rebuild the way that I want to run our credit checks and I can go and stitch together the data sets in interesting ways that allow me not to have to go and uh you know manually go and uh into them and it's a singular way where he can now go to um anyone to to say hey go and leverage the tools that have created to to to go and um and really drive um and drive change there. Um on our uh on our core dev team, it's it's been really fun to see like how they you know both will uh you know can you vibe code a product is one of the one of the checks that we use in diligence now um which is actually really fun to see um like how how deep is your remote how how interesting is the technology there and I think you know everyone should be able你们会尝试去重新复刻那个产品？哇，好吧。
英文原文
>> you'll try to recreate the product. Wow. Okay.是的。我认为我们的团队在这方面已经变得非常厉害，因此可以借此排查一些问题：这东西到底靠不靠谱？技术门槛究竟有多高？虽然这些构想中显然存在很多细微差别，但这能为你提供非常扎实的直觉。
英文原文
>> Yeah. Um, and you can I mean I think our team's gotten really damn good at at some of this and so you can get through some of the you know is this a real thing? Is it not? Like how how difficult is it or not? And you know there's obviously nuances to to a whole bunch of these ideas but it gives you like a solid intuition there以前你们做这部分尽调时，必须去找产品团队协助吗？
英文原文
>> and before would you have had to go to the product team to do that part of the diligence?是的，我们依然会与产品团队合作，但现在你可以拿出更多实质成果，比如“看，这是我搭出来的东西，这是我观察到的”，这建立了极高的信誉。
英文原文
>> Yeah I mean we still partner with the product team but you can take a lot more of like hey this is what I built this is what I've seen and it builds a lot of credibility there. um are我插一句，那些做复刻的是战略投资团队的人，对吧？不是产品人员，他们是有财务背景的人，对吧？
英文原文
>> I can pause and those are some of the corp people right not product people their finance background right是的，这非常引人入胜。完全是这样。而且我们是一家非常注重文档文化的公司，大家会写投资备忘录、并购备忘录等。最终阅读和消化这些文档的是我们的领导团队或其他产品负责人。通常情况下，大家看问题总会有一些固定的主题、类似的问题以及领导层特有的审视角度。有趣的是，他们为我们每位管理层成员创建了独有的性格特征模型，并作为一组技能来运行测试他们的投资备忘录，以此切实改进备忘录质量。
英文原文
>> yeah fascinating yeah um totally and they um we're like a very documentoriented culture so they write investment memos they write um uh uh you know M&A memos whatever it may be and you know amongst them the end the end readers or the the the folks that are digesting it are going to be folks on our leadership team or other product leaders and um you know more often than not there are themes there are similar questions there there are ways that we all as leaders tend to have a way of looking at problems and what's what's been interesting is they've gone and created unique p personalities for each of us as leaders um uh as you know a set of skills that they can go and run their investment memos by um to actually improve the process你们拥有类似于一个投资委员会的模拟系统，包含成员之间的互动？
英文原文
>> you have a like an investment committee that exists like the interaction between all of them完全正确，这正是他们最终演示的内容。他们会坐在那里反复模拟推演：“好吧，这是 Praveer 会提出的问题，这是 Dylan 会说的，这是我们的首席营收官 Sean 会说的。”这样你就能找出文档中需要强化或深入尽调的部分，了解需要提前准备回答的常见反馈，从而缩短大家坐在一起开会的讨论周期。因此，在很多领域我们都感到这不仅是一个协助增强的工具，也是一个帮助替代工作流的工具，更是一个帮助你理解和学习的工具。我时常对我们快速迭代演进的速度感到惊叹。对我而言，这让我处于一种极为乐观的状态：我认为我们将能够吸引并打造出世界上最优秀的财务组织，因为人们最渴望的正是学习的能力，感受到自己被持续投资，并被那些渴望在自身专业领域做到极致的人所包围。只要我能竭尽所能去促成和推动这种转变，我相信随着时间的推移，我们就能更好地留住、培养并壮大这些人才。
英文原文
>> throughout exactly so so that's exactly what they'll end up demoing um or rather they'll they'll sit there kind of going back and forth and saying okay this is what Pierre would say and this is what Dylan would say and this is what like you know Sean would say our CRO and so you kind of get you get you get like the areas of the document that you either want to go and plus up do more diligence on what are common sets of feedback that I need to go and make sure I ask the questions over um and then it reduces the cycle time that we're all sitting together on. So you know I I there's a lot of these areas where you know we're both feeling like it's it's a tool that helps you augment. It's this tool that helps you go and replace workflow um and then it's this tool that helps you seek to understand and learn. And I um uh you know I'm constantly surprised at like how um you know quickly we've been able to riff on things. And you know for me it's it's actually put me in this really optimistic spot of saying like um I'm going to I I think we're going to be able to attract and build the world's best finance organization because so much what people want is the ability to learn and feel like they're being invested in and surrounded by people that want to go and um you know become the best at their craft. And as much as I can do to go and effect and enable that change, the better I think we're going to be able to go at retaining and growing and grooming that talent over time, too.这些工具是你们直接基于大语言模型构建的，还是你们也在寻找市面上可以合作的现成 AI 工具？你们更倾向于自建还是外购？
英文原文
>> And are all those tools, did you build those right with the LM directly or are you guys also looking at um off-the-shelf AI tools that you can partner with and would you rather build or buy?我刚才描述的那些主要是混合模式。有些团队成员直接在 Codex 中编写代码并逐步摸索，比如战略投资团队就是这么做的；我们也看到有人在工作流自动化工具中，利用 ChatGPT 围绕 Agent 构建技能，部分信贷与催收工具就是这样搭建的；此外，大家也开始基于我们自己的工具进行构建，比如我们有一款名为 Figma Make 的产品，允许用户通过自然语言 prompt 生成应用程序或网站，提供托管服务，并配有后端支持。这类案例也开始涌现。我对这一点非常兴奋，因为这样我们就可以反哺产品，提出产品反馈，这是大家都乐见其成的。我认为我们目前正处于一个周期节点，大家开始指出其中的大量“废话内容”（slop）。很多讨论从拥有无限 token 转向了在会议中直接将其斥为无意义的产出。我听到一位 CFO 提到，使用 AI 导致每天生成无数文档，你根本找不到任何东西，因为现在每个人都能在几分钟内写出一份甚至十份文档，而以前这需要花上一两周。这自然也引发了关于招聘的讨论：你是招募更熟悉这些工具的年轻员工？还是招募经验丰富且兼具两者的员工？或者招募懂得顶级水准究竟为何物的资深专家？我有点担心的是，我们会忘记顶尖水准究竟是什么样的，就像在 AI 时代之前大家清楚一个真正优秀的模型该是什么样。在这方面你认为未来会走向何方？对于当前用 AI 产生的一些内容，你是否也曾当面直斥其为粗制滥造？
英文原文
>> Yeah. um the ones I was describing uh you know it's a mix like some some of them folks were sitting in codeex and you know I think that's what the team is doing on the um uh on the corporate development team they're sitting in codeex and trying to you know work their way through something um we've also seen folks like sitting within um you know some of the workflow automation pieces and building skills around agents uh using chat GPT and so that's you know how some of the credit and collections pieces were were being built and we've also started to see folks being able to embrace it and build it on our tooling as well so we you know we have a tool called Figma make which allows folks to be able to prompt through natural natural language and create, you know, um an application, a website, uh you can then host it. You have like a, you know, backend behind it. And so, you know, some of those instances are starting to crop up too. So, um I'm actually excited about that because then we can then we can ideally flip into this world where we can give product feedback as well. Um which, uh, you know, I I I know everyone would love. I think we're in the era right now in the period in the cycle where a lot of slop is being called out. Um there's been a lot of conversation too where we went from unlimited token to people just calling it slop um in meetings or I've heard one CFO talk about how using AI just led to unlimited amount of docs that are being created every day and you just can't find anything because everybody can write a doc now or 10 docs in in in in few minutes it took it would take you a week or two. Um were there some of the areas where u and that naturally created also the conversation around hiring. Do you hire younger people that are more versed in these tools? Do you hire um people more experienced that maybe both or do you hire more senior people that know what the world class looks like? And that's what worries me a little bit is that we will forget what the world class looked like just like in the preAI era, you know, where a really good model looks like. Um like especially with some experience and having created models. >> Um where do you think this is going? Like have you at the coal slop on some of the stuff that's being created right now with AI有时确实会。
