从零到 100 万美元以上：深入 Tiiny Host 与 Elston Barretto 的自举创业之旅

但这直接提升了我们顶级套餐的收入，效果非常好。

But it's instantly just boosted revenue of our top plan, which has done really, really well.

然后我们把这项设置推广到了低阶套餐的非工作邮箱用户，也就是那些太阳能套餐的用户。

And then we introduced it to non work emails on the lower plan, so the solar plans.

我们一共有三个套餐：基础版、中级版和高级版。

We have three plans like bottom, middle, and top.

所以，这招真的特别有效。

And so, yeah, it's been super effective.

用户该有的功能都有了，也都挺满意的。

People have got everything they needed and just have been happy with it.

抱歉。

Sorry.

我真的没太明白。

I really don't understand.

所以我来到 Tiny Host。

So I come to Tiny Host.

是的

Yeah.

我用自己的文件做托管。

I host my file.

我用公司邮箱登录。

I sign in with a business email address.

是的

Yeah.

你们给他们提供反向试用。

And you give them a reverse trial.

就是不收他们钱，还是得让他们填信用卡信息？得填卡。

So you just without even charging them or do they have to put their credit card in You put your card in.

七天

Seven days

对

Yep.

免费试用七天。

Seven days free.

对。

Yep.

那这个转化流程是怎样的？

And then what does the funnel look like?

就是说，有多少人一上来就取消了？

Like, how many people cancel at that, like, straight away?

又有多少人最终转化了？

And how many how many people convert?

我觉得，只要看到弹窗的人，有一半都会选择试用。

I think one in two people convert to the if they see the pop up, they convert to the trial period.

所以，大概就是这样。

And then really so okay.

有人进来，上传文件，用的是企业邮箱，然后就会弹出提示：享受七天免费试用。

So someone comes in, they host their file, they've got a business email address, they get a pop up saying, have a seven day free trial.

是的

Yeah.

有一半的人会输入信用卡信息。

And half the people put their credit cards in.

对。

Yeah.

我得再确认一下，但转化率确实非常高。

I need to double check that, but it's very, very highly converting for sure.

真厉害。

Wow.

明白了。

Okay.

那这就真的很不错了。

And then so that's really that's really cool.

一半的人填了信用卡，接下来会发生什么？

So half of put their credit cards in, and then what happens next?

所以到这个时候，很多人会取消吗？

So a bunch of people cancel at this point?

我认为大约有30%到40%的人会在试用期结束后继续使用。

So I believe it's it it varies between 30 to 40% of people continue on past the trial.

太惊人了。

Amazing.

显然，搜索引擎优化在你的成长过程中扮演了非常重要的角色。

And so obviously, SEO is a huge part of your journey.

我很想了解一下，对于不太熟悉SEO的人来说，你们一般是怎么做SEO来推动增长的？

I'd love to know just what is your general approach to SEO for growing tiny for those that are less familiar with with using SEO?

是的。

Yeah.

这是一套很多公司现在或过去都在用的标准做法。

It's a standard playbook that I think a lot of companies are using now or have been using.

我不确定在人们搜索方式发生这么大变化的今天，这套方法还像以前那么有效。

I'm not sure if it's as effective as it is now with the whole change in how people search.

最重要的是要做扎实的关键词研究。

But it's number one, having really good keyword research.

比如投入使用HRAS这样的工具，

So looking at, you know, investing in HRAS, which is

HRAS是我们的赞助商，对吧？

HRAS is the sponsor of us, don't they?

是的。

Yeah.

确实值得用。

Really do.

或者用SEMRush也行。

Or SEMRush.

关键是找出搜索量大但竞争难度低的关键词。

And basically finding out which keywords have high volume and low search difficulty.

所以PDF上传就是一个很好的机会。

So PDF upload was an opportunity.

我查的时候，搜索难度大概只有9，但每月在美国的搜索量有1500到2000次。

I think it had like a search difficulty of like nine when I was looking at it, but a volume of like 1,500 or 2,000 people monthly searches in The US.

于是我就做了个简单的事，去了Tiny Host。

And so simple thing, I just went on Tiny Host.

我建了一个落地页，叫tinyhost.com/p-upload-pdf，页面就六七个板块，专门讲什么是上传PDF、什么不是PDF，重点是怎么上传PDF，结合Tiny Host和PDF上传这些关键词，写了800字，加了些常见问题，然后就交给Google慢慢抓取。

I made a landing page called tiny the host slash p upload PDF, and you just have, like, six, seven sections targeting, like, what is upload p like, what is not what is a PDF, but how do you upload a PDF, like Tiny Host and uploading PDF, 800 words on that, some FAQs, and then you just leave it to marinate with Google.

这跟Veed的做法还挺像的。

It's kind of like the the Veed approach as well.

这些更像是针对用户搜索需求做的落地页，而不是普通的博客文章。

So these are more like landing pages for features that people are searching for rather than like blog posts

对，就是这样。

Yes.

没错。

Kind of thing.

落地页的效果真的很好。

Landing pages do really well.

然后还有外链，内部链接，还有其他一些东西。

And then you have like backlinking, then you have internal linking, then you have yeah.

周围还有很多其他工作要做。

A lot of other stuff around it.

并不复杂，但特别耗时。

It's not complex, but it's it's very laborious.

需要很多人协同配合才能跑起来。

It needs a lot of actors working to make it work.

之前我们聊过产品驱动增长和SEO，另外两个是YouTube和Reddit吧？

And then the early days we talked about product led growth, SEO, the other two things, was it YouTube and Reddit then?

对。

Yeah.

说实话，我前一百个用户都是从Reddit来的。

I mean, Reddit was where I got our first 100 users, I'd say.

Reddit非常有效，现在依然如此。

So Reddit is super powerful, think, and still is.

我觉得它特别适合获取非常贴近产品的反馈。

I think it's a really good sounding board for get like, to get really close product feedback.

