# Building a profitable Chrome extension | Lessons from Pretty Prompt

## Метаданные

- **Канал:** Chrome for Developers
- **YouTube:** https://www.youtube.com/watch?v=7i2N7Hqyqyg

## Содержание

### [0:00](https://www.youtube.com/watch?v=7i2N7Hqyqyg) Segment 1 (00:00 - 05:00)

We are yet to understand why they choose you know an annual versus a monthly pay plan and it's really interesting data for us but again we are still learning from our users. Hello and welcome back to developer mode. Today I'm joined by Charlie and Eli, two co-founders who are earning money from their Chrome extension. How to successfully monetize an extension is something a lot of developers ask about. And so today I really want to focus on everything they've learned about how to build something users will actually pay for. Welcome both to the show. How are you doing? — We're very good. Thanks for inviting us. — I would love to just hear a little bit about sort of what your product is and how it's going so far. — Yeah. So we uh we build pretty prompt which is a chrome extension that sits on top of uh your AI providers think sort of chat board — uh Gemini uh and we uh we help you improve your prompt as you write it. — Nice. — And it's really easy to get started once you install the Chrome extension. People just press this improve button and they get this small prompt that they wrote into a whole enhanced uh piece of text that hopefully gets them better results and it's been really an interesting journey. and sort of how's it going? Where are you at the moment? Sort of what's your journey been like so far? And what milestones have you hit recently? — So, we recently passed 15,000 users, which is really amazing. — Congratulations. Thank you. Uh I remember when we started this a few months ago and uh yeah, we only been live as a product for a few months to be honest. So, it's been pretty fast. — So, I think we started around about early June and we've been going since then. And uh yeah, we it started off as a sort of internal tool uh we were both using and we thought it'd be a great idea to uh just launch this out and release it. And the reason was because Charlie who is actually the main technical co-founder from both of us. I deal mainly with customers. Charlie deal deals mainly with the product and the building. And I think that the good thing is we were building a separate product before and internally we were constantly prompting and changing things. It took a whole lot of time like super like should we just build something to solve this problem for us — which I think is a great learning you know if you have a problem just solve it yourself first. — Yeah. — Um yeah and we released it and it had a um it had such a good response. We uh we decided to sort of burn the boats and go completely onto this. Uh and I think it was a good decision since then — 100%. So since June within the first week I remember of launching we had about 3,000 users — which is a pretty good milestone. That's already Yeah. a lot of progress in each of time. Yeah. — But again, we didn't launch trying to make money. Again, it was an internal tool. — Y — and we wanted to almost like validate if people would first of all use it — and second of all want to pay for it because there's no point to build something no one wants. It's like, you know, everybody talks about like the main reason why products fails are not because of the product are because customers don't need, don't want, maybe they don't get it, which is completely fine. it doesn't find the product market fit. — Exactly. So, did you find that users were actually willing to pay for your extension? — Uh, yeah. So, in the uh when we first launched it, it was completely free to use, but within about a week, uh we launched it around about the start of June, — 4 days based on the email. — Four We got an email from a user saying, um, yeah, we really want to pay for this. And that was kind of like the gold standard for saying, okay, we should really implement a payment provider. — Yes. — And again, it was part of validation. It's like level zero, level one, level two. Like we were still in level zero and you know, we are now in level one or two, no more than that. You know, it gets harder and harder. But the fact that someone emails you and say, "Hello, I was trying to subscribe and potentially pay for the service is like we should have a payw wall, you know, we should get on this like as a priority. " — And we talk about sort of adding a pay wall as something you just sort of decide to do and then it's done. But I know realistically like there's a lot of work and a lot of complexity to handling chargebacks and uh maybe refunds. So sort of upgrades — or upgrades even. Yeah. How do you handle all of this? — So uh so yeah I think we uh so we started off okay what's the most basic way we can do this and that's really taking um taking a payment from a user sending them to a checkout. So we just integrated with payment provider Stripe um and then we linked that to our database and then that would update um the user just as a flag which says is paid. — Okay. — Um really super basic. Uh there was no way to you know uh refunds we did that all manually. Upgrades — When they would pay I would send them a message because we would get a notification and I would see someone paid. I would go manually to send them an email. There was no automated email in the beginning. I was like, "Welcome to Pretty Prompt Pro. " No, there was no that. So, I was like, "Hey, I'm Eli. I'm one of the co-founders. I'm super excited to welcome you and thank you so much for paying. If you have any issues, let me