英文原文
>> sometimes? Yeah.防范内容膨胀与 T&E 自动化案例
是的，我认为我们在内部制定相关政策方面做得不错，比如要求注明哪些内容是由 AI 生成的。极其重要的一点正如我之前分享的：绝不能把思考外包出去。任何你发布出来让别人阅读的内容，都会耗费他们的时间、精力和体力。因此，你必须设定同样的门槛去审核、阅读、总结并精简提炼。你可以以不同的方式起步，我对此完全赞同；但我们所有人都有责任对如何占用彼此的时间保持深思熟虑。我确实遇到过一些半生不熟的交付物，当你深入追问时会发现：你真的理解这里面是怎么回事吗？但我也见识过很多惊艳的场景，惊叹于他们搭建的体系所带来的极高迭代速度。比如几周前，我们正在做 2027 年的规划，开会审查差旅与业务招待费（T&E）组合，探讨在哪里有机会调整政策，或者在哪里我们没有按既定政策执行。我们的企业财务团队在拼接不同数据源的信息方面做得非常出色。我们使用 Navan、Zip、Ramp，为了让这些数据真正发挥作用，需要进行大量模糊匹配。例如：某人出差真的去参加了团建吗？如果他们本身就住在旧金山只是过去拜访呢？你到底该如何标记和处理这些数据？其中 95% 的挑战仅仅在于把数据梳理归位。模型非常擅长替你做出这些初步假设。我不需要它做到百分之百完美，只要在方向上与我的目标大体吻合，能提供足够的信息让我明白哪些领域需要重点发力即可。非常有意思的是，他们不仅输出了一组我们可以实施的建议调整方案，我记得那天早晨我给出了 30 分钟后的视觉结构反馈，等开会时所有东西都已经可视化呈现在我面前。我当时的反应是：“天哪，这真是一个极其绝妙的用例。”而且他们达成这一目标的方式，靠人工是根本不可能完成的。因此，尽管可能会产生很多粗制滥造的内容，但如果你能用正确的工具做正确的事，明白什么是“足够好”、什么是必须做到“尽善尽美”，一些真正的突破性价值就会被释放出来。
英文原文
>> Yeah. I I think like you know I think we we've also done a good job internally um about building policies around like branding when you've written something with AI. Um I think like you know one of the things that's really important I think I was sharing it earlier like you should not be outsourcing your thinking. Um you anything that you're putting out there that you're asking someone to consume it's going to take them time, effort and energy. And so I think you want to make sure that you know you put a similar bar to go and review, read, summarize, make it simple. You can start in a different place. I'm totally cool with that. Um, but I do think it's on all of us to make sure that we're, you know, being really thoughtful about how we how we use each other's time, uh, as well. Um, so yeah, I mean, I've been in I've been in instances where I, you know, things look a little bit halfbaked and you really poke around. You're like, did you do you really understand what's going on over here? Um, but I've been also been in instances where I'm like, oh my gosh, like your speed of iteration here is like pretty damn incredible by virtue of how you set up. Like couple weeks ago, I was sitting in a we're going through 27 planning. you know, sitting in a T& review, we're looking, you know, we're looking at our T& portfolio and saying, okay, where where do we have the opportunity to go and make changes to policy or maybe where are we not like upholding policy where we should um and our corporate finance team, they did a really good job about stitching together information that lived in different data sources. We use Non, we use Zip, we use um RAMP and um there's a lot of fuzzy match that you have to go through in order to like really get that information working. Like was someone actually traveling for for an off-site? What if they lived in San Francisco and were just visiting? like how do you actually want to go and like tag and treat that data? Like 95% of the challenge there is just like getting your data in the right spot. Uh and these models are good at making making these assumptions for you. And I I didn't need something 100% perfect. Um I just wanted something that was directionally, you know, in the right, you know, in the similar vein for where I'm going that gave me enough information to understand, you know, what are the areas I really need to crank the dial on. Um it was really interesting. They both like outputed a set of suggested changes that we needed we needed to we could make. Um, I gave I remember giving feedback that morning 30 minutes later on like some of the construction of how I actually wanted things to look look like. Like like it was all visualized for me when we got in the meeting. I was like, "Holy this is like a really fantastic use case here." Um, and they did it in a way that like um you know uh I don't think they would have been able to do it by hand. Um so you know as much of that slop you know I think can happen. I also think some magic can really get unlocked if you if you if you kind of like use the right tool for the right purpose. um and understand you know what is what is good enough and what needs to be like really really picture perfect.对此我有两点看法。首先，核心突破点在于数据。现在借助 AI，你可以消费所有这些数据源，迅速完成模糊匹配，获得跨数据源大体正确或精确的对账；而在以前，你必须运行 SQL 查询、苦苦等待结果，本质上你必须自己充当商业智能（BI）分析师。第二点是，围绕自动化正在形成一种组织惯性。即使存在一些粗糙的内容，例如我与我们一家成长型公司的首席财务官交流，他说在一两个月内团队创建了四五十个 Agent。显然这些并没有全部投入实际使用，但这让每个人都进入了一种状态：如果我在做繁琐重复的手工劳动且缺乏好工具，那就动手去构建。虽然大约 90% 的工具和 Agent 可能不再被使用，但它促使人们开始以“自动化优先”的思维去思考，我认为这几乎与真正把 Agent 构建出来同样重要。
英文原文
>> A couple of comments to that. Um I mean a key unlock is the data. Right? Now with AI you can consume all these sources of data and quickly do the fuzzy match get u a a directionally correct or precise uh you know reconciliation across these sources whereas before right you had to run the SQL query it had to you had to wait and then and see if and then and then essentially you had to be your own BI uh analyst on on top of it. Um the second piece is there's this muscle that's being developed around the automation right and so even if there is slop like for example I was talking to one of our growth stage CEO CFOs who was saying that in I think a month or two there have been you know 40 or 50 agents created by their team which obviously are not all being used but it was getting everyone in this motion of of okay if I don't um if I'm doing something manual or tedious I don't have a great tool for it let build and so a lot of those maybe 90% I would guess of those tools and agents are not being used anymore but it got people to think okay instead of doing this manually let me think automation first around it um which I think is almost as important as getting the like uh the the agent built自下而上构建与自上而下平台化
我百分之百赞同。此外还在于你如何分享它，如何让其他人了解别人在构建什么。否则就会出现这样的情况：比如我们的内部审计团队花费大量时间跟进员工，确保他们在工单中填写符合 SOX 审计要求的信息；然后你发现，内部审计团队并不是唯一这样做的，GRC 团队在进行 ISO 和 SOC 2 审计时也在做同样的事情；在应付账款端计提预提费用时也有大量类似的往返沟通。既然大家都需要这种分流、拉取信息、对账的操作，为什么不建立一个统一的平台来完成呢？这正是系统团队介入并整合现有实践、将其真正产品化的地方。这样就不会出现很多人各自构建相似组件的情况。我认为这正是 CIO 职能演变的方向：如何将员工自下而上创造的东西，以一种超越最初创建者生命周期的方式进行重构和复用。我们正在全公司大会上做一些跨组织的分享，但更多是在团队内部开展。
英文原文