就是一长串帖子，大家说喜欢什么、不喜欢什么，而且通常二十四小时内就能收到反馈。

So, like, it's it's just a long thread of, you know, what people like about it, what people don't like about it, and it's within, like, twenty four hours you get this feedback.

于是有人问，我能绑定自定义域名吗？

And so people are like, oh, can I connect the custom domain then?

我就只是说，好吧。

And I just like, okay.

他们根本没想到这些是重要功能，比如加个数据分析，或者做点别的。

They didn't think of that was like an important thing or can I add analytics or can I do this?

所以这种视角真的很有帮助。

And so it is really useful for that perspective.

我知道这听起来有点小众，但我之前做另一个应用时，用过一个特别棒的办法——直接在某个特定的子版块上线。

I know a bit tiny, but I also did this really cool trick with, like, another app I built before where I launched on the specific subreddit.

那是个从零开始跑五公里的训练项目。

So it was like a couch to five k program.

我就跟他们说，嘿。

And I said, hey.

我做了一个应用。

I built this app.

能给我点反馈吗？

Can you give me feedback?

这是为社区做的。

It's for the community.

作为感谢，我送了十张付费版的优惠码。

As a thank you, here are, like, 10 free promo codes for the for the paid plan.

当时也没设什么使用上限。

And then there wasn't really a limit on it.

结果第二天，他们就说，不好意思。

And then the next day, was like, oh, sorry.

优惠码基本都用完了。

These are these are basically, like, all done.

我又发了一次。

I post it again.

不过我又给了10个。

But here's another 10.

我这么做了三次。

And I did that three times.

结果连续三天，或者说是三周，我的帖子都上了那个子版块的榜首。

And I got to the top of that subreddit in three separate days or three weeks, whatever.

对吧？

Right?

就因为这个，我一下子拉来了上百个用户。

And that got me like a 100 users just because of that.

是的。

Yeah.

但那是六七年前的事了，那时候的Reddit可不像现在这么混乱。

But this was like six, seven years ago and Reddit is a wild wild west at the moment.

是的。

Yeah.

现在你怎么做才能不被立刻封号呢？

How do you do that without getting banned immediately these days?

对。

Yeah.

你得

You have to be

稍微动点脑筋，讲究方法。

a bit clever about how you do it.

我觉得现在大家都是高度警惕，这也难怪。

I think that people are just on red alert and so as they should be.

但说到底，道理还是一样。

But again, it's the same thing.

就是得真诚、透明。

Like, think you just gotta be honest, transparent.

人们会去翻你的历史记录。

People are gonna look in your history.

对吧？

Right?

这其实是个明显的信号，说明你是不是一直在推销自己的产品？

That's a big telltale of like, are you talking about basically your product all the time?

但如果你分享真实的故事、真实的见解，真正为社区着想，我觉得反而有帮助。

But genuine stories, genuine like, you know, insights and things being held for the community, I think helps.

我觉得，如果你从来没被任何一个子版块封过，那说明你还不够拼。

I I generally think if you don't get banned from at least one subreddit, you're not going hard enough.

而且这一点。

And that.

没错，你确实得被封一次。

Yeah, you have to get banned.

我完全同意。

Totally agree with that.

我们换个话题，聊聊我和萨巴最关心的客户流失问题。

And switching gears to mine and Saba's favorite topic churn.

我们知道你们都特别关注客户流失，不然也不会坐在这儿了。

And we know you all love churn, that's why you're here.

说说看，你是怎么发现客户流失开始变成问题的？如果确实有这情况，后来你是怎么想办法解决的？

Talk to me about how you first noticed churn becoming maybe an issue if it did and like what kind of approach did you take to getting that bad boy down?

是的，这确实是个特别棘手的问题。

Yeah, I mean, it's it's a very difficult problem.

在运营过程中我才意识到，网页托管行业的客户流失率其实非常高，我本来以为不会这样，毕竟你把东西上线了，不就该一直留着吗？

I mean, well into the journey, learned that churn is very high in web hosting in general, which I I didn't think it would be because you put something online and I thought, hey, just keep it there.

但确实如此。

But Mhmm.

像Bluehost这类大公司，定价策略就很奇怪：第一年只要两美元，第二年直接涨到二十美元左右。

There's a reason some of the big plays like hosting or Bluehost have like this weird pricing strategy where it's like $2 a month and it shoots up like, I don't know, $20 a a month the next year.

但我们确实也...

But we yeah.

我们的流失率在任何标准下都不算低。

We we don't have low churn, I say, by by any standards.

我觉得我们永远不可能做到净负流失。

We I don't think we'll ever get to net negative churn.

但我们更关注用户群体的质量，这可能是我们目前看待流失问题最好的方式。

But we think about churn in terms of audience, I think that's probably the best way for us to look at it right now.

吸引那些真正做营销、经营业务、需要更专业工作流支持的用户，而不是单纯用网页托管服务的人，这样留存率会更好。

So like attracting the right type of audience who are doing more like, you know, marketing things or, you know, running businesses and that need more, you know, specific workflows revolve around web hosting is a better retention.

该做的基础措施我们都试过了。

We've done all the basic stuff.

对吧？

Right?

比如退出时给50%折扣、劝你降级套餐、展示使用数据，这些我们都做了。

Like give you 50% discount on exit, try to lower your plan, show you usage, all that kind of stuff.

这些办法有点效果，但根本改变不了大局。

And that kind of moves a little bit stuff, but it but it doesn't really move the needle heavily.

我觉得其中一件大事就是让更多人升级到专业版。

And I think one of the big things has just been getting more people into the pro plan.

对吧？

Right?

最低档的套餐，每月9美元，流失率特别高。

So like the lowest plan, which is nine docs $9 a month is is high churn.

这档套餐的流失率肯定很高。

It's it's gonna be high churn.

用户付的钱越多，流失率反而越低，说来奇怪，但他们是更认真的客户。

The more people pay, the less kind of churn there is in a weird kind of way because they're more intentional customers.

是的。

Yeah.

这其实特别有意思。

Actually, super interesting.