### [5:00](https://www.youtube.com/watch?v=7i2N7Hqyqyg&t=300s) Segment 2 (05:00 - 10:00)

know. " So if you got an email in what sort of the first weeks or months of pretty prompt then maybe it was directly from you — and it's still from me because you know I messaged people directly and funny enough even technical founders should and not that this is for everybody but like should feel that sometimes part of building a product is not building but it's actually talking to the people and like Charlie does support. Yeah, I I mean I think that's a big thing. I think all engineers should do support. They should get on a call with the customer and they should really feel like okay, if the customer is asking for a certain feature — um yeah, why are they ask why are they asking for it? Dig a little bit deeper. Don't just build it, but understand, okay, what is the underlying principle as to why they want feature X? Yeah. — And can this be solved in a — maybe not even a automation way? Is this something that isn't even a problem to start with? — Yeah. And going back to kind of like the payment flow, we put together this early on to understand how we would do it. And we came up with literally kind of like this four step which is from the extension you have this button that says upgrade. — Yeah. — Which you can click anytime but it would kind of like come to the front once you hit the limit that we are testing. What is the limit that user wants to upgrade — and that would take you directly to a checkout — in the very beginning. — Yeah. So it take you to a checkout and then you'd have like a web hook uh that would call sortly a superbase back end and then update the uh update the tier on the uh it's actually on the Chrome extension. It wasn't even on the back end, right? We only logged when you get paid. — Okay. — Um so we sort of capture the customer ID. So if there were any problems on our payment provider, we had some kind of indication of who they are. Something interesting with Chrome extensions specifically is when they would come back from that payment into the Chrome extension, we would reopen the popup which is kind of like on the top right corner of your browser — to refetch, you know, and make sure that is updated now they appear as — to get like the latest payment data from your server or wherever you need to check. — Exactly. So, every time that you kind of like upgrade and change your tier, — that gets reopened. Yeah, I think making sure that flow is uh works quite well is important because there's nothing worse. You've paid a you've given over your debit card uh details, you've paid for something and then nothing happened. — Nothing that happened. So you have then email them and it's like so we really tried to test that to make sure you know it really does work flawlessly. So there's no kind of break in what they're doing. — Yeah, I think that's where customer support becomes really important. I think if you have an issue and you get back to someone really quickly and say something went wrong with our system and we're going to fix it, then all is good. The longer you leave it, the more users start to think maybe they did just take my money and something is not right. — Exactly. You want to keep that kind of almost sort of uh a asynchronous communication as opposed to just synchronous. Um and I think that's uh that's maybe um worked quite well for us. — Yeah. And we we've spoken in the past and I think something we all agreed was there just isn't much data for developers to aspire to around sort of what's possible with extensions and what like size of an audience can you reach. How willing are users to pay for extensions? So I'm curious are there sort of any numbers you can share any insight into sort of what you found is possible and what you think is realistic. — Yeah. Uh so [snorts] over 15,000 users right now and within the first 60 days so we got a user we launch that was kind of like first uh stage. — Yeah. — Second one was what Charlie said which is — getting a paying get getting someone to ask to actually pay for it. — Exactly. So like — ask to pay great thick. So very quickly — Mhm. And then afterwards we actually got some um sort of uh it was like Tik Tok organic growth uh from just people building um videos whether it be on YouTube, Tik Tok or wherever. And that was uh that was also an amazing kind of validation for us because — we didn't ask these people to do it. They were just like this is such an interesting product. We really like to do it. And that sort of drove even more growth with it. — Yeah. It's a nice cycle where maybe a creator feels like if I share this and this is useful to my audience and of course it's useful to you because it brings you an audience. — Exactly. But I think that the very first thing is we all — as builders are trying to think about what the market wants and like what is the big opportunity and I think that is a great way of looking at it — but we took almost like the counterintuitive way of doing it and I and it's easy to say now you know when you're in the moment you don't think about I'm doing the opposite. No. Yeah, — but right now we built something that we wanted. There was no like big audience. — We actually launched on birthday. — It was Saturday. — I was actually in Greenwich uh Greenidge Park having a pizza. Um and then