>> yeah I 100% agree with you I also think like how you share it how you get other people to understand what other people are building and then you end up in these scenarios where where you know our internal audit team is like I they spend a ton of time going and like following up with individ iduals to make sure that they're adding information, you know, on the tickets that they need it so it could be trackable for our sock or socks audit. And then you're like, well, you know, that internal audit team is not the only team that does that. You know, our GRC's team is doing the same thing as they go through like all of our ISO and sock 2 audits as well. You know, you have a bunch of this back and forth on the AP side when they're going and writing acrals. Like you you wanted to go and build this motion of going and triaging, pulling information in, reconciling it, and why not go and build a singular platform that does that. This is actually where I think like you know a systems team is re is really really interesting to come in and and then harmonize some of the things that you see are happening really productize it um so that you don't have all these individuals going and building similar things that I think is like the evolution of what a CIO function should or could be. It's like how do you go and you take a lot of these bottoms up things that people have crafted and you have it you have it re be rebuilt and repurposed in a way that can live on beyond the creator uh and the and the initial builder. Um, so I I I uh you know I think that that that crossorg sharing we're doing some of it at our at our you know um at our all company all hands but we're also doing a lot of it internally across our teams.我之前还和另一家成长型公司的客户成功负责人聊过，他说他的每一个业务代表都用 vibe coding 搭了一个不同的看板，上面显示的数据各不相同。所以正如你所说，必须有某种对账和协调机制。这正好引出了一个问题：关于谁应该负责构建这些系统，大家目前是怎么做的？是由谁来主导？是你们团队的人最初搭建一个 Agent，然后交由中央工程团队接盘吗？我们之前深入探讨过“财务工程师”（finance engineer）这个概念，很想听听你们的看法。
英文原文
>> I was also talking to another one of our growth stage companies um the head of customer success was saying every one of his reps had a different dashboard that they had vibe coded and they had different numbers. Yeah. Um and so I think to your point uh there has to be some reconciling function. Um which would which is a great segue to hear from you guys around uh what are people doing in terms of the teams that should be building this are and who is who's taking the agency? Is it, you know, someone initially builds an agent from your team and then you're taking it to um you're taking it to an engineering team centrally and we talked a lot about this idea of the finance engineer which we can unpack further but we'd love to hear a little bit about我认为由业务专家提出想法并进行迭代一直是非常有价值的。他们知道顶尖的标准是什么，清楚自己想要什么，也熟悉需要理顺的业务流程。在构建最初的原型时，他们表现极其出色。但我是否希望同一个人去负责检查所有的上游依赖项，并确保在他们离职后系统依然能被妥善托管？整个组织中是否还有其他人有同样的需求？这些环节正是世界一流的 CIO 团队能够介入并发挥巨大作用的地方。我们昨天刚迎来了一位新的 CIO，是从 Coinbase 加入的 Casper Sorenson。我们在面试过程中深入讨论过：如何在 CIO 组织内部建立起集中的产品经理职能，与各业务团队协同合作，找出真正奏效的方案，并搭建起他人可以基于此进行开发的底层基础。这与传统系统组织的运作并没有太大不同，但迭代的速度以及赋能他人二次开发的方式，正在发生深刻改变。
英文原文
>> yeah I I think like you know I think we've always had I think having specialists come up with the ideas and iterate. They know what good looks like. They know what they want. Um they know the process that that they're trying to uh make look right. like they're fantastic in going at going about building these initial prototypes. Um, but do I want that same individual to be the individual that has to go and make sure that they're checking for all upstream dependencies and making sure that like, you know, if they were ever to leave that like this thing is hosted in the right way. Um, and hey, does anyone else across the organization have the same need? Probably not. This is where I think like, you know, a world-class CIO org can can come in and really show some magic here. Um, yeah, I'm really excited. We just had a new CIO join yesterday, a guy named Casper Sorenson who joined over from Coinbase. Um, and a lot of like what you know we talked about a lot in in the interview process is like how do you how do you create these like you know centralized PM type functions within the CIO organization that can partner with the rest of the business teams find the things that are really working and then build these like um foundations that other others can build on top of which is not too dissimilar from like how a systems organization would have functioned otherwise but the pace that you can you can um you can iterate on and then how you can empower others to build on top um is what is changing right Yeah.目前公司内部是集中管理的吗？还是说你预计这最终会发生变化？
英文原文
>> Is it centralized right now within the company? Do you or do you see that changing? Would >> eventually就像 Figma 的所有事情一样，我认为我们的业务扩张速度超过了支撑工具的建设速度。我的职责始终是赋能组织去构建，我不希望因为必须先建立底层基础设施而拖慢大家的步伐。因此我们现在有点在追赶这种平衡。在某些团队中，我们在集中化方面做得更好；而另一些团队构建的速度快得多，我们现在正在探索如何提供支持。所以我们目前正处于“混乱的中期阶段”（messy middle）。
英文原文
>> So, um, as is all things at Figma, I think we've grown faster than we've built, you know, the tools to support it. Um, and you know, it's on me to always empower the organization to build and then I'll I don't want to slow people down by creating the infrastructure on top of it. And so, I think we're chasing a little bit of that right now. Um I think with some teams we've done a better job um centralizing that and with other teams I think um you know they they built far far quicker and I think we're now having the process of saying okay how can we go and help? So um I think we're in the messy middle right now.“财务工程师”与数据科学的融合
这与我们从很多投资企业那里听到的非常吻合，无论是 GTM 工程师、人才招聘工程师，还是“财务工程师”的概念。这是一批分布在不同团队中的去中心化构建者，具备不同的背景画像：有时是有工程背景、善于思考哪些工作流需要自动化的财务人员；有时是从其他团队借调过来协助构建的产品工程师。在具体的人才画像上，大家目前都处于探索阶段。
英文原文
>> I think that resonates with a lot of what we're hearing from uh our companies whether it's a go to market engineer, a talent engineer, this concept of the finance engineer. It's a decentralized um set of builders across teams. Um it could have different profiles. You know, some sometimes it's someone who has an engineering background on finance who's good at kind of thinking about what workflows need to be automated. Other times it's you borrowed someone from a a product engineer from another team um to go build. Um but it, you know, I think I think everyone's in this messy middle right now in terms of of what what that profile of person looks like.这让我想起数据科学团队以及他们与组织内其他部门的合作方式。作为数据科学伙伴，你对所合作团队的业务背景理解得越深，你的工作成效就越高。比如我们有专门对接产品团队的数据科学人员，他们建立的产品直觉不亚于甚至超过产品经理和工程师，他们被视为推动变革的核心成员，只是在使用不同的工具赋能业务。同样，我们也有支持财务组织的数据科学专家。有意思的是，其中有几位实际上拥有会计背景。他们深知财务数据必须极其严密、绝不能出错；他们理解核对与平衡、什么是退款、什么是坏账等专业术语与细微差异。这使他们成为更优秀的系统构建者。同样，这也适用于你所说的“财务工程师”。双方需要对业务诉求有共同的理解，这种认知可以随时间逐渐积累，但建立起核心的专业基础至关重要，这样你才能去推动并拓展其他人的思维框架。
英文原文