你刚让我想起我16岁那会儿注册iPage账户的事，就是那个老牌主机服务商，对吧？

I didn't you you just took me back to when I signed up for an iPage account at the age of 16, which was like, do you know iPage like OG web host Yeah.

像CPanel那样的界面。

CPanel style.

没错。

Yeah.

对的。

Yeah.

我记得第一年只要两美元，第二年就涨到二十美元了。

And I remember it was like $2 a month for the first year and then $20 a month for the second year.

后来我就转到HostGator了。

And I've just got on Host Gator.

是啊。

Yep.

它默认给我订了四十八个月的套餐。

And it defaulted to a subscription length that was forty eight months.

是的。

Yeah.

你听好了。

Get this.

你听好了。

Get this.

你听好了。

Get this.

只要每月3美元，还送三个月免费。

Get It's $3 a month and there's three months free.

没错。

Right.

之后就涨到每月14美元了。

And then it goes to $14 a month.

我也不太清楚它是怎么运作的，

I don't quite know how

但合约期是四十八个月。

it works but like For forty eight months.

持续四十八个月。

For forty eight months.

是的。

Yeah.

我刚才说的，就是这么回事。

What what I said okay.

所以很明显，这个托管平台肯定在 monetization 方面做了些什么，没错。

So the obviously, the hosting space has obviously done something or doing something Yeah.

通过变现手段来减少用户流失。

With monetization to combat churn.

对。

Yeah.

这基本上意味着，不是产品本身不能满足需求。

Which basically means that it's not like the products doesn't serve the need.

所以问题不在产品，而是在这个品类。

So the product is not the problem, it's just the category.

是的

Yeah.

你看到这个领域里有哪些具体因素导致用户流失率这么高？

What are what are the actual dynamics that you see in the category that lead to high churn?

是因为一次性项目，还是创业者没能坚持下去？

Is it like one and done projects, entrepreneurs not following through?

到底是什么原因呢？

What like what is it?

是的，说到底，我们用户流失最主要的原因就是一次性使用场景。

Yeah, I mean all of that, it's basically The biggest reason we have churn is like one time use case.

我有个项目，做完了，是个活动，之后就再也没用过了。

So I had a project, I did it, it was an event and then I'm done.

第二个原因是有很多独立创业者，他们建网站、做小生意或者个人应用，但之后就再没下文了。

The second thing is you get a lot of solopreneurs, people creating websites or like small businesses, indie apps, and then never kind of went anywhere.

我们最意想不到的流失原因之一，说你可能不信，有人发邮件说：我爸爸发现我偷了他的信用卡，买了Tiiny Host的订阅，现在他要求退款。

I actually we get one of the most surprising reasons of churn is I kid you not, the message I get in the email is my father found out I stole his credit card and used bought Tiiny Host's subscription and now he wants the refund.

天哪。

Oh my god.

我们还遇到过奶奶，还有妈妈来联系我们，说孩子用家里的信用卡订阅了，根本不该这么做。

And I think we've and then we've had the grandma, then we've had like a mother come in and say, my kids signed up with our credit card, they shouldn't have done that.

这些人确实想托管内容，但随后……

So these people really wanna host stuff but then

应该有人资助这些孩子，是的。

Someone should fund these kids by Yeah.

所以他们会成为某类用户。

The So They're gonna be some someone.

然后他们就得取消订阅。

And then they need to cancel.

这是其一，但我确实觉得有一部分托管用户是稳定的，比如我们有个客户是标普500强的房地产公司，三年前我们就为他们定制了方案，到现在还在付费。

So there's that, but but I do think there is a segment of hosting which is like so we have customer, real estate company who's S and P 500, and we've made a custom plan for them probably like three years ago and they're still paying.

他们用这些服务为待售房产创建小型宣传网站，进行推广和分发。

They host like micro sites for like properties when they market and distribute a property.

这是为该房产专门搭建的独立网站。

It's a dedicated like website for that property.

他们还在持续付费，因为已经深度融入了工作流程。

And they're still paying because they're heavily embedded in the workflow.

这是他们的市场部门在操作。

They have this they have it's the marketing department.

他们直接用公司信用卡注册了服务。

They came and put their corporate card in.

随着卖出的房产越来越多，他们会不断创建新的网页，一直需要这项服务。

And they're just as properties, you know, as they sell more properties, they'll create more web pages and they continuously need that.

这种使用场景，你真的得好好跟进。

So it's the use case kind of you you need to follow.

对吧？

Right?

企业客户在多大程度上推动了你们的收入？

To what extent is has enterprise contributed to your to your revenue in your Yeah.

没多少。

Not a lot.

说实话，我去年底试过一把做企业市场，还打算雇个销售帮忙。

Honestly, I I tried to do the whole enterprise stuff maybe towards last end end of last year, tried to, like, hire a salesperson to kind of help us out.

但我发现我们根本不是做企业市场的料。

But, you know, I realized that we're not built for enterprise.

我们就是个现成的成品软件。

Like, we're off the shelf product.

企业那套来回扯皮的合同太麻烦了，我们靠这种模式反而做得挺顺的。

It's the enterprise stuff is just, like, back and forth contract stuff, and we're pretty we're going pretty well with that kind of stuff.

我最接近企业客户的一次，是美国有所大学想用我们，但他们非要我们填什么W-8BEN表格之类的税务文件。

I mean, the biggest the closest thing I went to we got to enterprise was then there was a university in The US who wanted to use us, but they want us wanted us to sign a w eight Ben form or some kind of tax form.

我就跟他们说，这种流程我们真不搞。

And I said, look, we're not gonna do that out of a process.

这完全是企业级合同的套路。

It's pretty from an enterprise contract.

这一个月才两美元。

Like, this is gonna cost $2 a month.

他们拉上了好几个部门的人，我把合同发过去，他们马上就同意了。

And they looped in all these kind of people from different departments, and I sent them the contract, and they were like, yes.

是的。

Yes.

对。

Yes.

还真有人签了字。

And someone actually signed it.

我当时就想，行吧。

And then I was like, okay.

挺好。

Great.