### [10:00](https://www.youtube.com/watch?v=7i2N7Hqyqyg&t=600s) Segment 3 (10:00 - 15:00)

suddenly um — Slack notification started going off. Um funny enough, we uh later that day, we actually got kicked out of Slack, which has never happened before in any other company I've been in. — Um so we had to switch very quickly. Why did we get kicked out? — So, we got kicked out because we had um two high rate limits on our kind of analytics channel — and you were posting like every time there was a new user or — so given that it was a new uh new product, you never know what to track. So you just try and track everything and — you were getting the notifications on Slack essentially — like every single you know second you were getting like 10 20 notifications — like for example sorry but like we just got like a customer like upgrading and sending some feedback. So like it's really useful because you get real time. — Really tapped into sort of what's happening and yeah how things are doing. — But then we got rate limited. — We got rate limited and then we had to get kicked off of Slack. Uh so uh it was only for about 24 hours but um it was a lesson to be like okay uh this is something you know people want they want to pay for — um and yeah — yeah so within the in terms of a little bit of metrics so um we started without the payw wall within the first week people sent kind of like these emails and messages with a lot of requests lots of questions a lot of stuff that is like fluff oh I love this that's no one cares about that what you want is actually what is not working and why they care. Why they like it, not that they like it. Tell me why you like it. What is the single most important feature or capability or unlock that this product does for you? If you tell me that, then we know where to double down. And if you tell me what doesn't work for you, we also know what we need to work on. So, it's like kind of like the good and the bad, not the neutral that no one cares. — I mean, it's nice to get that neutral feedback, but it's just not very useful for building a product. — Yeah. But for other developers like don't seek approval. — Seek for want to pay. That I think that is very different from like I love this bat. — Yeah. You know. — Yeah. No. I completely agree. I think it's um you can have a lot of reviews saying giving you such positive things, but ultimately reviews are great. — Yeah. — But if you want to monetize a product, um the only hard evidence is are people willing to pay for this? Yes. — Putting the credit card, putting your money where your mouth is. — Were people willing to pay? What did you find? — They were uh so yeah, they were they were really willing to pay. Um it immediately solved their problem. I think one of the big things we did which worked out quite well. We didn't think was super big at the start was the on boarding was really seamless. Like it takes about 10 20 seconds. — No on boarding. — Yes. Like add to Chrome. — It takes about 10 20 seconds to get any kind of value from it. So they can immediately see uh where it is. And what was quite interesting is then we started adding an annual plan to this. — Yeah. — And we just threw that out. We thought oh — what happen if we can you know give them a discount for like as usually you know SAS tools — you know are doing and — for say you know you get two months free over the whole year and then suddenly loads of people were signing up because they were like we want to take advantage of this. — I learned my metrics as a founder. So I want to say this 25% of users today pay for annual subscription which is huge 25% that's a quarter of our user base — which pays for annual. Now we are yet to understand why they choose you know an annual versus a monthly pay plan and it's really interesting data for us but again we are still learning from our users but we definitely know that it works because we didn't have it like Charlie said we added it now 25% of users are choosing annual versus the other one and as a startup as a small company this is huge to be able to — take this because you don't just take the money you take their trust for a year — so it's a lot and yeah I imagine the user then feels more invested to continue using the product for a year and sort of see it through while they have the premium access. — Absolutely. I think one of the other big things is getting users who are engaged and willing to give feedback and deeply think about okay is this actually solving my problem and uh if they are invested for the uh entire year that that's uh really helpful and we try to do like yesterday we did three calls with different users we try to do every week since that you know beginning of June — calls with customers to understand exactly what's working is not working what they like, they don't like. — Yeah. — And not only it helps to create a cycle back to the product, — but actually it makes them — be more invested in the product themselves, which mess means it drives more word of mouth, which means you have more probabilities of charging more money eventually — because they became your biggest advocates. So, we've actually haven't spent a pound, a dollar on marketing so