>> A lot of it reminds me of data science teams and how they partner with other parts of the organization as well. Um, and you as a data science partner are like, you know, the deeper you get into understanding the context of the team that you're working with, like the better I think you you operate as a as a as an individual. Um, so, you know, we've got data science individuals that are partnering with our product team. They build as much if not more product intuition than our than our product managers and engineers and they're viewed as part of that core team um that can go and help ideate and push change and they're just leveraging a different tool in order to enable that. Similarly, we've got folks on, you know, the data science side that are supporting our finance organization. Um, it's actually really interesting like the folks that I think have been devoted there actually come from accounting backgrounds, a couple of them. Uh, and they understand that like, hey, this stuff has got to be like really really hard and can't and and and always correct. Uh, they understand the nuances that we're describing on like checks and balances, what is a refund, what is it, you know, uncollectible, all that all that kind of um uh all that kind of nomenclature. And so, it actually enables them to be better builders. And similarly I could see that translate to like you know the you know the finance engineer that you're describing. You have some shared understanding of what is necessary. You could build that context over time but at least like you it's important to at least like build that major so that you can also go and push other people's frame of thinking as well.谁来拥有这些指标的定义权？我之所以问这个问题，是因为尤其在初创早期，团队可能只有 20 个人，全是创始人，没有任何财务或商务人员。他们在定义服务于报表、投资者以及业务决策的关键指标，直到后来才招募财务人员。过去大家可能每隔几年才重新审视一次指标定义，而现在借助 AI 可以更容易地查看代码逻辑或 Stripe 逻辑，这是否改变了指标审视的频率？
英文原文
>> Who owns the definitions? Like the reason why I asking this question um especially we see it at an early stage where it could be a team of 20 people and just uh founders and no finance or business people and then eventually they bringing them on um but they're building uh definitions of key metrics that serve reporting that serve um investors and really business decisions um and eventually finance gets hired. um where the definitions how often would you it feels like in the past we used to revisit them maybe every couple of years where now because you can access the code or you can access the stripe logic much easier with AI has that changed much how often do you visit it我的团队拥有最终输出的定义权。相比那些早期企业，我们的发展阶段稍显成熟，在口径表达上拥有更多控制机制。对我而言，我们在业务沟通上保持一致性始终至关重要。这让我无论面对董事会、领导团队还是广大员工，都能挺直腰杆，因为他们知道我今天提供的数据与一个月前完全一致，历史数据绝不会随意变动，我也没有隐瞒任何事情。随着公司的发展和迭代，以及财报季中关于各类信息披露的讨论，我们希望保持一致，并确保我们做出的决策在未来回首时依然站得住脚并具备长久价值。因此，如果人们为了更好地发展业务而探索不同的思考维度，这很好；但我认为数据的一致性同样具有不可替代的价值。
英文原文
>> I mean yeah my my team owns the outputs there um maybe I'm maybe I'm a little we're a little bit later than some of those early earlier businesses we have you know more controls in how in how we want to talk about it it's always been really important to me to have consistency in how we talk about our Um, you know, it's one of the things that I felt has always been empowering to like how I can show up for with our board, with our leadership team, with, you know, other folks broadly is that they know that what I'm putting out there is the same thing that I would have put out there a month ago. And, you know, the the historicals never change. And, you know, I'm I'm not trying to hide anything. Um, and I um, you know, I I I would just want to make sure that, you know, as we grow and as we give ourselves the ability to iterate and, you know, we constantly have this conversation on different different sets of disclosure as we go through earnings. We want to be consistent and we we we want to have the decisions that we make um be things that we can look back on and feel um uh feel good about um and have longevity uh further beyond. And so I I you know I think it's it's great it's people finding different ways to think about their business if it's empowering them to go and build their business better. Um but I do think that there's some value in consistency as well.在你的团队中，谁自动化的程度最高？什么岗位？
英文原文
>> Who is it on your team that's automated the most? What role? Uh我可以说出某个人的名字，但我猜其他人很快就会拿出更新的成果。坦率地说，全员都在参与：从负责预测模型具体模块的成员，到业务运营（BizOps）团队中将日常竞争对标和定价研究自动化的员工。我认为正是这种相互激励、彼此超越的氛围，促使整个组织共同进步。因此，要单挑出某一个人会让我觉得不合适，因为这确实是一个团队在共同推进的事业。
英文原文
>> I you know I I I I could say someone but I I'm assume that that someone else is probably coming up with something different really quickly. But you know very very honestly it's like all like individuals who are owning certain parts of our forecast model all the way through to you know uh individuals on our bisops team that are automating their day-to-day um sort of like you know how they're actually going and running um uh you know competitive benchmarking and pricing studies. So it's like uh I actually think that you know it's the it's the one uping of each other that actually enables you to create an organization that wants to better one another. Um so you know I I would feel uncomfortable saying there's one individual um because I truly do think that this is like a a team that's um trying to do this together.你认为这更多来自于在某一领域拥有深厚专业知识或资深经验的人，还是普遍发生在全员身上？这其中有多少是由你以及你自己亲手实现的自动化所驱动的？
英文原文
>> Do you think it tends to be people who are more expert like who have more expertise or experience in an area or is it really across the board and and how much is it coming from you and the role that you've automated yourself?在早期阶段，由我来展示“如果我都在尝试以一种全新的方式做事，这就是向你们传递一个信号，即我期望你们也这样做”，这确实很有帮助。但在很大程度上，我认为关键在于对你试图提取的数据集有深刻的理解。能够走得更快的人，往往对这些数据的拼接方式了如指掌。当你理解了数据架构，你就能像架构师一样思考：如何以合乎逻辑的方式将其打通？在哪些环节模型是在做确定性的决策，而在哪些环节是在进行推断？在哪些任务中允许这种推断发生，而在哪些任务中绝对不允许？其他人就会从中观察并学习。我们建立的展示节奏、内部黑客马拉松，以及引入外部专家的机制，都推动了这种学习。我经常与其他财务负责人交流，并尽量将外部见解传递到内部；我的管理团队也在这样做。大家都在相互学习，这非常令人欣喜。
英文原文
>> Yeah, I I mean I think like early days it was helpful for me to go and show folks like hey if I if I'm coming and trying to do things in a different way like hey that should be your signal to you that you know my expectation is that you are as well. Um but I think you know in large part um I think the things that are important is have a solid understanding of the data sets that you're trying to pull from and the folks that have been able to move more quickly have like an understanding of like how these things are stitched together. Um, and when you understand things that and how they're architected, you can better think as like an architect of like, okay, how do I actually bridge these things in a way that's logical? Where do I know where the model is actually making a deterministic decision versus inferring? Um, and what tasks like is it okay for that to happen? What tasks is it not okay? Um, and I think then, you know, others kind of pick up and learn from that. Um but uh you know I I I do think that like the the rhythm that we've created, the showand tell rhythm, the you know the internal hackathons, this ability to bring in other experts. We do a lot of like I do a lot of conversations with other finance leaders and I try to bring in some of that intel internally so that it can get disseminated and I think my leaders are also doing that too which is really cool to see. Um I think everyone right now is trying to learn from one another which is which is really cool to see.太棒了。
英文原文
>> Nice.自建与采购的权衡，以及人才吸引力
在我与诸多 CFO、财务总监甚至基层员工的交流中，一个普遍的主题是：大家发现工作重新变得有趣了。过去你要花两三周甚至连续不断地进行对账或结账；而现在，人们真正将目光转向业务本身，能够提炼出商业洞察。大家对 AI 及垂直解决方案带来的可能性感到兴奋。我想回到刚才关于“外购”的话题：我前几天和一家超高速增长公司的负责人交谈，他们表示因为跑得太快，必须直接购买现成方案。你是否有过这种感受，觉得自建比购买现有方案的周期更长？
英文原文