这事儿还挺简单的。

Like, this was easy.

然后我发了账单，结果有人回过来说：这玩意儿是啥？

And then I sent them the invoice and then someone came back and was like, what the hell is this?

他们就说

And they were like

天哪，不会吧。

Oh, no.

对啊。

Yeah.

我们不想付这笔钱。

We we don't wanna pay for this.

我就说：你们不是刚签了合同吗？

And then I was like, well, you just signed the contract.

于是我回去问：那你们到底想怎么办？

And so I came back and I was like, well, what do wanna do?

最后我们只好给他们升级到一个更贵的定制方案，但那根本不是原来的合同。

And then we had to get them into like a a higher custom plan, but it wasn't that contract in the end.

说实话，如果你是靠自己起步的，企业级方案真不值当。

I I honestly, I think enterprise, if you are bootstrapped,

确实不值。

it's not

真的不值得。

really worth it.

等你做到一定规模，比如年收入五千万、一亿、两千万美元的时候，再考虑也不迟。

Like, maybe once you hit like a decent scale $5.10, 20,000,000, you can start thinking about it.

但要是

But like

现在还太早了。

It's too early.

如果你团队小，靠自助服务就能做得风生水起，那就继续猛干下去。

If you've got a small team, you're doing self serve and you're killing it, just keep killing it.

因为一旦分心去搞别的，反而会偏离重点。

Think because if not, it just takes your eye off the ball.

你的生意里有什么东西正在起作用，对吧。

Like something's working in your business Yeah.

继续坚持下去。

Keep it going.

没错。

Yeah.

加大投入。

Double down on it.

别急着搞销售合同，不然你就得专门为他们开发功能。

Don't mess start messing around with sales contracts because then you have to start building product features that are specifically for them.

我觉得，比如根据你之前讲的那些孩子用爸爸信用卡注册的故事，我猜现在企业用户还没占到百分之八十。

And I think like if for example, and I assume not based on some of the stories you've told us about kids signing up with dad's credit card, but I I assume that this is not 80% businesses are using this within an enterprise currently.

没错。

Correct.

是的。

Yeah.

我觉得大概是四成吧。

We I think it's probably four yeah.

可能有四成的企业，不过呢。

Maybe 40% businesses, but yeah.

用量并不大。

It's not heavy.

是的。

Yeah.

主要是在付费方案里。

On paid plans,

我猜也是。

I assume.

是的。

Yeah.

是的。

Yeah.

对。

Yeah.

也就是说。

Which means yeah.

没错。

So exactly.

所以你的普通用户群体根本不是特别偏向企业用户。

So your general user base is not even highly skewed towards business.

如果你想想monday.com这样的工具。

If you think about a tool like monday.com

是的。

Yeah.

非常偏向企业用途。

Like Very business focused.

吉米这么个普通人，怎么可能用他爸爸的信用卡去注册monday.com呢，你说是吧？

Jimmy, lovely Jimmy, is not gonna take dad's credit card to sign up for monday.com, you know?

是的。

Yeah.

你得对自己诚实。

You gotta be honest with you always.

时间一点都不能耽误。

No time not.

而且没错，这是完全不同的团队，完全是另一回事。

And yeah, it's it's it's a different team, it's a different ballgame.

我最近才意识到，你们未来的赞助商Ahrefs居然在没有一个企业客户的情况下，做到了一亿美元的年经常性收入。

And I also realized recently that your future sponsor Ahrefs got to a 100,000,000 ARR without a single enterprise contract apparently.

所以，不靠企业客户，也能发展得很快、很远。

So it's possible to go pretty fast, go far without it.

你是什么时候意识到的？嗯，这事儿真要起来了，规模不小了。

At what point did you realize, okay, this is actually gonna you know, this is getting quite big.

这下是真的要成气候了。

This is going properly.

也许我该辞职了。

Maybe I need to quit my job.

那时候是哪个阶段？

Like what stage was that at?

那你具体做了什么，才达到下一个阶段的？

And like what would what did you have to like do to get you to like that next level?

我是在比较后期才这么做的，跟大多数人不一样。

I I honestly did it in a later stage and I kind of most people wouldn't.

我记得在砖匠酒吧的时候，大家都说，你得辞职，等收入到四五万的时候就该全职干了。

I remember being in the Bricklayers and people like, you need to quit your job, you need to quit your job when you reach like four or five

砖匠酒吧，肖迪奇最棒的酒吧。

The Bricklayers Arms, the greatest pub in Shoreditch.

是的。

Yeah.

我清楚记得萨博，还有蒂姆，他们都跟我说，必须全力以赴。

And definitely I remember Sabo, maybe Tim, he's like, need to go all in.

我当时想，是啊，但还没准备好。

I was like, yeah, will, but I'm not kinda ready yet.

后来我辞职了，主要是因为个人财务方面的考虑。

And then, I mean, I I quit at a point honestly was for personal financial kind of things.

比如我买了房子之类的东西，那时候就觉得该辞职了，诸如此类的原因。

Like I ended up like buying a house and stuff and then I wanted to quit and that kind of thing.

我辞职的时候，大概有八千美元的月经常性收入。

But I quit, I think, around 8 k MRR dollars Mhmm.

对，这在伦敦已经远超糊口的水平了。

Yeah, which is more than ramen profitable in London.

但当时我面临两个选择，这才是主要的原因。

But I the the main reason it was looming was I had two two things two options, basically.

我可以继续把它当副业，让它一直这么挂着，花掉我大量时间，但继续上班。

I could even leave this thing on the side, which has taken a lot of my time to just, like, be on the side forever and just continue working.

或者我干脆把它变成全职事业。

Or I could say, let me make this my full time thing.

当这两件事——我的主业和副业——变得压得我喘不过气时，我就想，我必须全身心投入了，因为我一直都希望做这件事。

And when it became, like, two overwhelming things, like my main job and doing something on the sides, then I was like, I I need to go all in because the future is has always been like, I wanna I wanna do this.