### [15:00](https://www.youtube.com/watch?v=7i2N7Hqyqyg&t=900s) Segment 4 (15:00 - 20:00)

far. — That's amazing. Which is amazing. — Yeah. And I think a lot of developers will be watching this and will be thinking I would be more than happy to do the customer calls. I would love as much feedback as I can get, but I just don't know how to even get in touch with the customers who could give me that feedback. So do you have any advice for sort of how to initiate that sort of thing? — So I so yeah I think um so for your existing cu uh customers make it very easy to contact you. — Yeah. So uh — sorry. And when they say you it's you. It's not like a team. No, put your name, put your face, put make sure people talk to a human. — Yeah. And uh yeah, make it very easy for them to contact you. — And uh reply quickly. Uh I mean I think that's a really underappreciated thing. Uh you know, sometimes you take uh 24 hours to reply. If you can reply within 10 15 minutes um and you — 60 seconds. — Yeah. 60 seconds — realistically. And if you can consistently do that and you can solve a bug for them within let's say 30 minutes uh suddenly they're like oh this is fantastic because they're not used to that kind of customer support because whether it's a big company you know there's so many layers that has to go through but — it's almost like very refreshing when you have some kind of very quick customer service and it's to the direct founders so they don't have to go to management to ask can we add this they can just go okay I can commit this straight to the GitHub repo test Yeah. And you can test it with them and say, "Is this fixed for you? " And they say yes and they give you a fantast review because especially for Chrome extensions, reviews are extremely powerful because they are first and foremost — center in your page of, you know, of your application. So if it worked out, ask them for a review. It's like, hey, thank you so much. I'm glad that it's working now, would you mind leaving a review for us? You know, and you know, fivestar review or not, but ask them for an honest review because they also will say what they like or they don't like. — And that is really, really powerful because — people are willing to share their experience whether it's good or bad usually, but they are willing to share their experience. And I think this is an important clarification because I've sort of seen some misunderstanding here. We have Chrome Web Store policies against sort of manipulating reviews or asking for dishonest reviews, but reviews are really important. And if you have a customer who's had a good experience, encouraging them, please just leave an honest review of how you found the product. Like that's completely fine and you should absolutely do that. — And I think you ask it exactly like that. Please leave an honest review of your experience. That's it. There's nothing more to ask or to say there. And you know, that's the best way to get people back into the news. — Yeah. — And I I think given that it's almost getting harder to trust uh what people say on their landing page, right? — And just on the internet in general. — So if you uh the fantastic thing about Google reviews, it is linked to someone's name. So it's linked to the date they did it and it is on the Chrome Web Store uh website. So you So we actually link it directly there. These aren't just reviews someone has given us to in an email. They're actual reviews that exist on the web store. — So touching a little bit of monetization and what Charlie just said. So we call that social proof. So social proof are reviews, what people say, you know, all the logos and all the nice smiley faces and all the five stars in the, you know, in the Chrome web store. — So what we actually learned is we originally had this button that said upgrade for paying customers once they hit a limit. So the way that pretty works is like you have you can try it without even logging in which is an amazing kind of like idea that Charlie had and it works amazingly because you add to Chrome you try it you like it only after if you want you go to our free plan which you get a limit per week — great if people need more and want to unlock all these amazing features which are amazing on pretty prompt they can click the upgrade button originally we would take them directly to the payment page which is just a checkout. — We realized through our analytics that people were clicking but not upgrading. So we were like — what is happening there? — Yeah. So how do we kind of fix this or get some kind of information? So we added kind of like intermediate page between there. — Uh so we could actually you know see okay uh if we give a load of features what features do people want? Uh we have those on the sort of pricing plans. Yeah. And the advantage — and social proof is also really a big thing because if you see 10,000 other people are using this, you know you're not the first guinea pig. — No one wants to be the first one. Yeah. — It's maybe worth explaining that in more detail like what is encompass when you say social proof? What does that mean in practice? — So I like to call it I like to divide