>> Yeah. I feel like one of the common themes that coming up in my conversations with CFOs or uh directors of finance um or even just individual contributors is that they're finding it's really fun to be working again because before it was like you're reconciling something for 2 weeks or 3 weeks and you're closing the books for two or 3 weeks or continuously and and now it just feels like people are really turning their eyes towards the business and are able to surface insights business um and all of that. Uh there's just a lot of excitement around uh what's possible with AI, what's possible with vertical solutions. Um one of the things I'm I'm I'm curious to go back on the um buy uh scenario. Um I was talking to one of our hyper growth companies the other day and they said we just have to buy everything off the shelf. We're moving so fast. Do you ever feel that way? Like maybe a building is a longer path than just buying one of the solutions out there.对于核心的记录系统（systems of record）以及那些绝对不能出错的环节，我们确实会去采购现成方案。而我们团队觉得非常有意思的领域，正如你所提到的，是工作流自动化——在日常工作中，哪些环节几乎是机械无意识的？我该如何去强化日常工作，从而重新规划时间分配？我完全赞同你表达的观点：工作一直都很有吸引力，但现在变得更有趣了，因为你能以更快的速度去探索。团队里有人来找我说：“天哪，我现在真的觉得工作太有意思了。”我认为我们制定的宽松政策以及给予大家自由探索的空间，正是某些员工表态“我想留在这里，我真的很喜欢现在做的事情”的原因。我不知道其他公司是否充分意识到，提供前沿技术的使用机会其实可以作为一种招聘手段。自从听到团队这样的反馈后，我至少在招聘对话中开始强调这一点。尤其是当你尝试吸引非科技行业的人才时，他们会觉得：“天哪，这是我进入一个全新世界的大门；如果我能在这里被赋能去学习，那真的太酷了。”
英文原文
>> I think for like the core systems of record solutions and like the things that like we just can't mess up like yes we're going and buying solutions for that. The areas where I think our team is having a lot of fun it's the it's kind of where you're going it's like the workflow automation like what are the areas where if I think about my day like I'm almost like not thinking um and how can I go and augment my day-to-day so I can you know go and re orchestrate how I want to spend my time. Um, I like I also, you know, fully fully agree with the sentiment that you're sharing that like the work has gotten, you know, the work the work is always going to be interesting, but the work has gotten more fun because you can explore things more quickly. Um, I've had folks on my team come up to me being like, "Holy like I actually am like having a ton of fun." And I actually think that like what the the policies that we created and the freedom we've given folks to explore are the reasons why like some folks are like, "I want to stay here. I really really like what I'm doing." Um, and I think, you know, I don't know if folks are actually treating uh access technology as like a recruiting vehicle as much as they could or should. Um, you know, I've at least started to do that in the recruiting conversation since I started to hear that from the team. Um, especially when you're trying to pull from folks who are not necessarily in technology. They're like, "Oh my gosh, this is like my my front end or my front entrance into like an entirely new world and if I'm empowered to learn, like this is really cool." Yeah.就像我们谈论福利待遇一样：加入这里你能获得各种支持；以前的雇主可能不允许你使用这些工具，而现在你能够接触到所有这些模型和其他工具。
英文原文
>> Like we talk about perks and benefits. You get X Y and Z if you come to work here. They might your former employee employer doesn't let you available and now you also get access to all these models or other tools.不仅如此，还在于你能让员工看到身边的同事正在构建什么。你学到的大部分东西来自于对他人成果的潜移默化。如果你感觉到自己加入的组织不仅嘴上说提供这些工具，而且大家都在实际动手构建，并且在面试过程中就能切身体会到这一点，这才是最吸引人的地方。
英文原文
>> Yeah. Yeah. And I I think it's not just that, but then it's also like if you if you have folks understand what are other people building around them. Um and so much of what you're going to learn is through osmosis of what others have built to. And so if you get this feeling that like you're joining an organization that's not just saying that they're that they give access to these things, but they're actually building and that you can see it and you can feel it through the interview process. Um those are the places I'd gravitate to.从数据对账转向战略决策
2020 年，我和我的合伙人 Angela 写过一篇文章，探讨 CFO 为什么需要现代化工具。我们在文章中画了一个金字塔图，指出财务团队将绝大部分时间耗费在底层的数据对账上，而在最顶层、耗时最少的部分则是战略财务或战略性思考。我想知道，你和你的团队是否真正迎来了时间的解放，不仅是能做更有趣的事，而且能够更具战略性、参与更多商业决策？这对你的角色产生了怎样的改变？
英文原文
>> So uh in 2020, my partner Angela and I wrote this piece on how CFOs needed modern tools. We had this pyramid diagram in there which was essentially say saying that the finance team was spending all their time on this like data reconciliation layer and at the top which was the smallest amount of time um was being spent on strategic finance or like strategy questions. Um, I'm curious if your time, you've, you know, actually seen your time, um, and your team's time unlock, not just on doing things they enjoy more, but like being more strategic or being able to weigh in on more decisions or like how has that changed your your role?我感到很幸运，随着我在公司的成长，战略层面的工作占据了我越来越大的精力。当我思考什么能让其他人感到振奋时，我也意识到，为了让我拥有这种空间，在组织内为其他人创造同样的空间是大有裨益的。虽然日常或每周的固定流程依然要做，但如果我能让我们分管 GTM 业务的财务子团队（比如负责自助服务业务的团队）迅速从上周或上个月的数据对账中解脱出来，转而专注于我们正在进行的业务实验、观察到的趋势以及我们应该追加投资的方向，这对他们来说要引人入胜得多。他们现在确实能更多地投入到这些工作中。很多细节至关重要：比如如何自动生成一套外观和结构完全一致的材料，从而把时间集中在批注、发现和业务理解上。我们看到这方面变得越来越快，质量也不断提升。负责该团队的主管甚至在全员会上展示了他如何在 Figma 中制作一组插件来快速进行此类构思。我希望我的团队告诉我，他们正在将越来越大的时间比例投入到这类工作中。但你始终需要应对业务的复杂度，平衡好业务复杂度与预留给其他战略事项的时间。
英文原文
>> I feel fortunate that like you know as I've grown within the company that like that has taken up more and more of my role and then as I then think about like what what is exciting for others like I also recognize like for me to have that there's benefit in me creating that same space for other people within the organization. Um so yes there are things that like we'll have to do as daily rhythms or weekly rhythms. Um but in the same breath like if I can get like take our take our we have a finance organization that's split in one of the one of the pillars is go to market. If I can take the team that's focused on our self-s served business and immediately get them out of reconciling last last week or last month or whatever it is and then very much more focused on what are the experiments we're running, what are the trends that we're seeing, where are the dollars that we should like what are the additional dollars that we should be spending on like that is way more engaging to them. Um, and I think they're actually able to do more of that now. Like it's it's like such it's some of the little things here matter so much. It's like how can I go and automate the creation of a set of materials that I know it's going to look and feel the exact same way and you know then focus my time on the annotation, the discovery, the understanding. Um, and we've seen we've seen that start to get a lot a lot faster. Um, and the quality of that improve and you know we even had one of like the I was actually the leader of that team get up and talk about like how he had gone and create a set of plugins to go and you know in Figma to go and idate on this stuff really quickly. Um, so you know, I I I would hope that my team is telling me that they're spending a growing share of that their time on that on that bucket of things. Um, but you're always going to have to chase, you know, the complexity in your business and managing the complexity with like how much time you're able to reserve for for the other pieces.确实如此。在我们投资的 CFO 工具中，我们看到的一个趋势是：产品正从单纯跟踪工作流，转向直接提供结果，并将审计追踪（audit trail）作为产品的一部分。财务团队必须确保一切准确无误，需要能够追溯产品是如何得出某个具体对账结果的，但具体计算已经由系统完成了。这是审计追踪与交付结果之间的平衡。我们的团队确实将更多时间用于做实验等以往需要耗费一个月研究的事情——以前你必须拉上数据科学人员，通过邮件催促各方，把大家聚在一间屋子里；而现在后台运行着 Agent，你可以极快地进行迭代。