我想说

Something I

我想问问你们俩，当你们意识到产品和市场真的匹配了，那种感觉是怎样的？

wanna ask I wanna hear from both of you is like, how does it feel when you're like, oh, I've got product market fit?

如果有人问我，你有产品市场契合度吗？

Like, I if someone asked me this, do have product market fit?

我可以肯定地回答，有。

I could say, yes.

我来解释一下原因。

Here's why.

到底是什么样的迹象让你觉得，没错，我真的找到了？

Like, what what was the kind of like signals that make you think, yeah, I've got that?

我懂。

I know.

我觉得很难说清楚。

I think it's hard to tell.

我以前觉得我们在某些阶段已经达到了产品市场契合，但最近三四个月市场变了，我才真正觉得我们做到了。

I I I thought we had PMF at certain stages and I was like, now we have PMF more recently in the past three, four months because the market has changed.

不过我觉得，好像是你或者别人说过，产品市场契合是会随着你的发展而变化的。

But I think, I don't know, it's you or somebody else said PMF changes on your journey.

不是说你一旦达到就万事大吉了。

It's not like you hit PMF and then you're done.

我觉得真正的标志是，当你意识到，你之前投入的所有努力——无论是产品流程、用户引导，还是用户群体——突然之间开始自动加速运转了。

And I think it's it's literally when honestly, it's when you realize that all the work that you put in, all the mechanics, the onboarding and stuff, and the audience, and then you you can just you just sit back and stuff just accelerates.

对吧？

Right?

我们第一年的时候，可能每周才签一两个客户，但现在就算最差的一天，也能签下一百个客户。

Like so we were signing up, I think in the first year, maybe, I don't know, one customer a week or two customers a week, and we now sign on on on a bad day, a 100 customers a day.

所以对我来说，这种增长就是产品市场契合的明显信号。

And so that's that for me seems like a decent PMF indication.

对我来说，PMF就是当你能去度假，整个团队都去休三到六个月，收入却还在持续增长。

I for me, PMF is when you can go on holiday like, you think the whole team can go on holiday for three to six months and revenue will continuously continue to grow.

对。

Yeah.

没错。

Right.

可公司里没人运营啊。

Well, no one operating the company.

不用，你只管去就是了。

No, you just go.

连客服都不用管。

Even support.

客服从来不能停。

Mean Support never leaves.

就像那种，真的就是当……

It's it's like it's like when like when yeah.

你就放轻松就行。

It's just like you you just chill.

就是说。

It's just

在自动运行。

working.

就像完全自动运行一样。

Just like autopilot.

我觉得你已经达到了另一个层次的PMF。

I I would say like, you found another level of PMF.

Veed最近也达到了另一个层次的PMF。

Veed has recently found another level of p PMF.

没错。

Yep.

不过感觉和第一次达到PMF时不太一样。

It doesn't feel the same though than the first p m f.

哦，是的。

Oh, yeah.

为什么会这样？

Why is that?

第一次找到PMF的时候，简直惊呆了，这是我这辈子遇到过最让人激动的事。

Like the first p m f is like, oh my god, this is like the most exciting thing that ever happened to me.

对。

Yeah.

而第二次，规模更大了，

And the second time it's like bigger,

影响也更深了。

more impactful.

又来了。

Here we go again.

但感觉完全不一样了。

But it doesn't feel the same.

是的。

Yeah.

没那么让人激动了。

It's not as exciting.

你总会记得自己的第一笔销售，记得客户是谁、从哪儿来的。

And you always remember your first sale, you always remember, you know, who they are, where they came from.

我跟很多粉丝聊过，当你在这个行业待得太久，公司能连续四五年的状态不变，其实非常少见。

And I think as I was speaking to a lot of fans, when you are in the game for so long, it's very rare for a company just to stay as it is for more than, like, four to five years.

市场在变，事物在流动。

Like, the market change, things move.

你得不断重塑自己。

You have to reinvent yourself.

看看世界上最大的公司之一，苹果。

Look at the biggest company in the world that's doing it, Apple.

他们把自家的iPod彻底消化，转型成了iPhone，诸如此类的改变。

Like, they swallowed their, you know, iPod and made it into iPhone, all that kind of stuff.

所以你得不断调整产品与市场的契合点，弄清楚它对你、对公司、对整个世界意味着什么，这真的让人精疲力尽。

And so you con you you have to constantly refine PMF, what it means for you, for the company, for the world, and it's it's it's exhausting.

对吧？

Right?

每隔几年就得重新打磨你业务的核心机制，真的挺不容易的。

Like, really trying to, you know, like, reinvent the mechanics of your business every every few years.

但这就是长期持续运营所必需的，非这么做不可。

But that's what needed that's what that is what is needed to continue operating for a very long time.

对吧？

Right?

所以很明显，如果你以前是从十万美金年经常性收入一路做到一百万美金，那确实挺厉害的。

So obviously, if you used to go from a 100 k ARR to a million ARR Yeah.

你从零开始，花了多久才达到一百万美金的年经常性收入？

What how long does take you to get to a million ARR from when you started?

大概两年吧，从十万美金做到一百万。

So like two years a 100 k ARR.

是的。

Yeah.

我觉得再过四年才能达到一百万的年经常性收入。

So I think a million is probably four years after that.

对。

Yeah.

没错。

Yeah.

所以，这么说吧。

So okay.

加起来一共六年。

So six years total.

是的。

Yeah.

当然，一路上有成千上万个小决定和其他因素促成了这一切。

And obviously, there's a thousand things which got you then tiny decisions and whatever.

不过，是的。

But Yeah.

如果非要归纳一下，从产品角度或者决策层面来看，你觉得要达成100万用户，主要得做哪些大方向的决定？

If you had to just like had to categorize just like the big categories or decisions that kind of needed to happen to get to 1,000,000, what what do you reckon those things were?

我觉得从产品角度来看，我们做了很多调整，虽然产品整体看起来变化不大。

I think from there were a lot of stuff we changed from a product perspective even though the product stand, like, fundamentally looks the same.