### [20:00](https://www.youtube.com/watch?v=7i2N7Hqyqyg&t=1200s) Segment 5 (20:00 - 25:00)

this in quantitative and qualitative. So quantitative is literally number of people who have left a review, number of good reviews, your star rating, you know, essentially out one to five, how many fivestar reviews you have. So we actually have over 90 five star reviews. — And so that's anything you see on some sort of landing page or interstitial that says, you know, this extension is trusted by this many people, — X number of people and you know 4. 9 over five on the Chrome web store. Like all those things are quantitative. Qualitative is actually the words you people use to describe that review. — Yeah. — And you need both because sometimes people don't trust just like you know great reviews. This is amazing. — They want something specific. — They want to know what they like. Specific. Yeah. — So yeah. The best reviews are not just I really love this. It's I really love this particular functionality. So, um, — so funny enough, like someone else the other day left a review actually yesterday on the Chrome web store saying it used to take me 20 to 30 minutes. Now, you're right, but now it was talking about people love a story. So, that's so nice. Like, this is the problem I had. This is the solution. This is how — I read that and I was like, wow. I sent to Charlie, you know, the screenshot and I was like, this is awesome because — it shows that someone else find the value that we find from our tool without us knowing who this person is. — Like I don't know. — And they give actual numbers and time to it. So it's more quantifiable. — So that's the qualitative part of the social proof. So when you are adding this, put it first, you know, on the front of your website. Also, don't just rely on the Chrome Web Store because you can control it. they can put it also on your website and what Charlie was saying which is the upgrade page. — Yes. — Yeah. So um yeah and I suppose the other advantage of having like an in between upgrade page is uh we all know with sort of Chrome extensions it can take some time to make any update because you know you guys have a fantastic review process which makes you really secure which is a good thing for customers. — We try our best. Yeah. — We love it. It's good. But sometimes you want to test some pricing things very quickly just to be like, "Okay, I have an idea. Can I just very quickly test this or you've got suddenly an influx of users? Does uh does this pricing structure work? " And having an intermediate web page where you kind of control updating it very quickly. Yeah. — Uh I think has benefited us quite a lot and helped us kind of get into that — good maybe pricing structure. I mean, we're not there yet. — We're still learning. We're always experimenting, but uh — there is no I sorry I don't think that there is a a linear path to your pricing and monetization strategy. — I think that there is what works for you, what works for them, what works for the other person. Each app will have its own story. — Yes. — But we learned that there is nothing better than just putting out there something and seeing what happens. better than strategy. It's better than research. It's better than talking to customers. The best thing is to have your own page where you can control what happens. You put it out there and seeing that actually people pay. — Yeah, that is like Charlie said the ultimate validation of you know making something that people want. — So we recently also tested a medium plan. So you have your free, you have your pro and then we added like a capped one every month which is a little bit cheaper to see what happens. So you're giving options to the people. and have the annual seeing what happens. People convert, great, let's not touch it. — So yeah, that's kind of like has been really interesting. — Yeah. — I wanted to ask you about internationalization. I think it's easy to build something that just works in English, but what have you found? Do you have users that are trying to use your extension in other languages? And are there interesting challenges you had to face there? I — I think so. Yeah. So when we first started um majority of our users were sort of English speakaking. Yeah. But as we sort of grew um got organic growth um we started getting people from all different places in the world and then uh suddenly we had to support all different kinds of languages and we got some maybe some numbers on the sort of be interesting. — Yeah. And it's interesting because the numbers follow usually when you get like a spike of users from say like Korea you know usually is followed by someone a creator or some newsletter or a Tik Tok video that someone created on this and it's really interesting to understand how the audience follows. So about 10 to 15% come from like Asia like Korea and this kind of like side of the world. We have 20% of users that are in Middle East which is really interesting because it's a complete different language and we are an AI based product. So we also had to figure out all the difficulties to kind of like translate these languages so they make sense to the user also. — I think so that that's really important like certain languages unfortunately are easier to um process than other languages. So sort of the quality of the AI output is different depending on the language. — Exactly. — And we are all about improving the pro