英文原文
>> Totally. One of the things that we're seeing in the CFO tools that we've invested in is the product is moving away from you know tracking workflow towards uh providing a result and then part of the product is the the the audit trail. So obviously finance needs to make sure everything's correct. You need to be able to uh track >> where you know how the the product came to um a particular reconciliation for example. Um but then it's done for you. So it's that balance of both having the audit trail and the um and the result. But we are definitely seeing our team spend more time doing things like running experiments and doing things that you used to take like a month-long research study. You had to pull in some from data science and like chase down a bunch of people over email, get everyone in a room. And now you can quickly iterate on this because you can um yeah, you can you have agents running in the background.甚至在会计端也是如此，比如波动分析（flux analysis）。作为会计人员，你花很多时间逐期审查总账科目。如果你能利用工具——市面上有很多工具能给出最初的变动分析，为他们指出一系列可以跟进的线索，甚至有些结果立即可用，那就太好了。然后他们就可以把时间投向别处。我完全赞同你的大方向：那种初稿、最初的预期和线索，完全可以通过工具呈现。这与我刚才提到的差旅费用审查一致：工具为团队提供了可供深挖的线索，并赋予我们足够的直觉去追查下去。
英文原文
>> And it could even it's even things on like the accounting side like flux analysis, right? like you know you're spending time as an accountant going through and trying to reviewing the GL accounts for like period over period. If you can leverage a tool and I know there's a whole bunch of tools now out there that give you that initial sort of like stab at what has changed that gives them like a set of things that they can go and chase. It gives them like an initial set of things and maybe it's right right away and I'd love for that to be more and more true. Um and then they focus their time elsewhere. And so I I I fully expect that like the spirit of where you're going like that that like um initial draft that initial sort of set of expectation that initial sort of idea of what to go and um and and pull the thread on um can can be surfaced uh via tools. It's kind of where I was going when I was talking about that T& piece, right? Like it was giving our team sets of threads that they can go and pull um and then us enough gut to go and and chase it too,没错，你拿到了数据，现在有了时间去解决那些二阶问题，看清背后的真实情况。
英文原文
>> right? you have the the data and then now the time to go, you know, address those second order questions and see what what's behind uh what's what's going on.我在想，在我们的成长过程中，“对账”（reconciliation）曾是一个非常消极的词，伴随着许多糟糕的回忆，以及耗费数周乃至数月去排查问题的痛苦；而现在它只是日常的一项普通操作，甚至变成了一个偏积极的词，因为借助现代工具和 AI，你可以更轻松、更快速地解决它。
英文原文
>> Yeah. I was thinking how um when we grow up the war reconciliation was a very negative word is associated with a lot of bad memories and a lot of weeks and months spent on trying to chase something where now it's just something you do or maybe it's even a positive word because you can uh you can solve it much easier quicker with the modern tools and with AI即使未必能完全定位出错误所在，但能更快地推进到这一步；然后你依然需要与人交流，因为财务规划与分析（FP&A）始终是一门关乎人的学科。
英文原文
>> and and find maybe not exactly where the errors are but just get there so much quicker and then you still need obviously to go talk to people as FPNA will always continue to be the关乎人的学科，是的。我认为沟通、建立关系以及施加影响力的能力，曾是我低估或没有意识到对这个职位有多么重要的特质。最初走入这个岗位时，我可能觉得自己只要做全场最聪明、能搞定一切的人就行了；但后来我意识到，无论如何这都是一项团队运动。你个人能推动的改变是有限的，你必须让其他人与你达成共识。如果你能有效做到这一点，你就能成为一名极其优秀的领导者。这些才是真正的核心能力。刚才你问到招聘筛选的标准，如果我能通过测试感受到你具备强大的号召力、能赢得他人的信任，这种品质是无价的。
英文原文
>> the people the people discipline. Yeah, I I you know, it's one of those things I think I underplayed um uh or I under I I didn't realize how important communication um you know, relationship building uh your ability to, you know, go and influence was to this role. Um, you know, I I probably walked in feeling like I could, you know, be the smartest person in the room and figure it all out, but then I realized like, you know, it's a team sport regardless. And you can only go and affect so much change and you then really do have to go and um get other people to to be on the same page with you. And if you can do that effectively, I think you can be a really effective leader. But, um, you know, those are sort of like the I think the true skills. If I can you you're asking earlier about like you know, what am I screaming for? or if I if I can test and get a sense that you're like you you have this ability to build really solid followership, you have this ability to go and build trust with others, like that's so invaluable.未来 12-18 个月的演进与 Excel 的地位
这顺理成章地引出了我接下来想聊的话题：在未来的 12 到 18 个月内，财务团队、该角色以及整体职能会发生什么变化？不知道你们对此有什么看法？或许可以先从团队的组织架构谈起。
英文原文
>> That's a a good segue into what I wanted to chat about next, which is what do we think changes in the next 12 to 18 months uh for the team and the role and the function overall. Um I don't know if you guys have any points of view on on what might change. Maybe let's start off talking about how your team even is organized.我希望达到的一种状态是：只有当业务复杂度的增长速度超过了支撑需求时，我们才增加人员。我认为很多日常的业务运作环节，现在完全可以向相反的方向演进，不再需要单纯为了跟上节奏而增加人手。对我们而言，由于不断推出新产品、收购团队和业务，业务复杂度仍在持续超越我们实现完全效率提升的步伐。但不可否认，我们在逐期变得更加高效。
英文原文
>> Yeah, I mean I I would love to be in a spot where um you know we're only adding individuals if like the complexity of the business is growing faster than than the than the needs to support it. Um I do think that we're in a spot now where like you know a lot of the the regular um sort of business rhythm pieces. I think you can start to move in the reverse direction versus needing to add people in order to keep up with it. Um you know I think for us we're going to continue to fight it because we're you know adding new products, we're acquiring teams and businesses. we're, you know, continuing to evolve what what we're doing. And so it's always more, I think, in that world, the complexity there is still outpacing, you know, where where we're able to truly drive, you know, you know, complete efficiency. Um, but we're definitely like, um, we're definitely getting more efficient, you know, period over period.你认为这种效率提升会遇到瓶颈，还是说会有越来越多的流程实现完全的端到端自动化？
英文原文
>> Do you think that the efficiency, do you think it'll like plateau like you've got the tools to do what you need to do or do you think that this more of these processes will just completely get automated end to end?我希望越来越多的流程能够完全自动化。或许我会从那些完全确定性的环节入手：如何向系统输入足够的信息，从而让我对输出的结果感到完全值得信赖。对于关键业务系统而言，要达到随时可供审计的输出标准，要求是极高的。如何让审计师感到放心，也将是一场有趣的对话。但我并不排除这种可能，我认为随着时间推移我们终将实现。正如我刚才所说，如果我们面临的业务复杂度保持不变，一年后我们可能不需要相同数量的人手；但这必须建立在你更善于在日常运作中建立更高一致性的基础之上。
英文原文