比如，我们有一段时间彻底转向了免费增值模式。

I think, like, we went truly freemium at one point.

这确实帮了大忙。

That kind of helped a lot.

去年的那个测试，我觉得对加速顶端用户的增长帮助很大。

The trial, like, was probably, like, within last year helped a lot in terms of accelerating people on the top end that was there.

然后我觉得另一件大事是我们重新设计了网站和平台，还请了一位专业设计师。

And then I think one of the big things is we just redesigned the the website and the platform, got a proper designer on.

从那以后，我们看起来就不像那种随便搭出来的草根网站了。

And at that point, we didn't look like this kind of bootstrap website.

你能感觉到有些人用的是那种‘ vibe code ’之类的东西，然后你就不太信得过他们。

You can tell when someone's like Vibe Code or something and then it's like, you don't kinda trust them as much.

这一点特别关键。

So that was really big.

我们把定价固定下来了。

And we we've we cemented pricing.

以前我老在改套餐和价格，但最后我们敲定了三个核心套餐。

Like, we I was playing around with plans and pricing a lot of times, but we we settle on like three core plans.

通过这个，我们弄清楚了怎么给用户分类。

We figured out how to categorize our users based on that.

那时候有好多类似这样的调整。

So there was a lot of those kind of things there.

是的。

Yeah.

不是。

No.

我记得我以前老是吐槽那个设计，

I think I remember bashing on about the design for for

好一阵子呢。

a while.

当时我就说，埃尔森，别这样。

Was like, Elston, come on.

但你说，没坏就别修。

But you were like, it ain't broke, so we're not gonna fix it.

还有几件事我觉得也很关键，

So a couple of other things that I think were also big things

是的。

Yeah.

对你来说。

For you.

你要是觉得有什么特别重要的，就说一声，比如招聘这件事。

Let me know if you just go it was just well, obviously one is around hiring.

是的。

Yes.

还有一件事，我觉得就是你持续不断地在YouTube和其他平台发布SEO内容。

And another another thing is I think it's just your absolute just consistency in putting out SEO content in YouTube and other stuff.

基本上，你六年从来没停过。

And you just didn't stop for six years basically.

说实话，如果我们拿了风投，别人肯定会问：你们到底在干嘛？

Honestly, if we were VC backed, people will be like, what the hell are you doing?

但这正是我们自筹资金的优势。

But it's a bootstrap advantage honestly.

我就一直顺着这条线往下拉，没松过手。

We consist I was just kept pulling the thread.

事情就这么一直往前推，越推越顺。

Like, it just kept going and kept going.

我心里就想，我们还在增长呢。

I was like, we're still growing.

咱们就继续这么干下去。

Let's keep on doing this.

有时候一个月增长百分之三，有时候一个月增长百分之二十、四十，但总之一直在增长。

We had months we were growing 3%, months we were growing 20%, 40%, but we were still growing.

我当时就想，只要还在增长，我就继续这么干，边干边找方向。

And I was like, as long as it's still growing, I'm gonna keep keep doing this and figure out where we're going.

现在回头看，我昨天遇到一个人，三个月内从零基础做到了年经常性收入一千八百万。

Like retrospectively looking, like now, I mean, I met someone yesterday who went from zero to 18,000,000 ARR in three months.

哇。

Wow.

十八还是八十？

18 or 80?

十八个月。

18 in

三个月。

three months.

这太离谱了。

That's insane.

简直离谱。

Insane.

太离谱了。

Insane.

不过，那是另一个世界了。

Anyway, different world.

是啊。

Yeah.

AI时代之前和之后，完全是两个类别。

Different category, post AI, your pre AI Yep.

这话挺有意思的。

Which is a funny statement.

回头看看那四年，从十万到百万，如果非得说有一件事能让你走得更快，你觉得会是什么？

Anyway, like as we look back at those four years going from a 100 k to a million, if there was one thing you could do that would have made that go much faster that journey, what would have it been retrospectively looking back?

还是说，你的成长速度其实没你想象中那么快？

Or did you grow as fast as you think you could human humanly possibly grow?

没有。

No.

我觉得这其实牵涉到自筹资金和风险投资这两种模式的区别。

I I I do think I mean, this is where like bootstrapper versus VC kind of game comes in.

有资金的时候，你可以多发视频、多做内容，很多事情都能推进得快很多。

Like when you have capital, you can like put more videos out, put more content now, kinda do things a lot faster.

我本来也可以更快地招人。

Maybe I could have also hired a lot faster.

我觉得很多事我确实可以做得更早一些，但我们在产品上积累的这些经验，我不确定能不能更快地获得。

I think I think there's a lot of those things that I could have done to a degree, but I think the learnings that we've got in the product, I don't know if we would have got there any faster.

关键还是得不断推出东西，看用户怎么反应，然后调整策略，再调整，不断优化。

It's really about like putting stuff out, trying to figure out how the audience resonates with it, tweaking the plan, tweaking it again, moving stuff up and that kind of stuff.

这些事都得亲自做一遍，得把各种条件都理顺了，才能真正搞明白什么是产品市场契合点。

I think that stuff that needed to be done, and we needed to get those, like, you know, ducks in a row to figure out, like, what what PMF is.

但毫无疑问，我们本来可以发布更多视频。

But for sure, like, we could we could have put a lot more videos out there.

我确实一直都在输出内容，只是没做到那种大家常见的密集程度。

I mean, I've consistently put stuff out there, but not at a volume that people kind of, you know, you see.

我记得很多公司，不是客户，而是其他公司，会来找我们说，嘿。

And I remember a lot of companies, not company customers come to us and say, hey.

比如 Tiiny Host 和 Square，他们是偶然发现我们的。

Like, Tiiny Host Square stumble upon it.

他们说，我们希望更多人知道你们。

We need I want more people to hear about you.

你们是怎么做到的？

How do you do that?

我就说，我其实一直在努力。

And I was like, well, I'm trying.

但说实话，我也不知道具体该怎么做。

But I think like but like, but I don't know of you.