>> I would hope that more and more can get completely automated. I think maybe I start with the pieces that are like truly deterministic and like how can I feed it with enough information so that I can feel that the output there um you know is completely completely trustworthy. Um I think getting something to that like audit ready output piece um uh for a for a system that's like truly business critical is extremely high. I think it's also going to be you know an interesting sort of conversation with the auditors on what gets them comfortable. Um, but I don't I don't I don't kind of take it out of the picture. I do think we'll get there over time. Um, I think, you know, like I was sharing, you know, as long as our business comp if our business complexity was like staying the same, I think we probably would wouldn't need the same number of people a year from now. Um, but uh, you know, I think we only get there because you're doing a better job of finding these places and rhythms where you can um you can build more consistency in in the way that you're operating.Excel 受到威胁了吗？
英文原文
>> Is Excel under threat?没有，老兄，那是我的“爱的语言”。就像我刚才说的，我依然认为案例研究非常重要，因为你将它作为一种工具来建立直觉。人们最终构建出来的很多产出依然是模型。那个模型可以托管在某个地方，或者采用其他形式，但你需要这种工具赋予你深化理解、亲自动手调试的能力。这就像我们在 Figma 中看到的魔力：你可以通过自然语言 prompt 让人们获得最初的草稿，但有时亲自动手进行微调修改要容易得多。以调色盘为例：在设计界面时如果你想更换背景，你可以坐在那里向 Agent 发指令“把背景从红色改成绿色”，或者你可以直接在调色盘上滑动，感受各种色调并获得对实际效果的直观感受。我对 Excel 也是同样的看法。这是我建立信任和理解的方式。未来或许会有更好的工具，但对我而言，要让我彻底放弃它真的非常非常难。
英文原文
>> No. It's like my love language, dude. Like I I I think in the same way I was describing how I um uh you know I still think that case study is so important because like you use it as a tool, right? Like you use it you use it to build intuition. A lot of the output people are still creating is a model. I mean that model could be you know you could host it somewhere could be something else but like I think you want this tool to give you this ability to refine your understanding um and play with things with your hands. And it's kind of like the magic that we see with Figma, right? Like we can give people the ability to go and, you know, have a have an initial stab in the first draft through, you know, prompting and through natural language, but like sometimes it's so much easier to go and riff and make changes um on your hand. Take a color wheel as an example. You've got like a a a screen that you that you've designed and you're like, "Hey, I want to go and change the background." You could sit there in an agent and prompt it, right? Like go change the background from red to green or I could take the color wheel and you get all different shades of it and you get the real feel of of what's going on. Similarly, I think about that with Excel. It's the way I build trust. way I build understanding. There may be something better in the future, but for me, I think it'd be really really hard to rip away from me.我也是在 Excel 环境中成长起来的，现在仍然用 Excel 来整理思路。如果要进行分析，其空间组织形式依然极具价值。我认为现有的模型在独立创建预测或财务模型方面还不够成熟，那种结构化组织的价值依然存在。不知道你是否也有同感？
英文原文
>> I also grew up in Excel and I still use Excel to like organize my thoughts. Like if I'm trying to run an analysis, I'll it's yeah, it's just even like the spatial um organization of it and and I think the the models aren't quite good enough to like create a forecast or a model quite yet, but I think that organization is still valuable. I don't know if you feel similarly.完全赞同。而且我现在尝试探索的一点是：如何利用这些模型帮助我更好地理解团队整理出的成果？如果你给我一个包含 20 个标签页的模型，我逐一去翻阅各个标签会非常吃力；但如果我把相同的上下文输入给模型，让它帮我提炼出哪些是真正的核心杠杆、哪些是对变动最敏感的区域，这实际上为我去协调和理解全局提供了一个更好的起点。我发现这非常有帮助且极其高效。
英文原文
>> Totally. And I also think, you know, part of the things that I've tried to experiment with now is like how can I use um uh you know, how can I use how can I use one of these models to help me better understand the output that a team is putting together? Like if you give me like a, you know, a 20 tab model, it's going to be really challenging for me to go and like, you know, tab through and thumb through everything. But if I can, you know, pass that same context to through through a model to help me understand, you know, what are the things that really are are are the the needle movers or what are the areas that are more sensitive. um has actually given me like a better starting spot to go and to go and orchestrate and understand things which um I've actually found really really helpful and effective.我们生活的世界也充满了挑战。我前几天和一位大学教授聊天，问他们大学里是如何教授 AI 的。他说他们现在根本不教。他们遇到的挑战之一是：在同一个学期内，因为不断有新模型发布，他们不得不数次修改教学大纲，最后只能放弃。所以他们重新回归基础，教授分析性思维和 Excel 等等。因为如果你过度依赖外部模型，你可能不得不彻底重做你的财报流程。
英文原文
>> The world we're living in is also challenging too. I was talking to university professor the other day and I was asking uh him how do they teach AI at the universities? Well, they don't right now. And one of the challenges that they they said they had a course and within the semester they've changed the syllabus several times because new models have dropped and they just have to completely change how they do it though. So they just sort of give up. So, they're going back to fundamentals again, teaching people analytical uh thinking and Excel and all of that because um I imagine you're in a similar world where if you were relying in too heavily, you're going to have to redo how you do your earnings.新时代 CFO 的核心职责与给初创企业的建议
很多东西变了，但也有很多东西没变。这或许正是你所指的核心。我认为成为优秀财务人员的核心能力依然未变；如何利用手头的工具做出更好的决策，永远是一个动态的目标。就像当年 SQL 对我而言是一次能力的释放，如今 AI 作为工具和能力的驱动力也是全新的突破。因此我认为，只要能持续适应，优秀的财务人员就能拥有持久的职业生命力。
英文原文
>> A lot has changed, but a lot hasn't changed. And I think maybe that's the spirit of where you're going. Um I, you know, I think that the core capabilities of what makes a great finance person are still there. Um how you actually you know leverage the tools at your disposal to make you know better better decisions is you know always going to be this like moving target for you. Um uh like the unlock that SQL was for me is you know now you know net a net net new unlock with like AI as a tool and capability um driver for me. So I you know I I think a great finance person um has longevity if they continue to adapt我们在《你需要一位新 CFO》中强调的核心观点是思维模式的转变：无论是人员本身的更换，还是思维上的进化，财务管理者都必须具备“自动化优先”的思维，根据实际情况自建或采购工具。我想知道，你们认为未来一两年内 CFO 的人才画像会发生改变吗？还是说核心维度依然不变，只是思维模式在不断演进？
英文原文
>> which you know we wrote this piece we said it was titled you need a new CFO the main point was that the mind there's a mindset shift so whether the individual actually has changed it's the that they have to think more automation first either build or buy the tools depending on what's what's available and what makes sense but it's really automation first um I'm curious if you think the profile either of you of a CFO in the next year or two is going to change in any way or to your point it's kind of the same the same dimensions that are important but uh maybe just the mindset continues to evolve.如果要给别人提建议的话，那就是去招聘一个和 CEO 一样具备强烈业务主人翁意识与运营严谨度的人。他们渴望胜利，想方设法去赢；他们拥有一套工具，并理解如何通过损益表（P&L）去推动变革。你需要的是一位真正的商业领袖，而不是像我之前说的那种“交通警察”。当然，扎实的会计基本功和主持规划流程的能力是必不可少的基础；但真正让人脱颖而出的，在于你是否理解业务、能否推动变革，以及能否将复杂的商业模式简化提炼，让全员朝着同一个方向发力。
英文原文
>> If I were to give advice to anyone it's like hire someone who feels a similar amount of business ownership and u wants to be the you know and operational rigor as you as the CEO does. Um they want to win. They want to win no matter which way possible. They have a set of tools that they're going to use and an understanding of a P&L to go and orchestrate that change. Um, but you truly want like a business leader versus like, you know, someone who um, you know, like I was sharing earlier, like a traffic cop, right? Like that that is just not and yes, I think there's a lot of there's a lot of things that you're going to need to have as like common place and solid understanding of like all the accounting fundamentals like you know your your your ability to actually go and orchestrate a planning process. But like I actually think the things that like make people truly unique in this role, it's like do you understand the business? Can you go and make make change happen? And could you get can you simplify the understanding of how the business model works so that others are all running in the same direction?无论对于像 Figma 这样的上市公司，还是正在起步、拓展 GTM 业务的种子轮或 A 轮初创企业，创始人真正需要财务团队提供的正是对业务的深度见解。这些洞察如今更容易获得，但核心依然回归基本面：你是否理解客户、理解用户行为，以及理解由此产生的输入与输出指标？这始终是我们的期望所在。
英文原文
>> Yeah, and I think it's true uh whether we're talking about a public company like Figma, we're talking about a seed stage or series A startup that's starting to scale their go to market and getting traction is that that's what the founders and the CEOs really need from finance is that extra business opinion opinion on the business um and insights that uh are easier to get but also still you go back to the basics like customers do you understand the customers do you understand the behaviors do you understand the metrics that come from that inputs outputs I That's that's will always be the same in what we expect.我想补充的一点是创造力。不是去随意捏造花哨指标的那种创造力，而是能否从略微不同的视角去审视业务，以此推动变革并提出旗帜鲜明的观点。如果能够提出这些构想与不同的思考维度，财务组织完全可以成为推动公司内部战略制定的极具影响力的层级。在投资人和董事会层面，我们确实看到了更深入的分析和快速的迭代，以往董事会前后漫长痛苦的往返解答过程被大幅缩短了。
英文原文
>> Yeah. Maybe the thing I'd add like I I I do think like you know creativity is important. Not like you know go and create like a fun funky set of metrics creative but like how can you actually go and think about your business in a slightly different light. Um and use that as like the way that you go and um drive change and be opinionated. Um and I you know I actually don't know how you know I try to do I try to do this. I don't know if I'm doing it as much as I, you know, could. Um, but if I come up with these ideas, if I can come up with different ways of thinking about things, I actually think that, um, I think some great ideas can come from your finance organization. And if you really set them up well and right, um, they can be they can be this like layer within your organization that's just as impactful. Um, and uh, really driving a lot of the internal strategy as well. I will say at the investor and board level we are seeing a much deeper set of analysis come through or the rapid iteration you know before it used to be oh we have all these questions at the board meeting or before and then it's a you know the team needs to run and cycle through that that process has just shortcircuited and hopefully is less painful finance team side最后一个问题问 Ivan：当投资组合公司或普通企业向你请教在自动化方面该从何入手时，你会给他们什么建议？
英文原文
>> um last question for you Ivon uh as our portfolio companies or just companies in general come to you and say where do I start in terms of automation What do you give them as advice?我认为起点始终如一：从数据的完整性与真实性（integrity of data）入手。你必须搞清楚自己是否真正拥有良好的数据访问权限，是否在进行数据追踪。很多初创企业早期根本什么都没追踪，你必须先搭建起来。这往往会与新招聘的第一个财务人员的当务之急产生冲突——他们可能想着赶紧结账、发送董事会简报等；但实际上，在拥有坚实的地基之前，你甚至无法做好这些基础工作。有时候你必须重新搭建甚至重写逻辑。很多新加入的人在几周内就会意识到，营收指标的计算方式可能根本不符合该商业模式的规范。
英文原文
>> Uh I think it's kind of the same place where we always started. You start with the integrity of data. You have to figure out if you actually have a good access to data if you're tracking things. Many many of the times especially early stage companies they don't even track anything. And so you have to set that up and that often goes in uh against the priorities that maybe the first finance hire has like I got to close my book sometime got to send the board update all of that. Actually no like you can't even do that until you're on the solid foundation. You might sometimes have to set it. You might have to rebuild it. You might have to change things. We've had a number of people that came in especially now very quickly you can realize within several weeks that maybe the revenue is calculated um in a way that's not how it should be calculated for that business model.早期阶段见过的“富有创意”的 ARR 计算方式，比我以往见过的都要多。
英文原文
>> More more creative uh ARR at the early stage uh than I've ever seen before.确实如此，而且变化非常快。但这必须始于数据，然后建立报告的严谨性，定义指标，最后提炼出业务叙事（story）。随着目标瞄准 IPO，保持口径一致的叙事能力至关重要；而这种叙事要求现在已经提前到了 A 轮和 B 轮融资中，不仅是描绘产品愿景，更是用具体数据展现商业前景。AI 让这一切变得更容易，但也因为过度思考而变得更复杂。Ivan 和我过去几年一直在举办 CFO 和财务总监晚宴，大概在九个月前，我们看到了真正的市场拐点：大家从最初讨论“用 AI 生成税务备忘录或入职文档”，转变为“利用 AI 处理核心数据对账并自动化大量工作流”。这确实是一次阶跃式的转变。管理层也意识到必须迎头赶上，否则就会在组织内显得落后。这又回到了你之前提到的在黑客马拉松中获奖的经历——几年前财务人员绝对不可能成为自动化黑客马拉松的获胜者，看到现在的彻底转变真的很酷。
英文原文
>> Sure. For sure. The it's changing fast too. But um but it starts with data and then um based on that uh you really kind of create the reporting rigor. you create the metrics and then you create the story. I think one of the things too Premier was um especially you've been on that journey as a CFO like storytelling becomes a really really important once you have your eyes set on the IPO and you have to be all telling the same story. I feel like that has moved early to earlier and earlier stages because you expect to tell the story now in fundraising for Sirius A, Sirius B and um not just the story of what the product vision could be but story with the numbers and what the what the business could really be like and that's gotten easier but it's also gotten harder because I think a lot of people overthink that with AI and you know Ivon and I have been hosting these CFO and controller dinners for I don't know a couple years now and we saw this real market change I'd say be nine months ago when it went from people saying, "Oh, yeah, I'm generating this tax memo or this onboarding dock with AI to saying, okay, yeah, I can actually I'm doing the core uh the data reconciliation um and automating many of the workflows." And it's really been a step uh function change. And if anything um the leaders are saying, I'm I I need to do this otherwise I'll be perceived within my organization as uh behind everyone else. This goes back to your point around you winning the hackathons. Like finance was never the one that was going to be winning the automation hackathons, you know, I don't know, a couple years ago and it's it's pretty cool to see that's totally changed.是的，我非常喜欢这部分。我们公司内部有“创客周”（Maker Week），也就是我们的黑客马拉松。在早期阶段我坐在那里完全不知道该如何参与；而现在只要有一点空闲时间，我就能冒出无数个想法，我相信我的团队也有同样的感受。
英文原文
>> Yeah, I I love this part of it. And I um I would we have like an internal maker week and it's our internally branded like hackathon and um you know I would always sit there early days and not really know how to participate and now I'm like if I had like a spare minute I think I would have a million different ideas and I'm sure my team has feels very similarly.好的，Praveer，非常感谢你的加入。探讨这些话题并直接从经历过这场自动化转型的从业者口中获取洞见，真的非常有趣。也感谢 Ivan，这是一次非常愉快的对话。
英文原文
>> Well Pier, thank you so much for joining us. This was really fun to talk about and uh understand directly from someone who's gone through this journey of automation over over your career and thanks Ivon and um this was a really fun conversation.好的，感谢邀请。
英文原文
>> Yeah, thanks for having me.好的。