我是通过别的渠道偶然发现你们的。

I just found you through something else.

我当时就想，确实如此。

And I was like, yeah.

但作为一家公司，你总觉得大家都知道你，因为你一直在推特、YouTube这些地方发内容。

But so clearly, like, as a company, you think everyone knows about you in some sense, but because you you you're putting, you know, as much stuff here on Twitter, you're on YouTube, you're this.

但事实上，就算到现在，我们已经有两百多万用户了。

But in reality, even now, we have over 2,000,000 users.

还是有很多人根本没听说过我们。

There's a lot of people never heard about us.

不过我现在发现，在我们的圈子里，去参加一些线下活动时，总会有人说：哦，我知道Tiiny Host，听说过。

But I'm coming to a point where, like, probably within our community, go to, like, meet ups here and there and, oh, I've heard of Tiny Host and that kind of stuff.

但这终究还是个小圈子。

But it's still a bubble.

外面的世界根本没人知道我们。

It's not like the wide world has heard about us.

所以我觉得，要是当时有资金支持，肯定能帮我们更快推进，也能更早地招到更多人。

And so I think, yeah, like, capital would have definitely helped and helped us move forward and and hiring people a bit more sooner.

但我还是想走这条自力更生的路，这正是它的魅力所在。

But it's the journey I feel like as a bootstrap that I wanted to take and that's the the beauty of it.

那些不了解埃尔斯顿的人可能不知道，他在发掘人才方面简直传奇，总能以极低的成本在全球找到顶尖人才。

So those that don't know Elston know that Elston is actually like legendary for finding diamonds in the as in just like he just finds really good people in all over the world for pretty good value.

你平时是怎么做到的？能分享一个最核心的建议吗？

What's your kind of approach to that like, you know, what's your like number one tip just for doing this?

是的。

Yeah.

我几年前带过一家公司，整个开发团队都在印度。

I mean, I I I basically ran a company a few years ago, which was the whole dev team was in India.

所以我特别清楚，怎么在海外找到真正靠谱的人才。

And so I got a really good understanding of, like, what it takes to, like, whistle out to good people when there's offshore.

很多人雇海外人才都踩过坑。

And I think a lot of people get burnt from, like, hiring talent offshore.

我觉得自己有个优势，因为我是做开发的，所以我招的离岸员工里，程序员占了大多数。

I think I've got an advantage because I'm a dev, so the the most people I've hired are probably devs offshore.

我一眼就能看出来谁是好程序员，看他们的思维方式，还有写代码的习惯，这些细节都骗不了人。

And I can tell exactly like there's there's little telltales when someone is a good dev, the way they think, the way they, you know, write the code.

所以哪怕面试流程很简单，我也能很快判断出这个人到底行不行。

And I can understand from a very simple interview process, like, you gonna be good or not?

这一点真的很关键。

And so that's really important.

我觉得确实得先搭好一个筛选漏斗。

I think it's there's there's definitely this funnel we need to set up.

对吧？

Right?

所以不管你在哪发招聘，比如在印度或者非洲，反正一放出去就能收到一百份简历。

So you put any, like, job listing out anywhere, maybe in India or Africa, whatever, you get a 100 applicants.

然后你得挨个去联系一下，结果大部分人根本都不回你消息。

Then you gotta, like, message, you know, a few back, and then most of them don't even kinda reply back to you.

你再发消息约他们面试，有一半根本不会来参加电话会议。

And you message them more for an interview, like, half of don't even show up on a call.

最后能剩下三四个候选人做测试，真正能留下的就那么一两个，整个流程真是又耗时又费劲。

And then you get to, like, maybe three or four people, they do the tests and then you get maybe a few people and then so there's this process you need to go through as laborious.

但作为一个自筹资金的创业者，这跟有稳定收入、能长期合作的团队完全不一样，那种感觉踏实多了。

But as a bootstrapper, it's it's it's nothing like having just, like, you know, people who can work with you for long time based on the revenues that you have and you can deal with.

这样一来，你就等于有了无限的时间和空间去慢慢发展。

And then you have effectively just an infinite runway to do that.

但现在，尤其是我们进入人生下一阶段时，确实会发现有些人跟不上节奏了。

But it is right now, especially as we're going through the next stage of life, you do definitely outgrow people.

当初招进来的人，可能已经不适合接下来的发展了，必须得提升和换血。

There are people you hired back then who are not good for the next journey, and you need to upscale and upgrade.

我们现在那位首席工程师在英国，能力特别出色。

And we've now, got our lead engineer is is in The UK, and he's amazing.

对吧？

Right?

不过，外面还是有很多优秀人才的。

And so but there is a lot of good talent out there.

现在更有趣的是，这些人才在AI出现之前就已经非常出色了。

And now it's also even more interesting because that talent also, I'd say, was very good pre AI.

对吧？

Right?

那时候根本没有什么大语言模型，也没有Claude Code能直接帮你完成这些工作。

And so, like, there wasn't an LLM out there that could just there wasn't Claude Code that could go and do this.

对吧？

Right?

所以现在你要找的是一个 offshore 的初级人员，得比Claude Code更强，或者能把它用得特别好。

And so that but now you gotta find a junior offshore person if you're looking for someone who's better than Claude Code or uses Claude Code very, well.

这才是现在的标准。

That's that's the bar.

以前呢，你花在token上的钱其实挺多的，但现在找个人 offshore 做，反而更划算。

But before it's like, this was like effectively how much you'd burn on tokens, but having someone, you know, offshore doing it.

所以情况有了一些变化，你需要换不同的角度来思考。

And so it's it's changed a bit and so you need to think through different lenses.

这真的很有意思。

That's super interesting.

我知道，不管你在哪里招聘，大概都要面20轮左右才能找到合适的人。

I know that you do have to I think you should assume if you're hiring regardless of what country you're hiring in, it probably takes 20 10 to 20 interviews Yeah.

才能找到真正合适的人选。

To find the right person.

很多人太急着想一下就招到最好的，结果就将就了，但其实不该这样。

And I think people are just so desperate to find the right the first person that they'll just settle and I think the important thing is not to.

对。

Yeah.

你刚才说工程领域的成本结构变了，这话特别有意思，能展开说说吗？以前写代码不就是靠堆人吗？

What do you mean that's really interesting what you said about like how the economics of engineering changed Because it used to be coding was a brute force thing.

需要更多工程师来开发更多功能。

I need more engineers to build more features.

现在这可能变成了一个技能问题。

Now it's potentially like a skill problem.

对。

Yeah.

你该怎么看待这个问题呢？

How do you how do you think about that?

AI 编程是怎么改变你的工作方式的？

And and like how's AI coding changing what you're doing?

你还在写代码吗？

Are you coding?

你已经不写代码了？

Are you not coding?

具体是怎么运作的？

Like, how how does it work?

是的。

Yeah.

我的意思是，过去一两年里，情况确实变了。

I mean, it's definitely changed in the past one or two years.

所以我给团队，或者小一点的工程团队定的底线就是：你们每个人都要比Claude Code更强。

So, like, the simple thing I've told the team or a small engineering team is the baseline now for all of you is be better than Claude Code.

这个工具就摆在你桌上，明摆着的。

Like, you've got this tool literally on your desk.

你不能比Claude Code还差，也不能比它还慢。

Like, you cannot be worse or slower than Claude Code.

你必须得比它强，因为它就在那儿。

Like, you have to be better than it because it's there.

这其实是个挺重要的变化。

And so that's kind of a big thing.

以前我刚组建团队的时候，最头疼的问题之一就是，出了问题，比如有个bug什么的。

And I previously, maybe when I hired a team, one of the biggest like issues I had was like, okay, there's a problem, there's an issue with a with with a bug or whatever.

那时候我只有两个选择。

So I can either I've got two options.

要么我亲自花时间写代码把它做出来，要么我就训练团队来完成它。

Either I can go in, spend my time and code and build it or I can get train the team to do that.

如果你真在带团队，那就该让团队来做，而不是自己亲自动手。

And really, if you're building a team, you need get them to do it rather than yourself.

有一阵子，我一直都在推动这一点。

And for a while, I was like pushing pushing that.

最近三四个月，客户支持那边在Slack里发消息，详细说明客户遇到了什么问题。

And now, like, in the past three, four months, someone in customer support will post a chat a message in Slack saying customer has this issue, this is the problem here, like a very detailed explanation.

现在直接复制粘贴到Claude Code里，发一个提示，生成一个PR，合并就完事了，效果好得惊人。

Now literally copy paste, put into Claude Code, one prompt, one PR, merge and like it's that good.

完全是另一个世界了。

Different world.

确实如此。

Yeah.

我觉得这对我们有效，主要是因为我们代码库结构很清晰。

And I think it works for us mainly because we've got this code base which is well structured.

我也熟悉这个代码库。

I also know the code base.

我大概能再补充点上下文，比如这个问题出现在哪个文件里。

I can probably like give it a little more context in terms of like, it's in this file.

但至少已经有四五次了，十分钟到二十分钟内我们就把bug修好了，我都参与其中。

But there's been at least four or five times wherein about ten, twenty minutes we fixed a bug, and I I've just been involved in it.

所以现在我会主动这么做。

And so now I'll I'll do that.

我还见过更进一步的做法，我们现在正试着推行：当有支持工单进来时。

And I've seen next levels of this, which we're trying to implement now, which is like like a support ticket comes in.

后台会自动启动一个代理程序。

There's automatically this agent, which is kinda spun in the background.

它会生成代码提交，工程师只需审阅并合并即可。

It raises the PR, and engineer just looks at reviews and merges it.

我觉得，工程团队处理工单确实存在一些额外负担。

And I think that's there there is an overhead to tickets in engineering.

对吧？

Right?

不管任务大小，你都得在自己的工程记忆里理清楚任务内容，读工单、写工单，或者提代码评审。

Like, regardless of the small or big, like, you have to, you know, in your own engineering memory, figure out what the task is, read the ticket, write the ticket, whatever, raise the PR.

对于那些很小的bug或工单，比如一些小调整或功能优化，就不该浪费工程师的时间。

And so for the small small bugs and tickets, which are just like little things need to be done tweaks or like enhancements, stuff like that, it shouldn't be wasting engineering's engineering's time.

对吧？

Right?

我觉得这些事应该在后台自动处理，工程师只需要处理代码评审就行。

I think it's it's something that it should be done in the background, and you just have PRs.

最近我们给了产品经理前端访问权限，他以前从没写过代码，但现在有很多想法，都是些用户界面相关的建议和改进。

And more recently, we've given our PM, like, front end access, and he, you know, he never coded before, but now he has a lot of ideas like oh oh, just very, like, user interface oriented kind of suggestions and improvements.

对吧？

Right?

现在他可以直接进去提代码评审，让团队来审阅。

And now he can just go in and raise a PR, and get the team to review it.

这太棒了，因为他不用再提工单、等工程师接手、等审核、再自己测试、最后合并，这一套流程走完，运气好也要两周。

And that's super cool because the friction from, like, him needing to, like, you know, raise a ticket, wait for the engineer to pick it up, wait for it to be reviewed, then him test it, then go to merge is like it could be a two week cycle if you're lucky.

现在呢，我直接看到一个PR，兄弟。

And now it's like, oh, I just got this PR, dude.

你能帮忙审一下吗？我们明天就能上线。

Can you can you just review it and we'll get it out tomorrow?

我很喜欢这种安排给团队带来的改变。

I love the unlocks this is giving your team.

你们那些技术背景较弱的成员，现在也能参与进来了。

So you're less technical members.

他们可以以自己的方式，做出小但重要的贡献。

It's allowed them to contribute in their own smaller ways.

同时，这也让你们的资深工程师能专注于更重要的事。

But also, it allows you to for your more senior engineers, you charge it.

那些简单的小改动，系统会自动建议创建PR。

So the simple stuff, it will just automatically suggest a PR for that.