### [25:00](https://www.youtube.com/watch?v=7i2N7Hqyqyg&t=1500s) Segment 6 (25:00 - 30:00)

you know the output. So we need to validate this — and yeah regardless of your language you should be able to use it. Uh — and then the majority of users are still English speaking like no the US, UK. However, Europe is huge. Like we had some users from the Netherlands the other day and Germany actually that I had a call with them and I was talking to them and understanding what they use it for and all of this. And I don't think there's a correlation where they come from to what they use the tool for. But I think it's always interesting when you build the tool and you start to see all these different people coming from all the world. Um, yeah, it's a really interesting journey. — Nice. And then I'm curious to ask a very technical question, but this is something that uh I've heard developers ask us before and I'm curious for your perspective on uh you know everything in a Chrome extension is running client side. So sometimes users are worried or developers are worried that users might find a way to sort of circumvent the pay wall that they have and get access to features that they shouldn't. Uh is that a problem that you have? you've solved or that you think about? — Yeah. So yeah, that's a great question. So, so when we started um — I know what you're going to say. — So, when we started, uh you know, we integrated with Stripe and um that basically updated the user to basically say uh we have a flag which says is paid true. — Yeah. — Now, when you first started using us, if you jumped into the dev tools, you could switch that over. — Is paid false? — We were testing that. — Yeah. And we were like, okay, it doesn't need to be super complicated. But as we grew and as we got more users, we needed to be a bit more sophisticated. So the way we kind of do it is and it's the standard way you do it on web apps which is the database is the source of truth. Yeah. Right. Uh so they technically you could jump into the dev tools and you could change it but the moment you reload it will revert back to your correct plan. Um, and I think one really interesting thing certainly with monetization is uh you have different tiers of users. And that's always a hard thing. And I I've never seen a sort of internal company really solve this fantastically because really it's just a matrix of uh your tiers and the features people have access to. But uh — I think — starting off really basic um and accepting that it's not you know super uh super secure in the sense users can get free stuff. — Yeah. — And then validating that and then going over to being like okay now we're going to spend a week making this a bit more secure because we've got more users and there's more chance of this happening. — Yeah, that makes sense. I feel like it's very much just sort of picking your battles like to begin with we just want users and so we don't even need as you said to necessarily have the monetization working and then we get the monetization working but maybe it's possible to circumvent and then kind of the last hurdle once everything else is working is maybe we can add some more checks and balances here. — Exactly. — I like to think about like as a founder or as a builder as a developer you need to earn the right to have that problem. So your first challenge is to just build it. The first the second challenge is to get people to pay. If you earn the right because someone paid now you can worry about the next problem. — Yeah. — If no one is paying yet then why would you worry about even the problem that is not there yet. So it's almost like earn the right first to you know to unlock your problem. It's like playing like Mario games, you know, when so you have your level and then you win against the big guy in the end of the level, but then you know the princess goes to the next castle, so you need to go to the next castle. So we think about building a product or a company in the same way. You have like levels and each level gets harder because the problem problems get harder. — Yeah. — But don't no point to jump to the next level if we haven't, you know, fixed the first one yet. And also if I also think if a developer can circum circumvent your uh systems I'm impressed like you can have it for free. Uh I respect your curiosity of diving into those dev tools. And I think it's usually the minority, right? Most users who are using your extension aren't going to have that technical ability and aren't going to try and do that. — Which also shows and we are a balance because we're like, you know, the black and white of the two sides of the coin, but developers sometimes thinks in develop think in developer terms. — Customers are not always developers. Yeah. Especially for a lot of Chrome extensions. — Yeah. So don't think what is good for you as a developer might be good or the same way as a customer is thinking. — Yeah. — Right. You know it's like — I think Yeah. — We were in a call with someone that we got them to open the dev tools and they were like — it's Yeah. It's like opening a you know going into Nano you know suddenly this entire world and it's completely normal to us but um yeah quite a lot of the time. Um I think yeah you do have to consider uh there are things you know that a lot of other people may not know

### [30:00](https://www.youtube.com/watch?v=7i2N7Hqyqyg&t=1800s) Segment 7 (30:00 - 35:00)

and you can't assume that. So if you just ask them to especially if you're in like a customer uh customer support ticket or something uh you've got to be like very kind of basic and assume almost knows — zero knowledge I suppose. — I was just thinking about monetization. So in the first 60 days so in under 3 months we actually hit the 10,000 users. — Yeah. Right now we are in 15K something like this. Within the first 60 days we actually managed to get a conversion rate because not every user is paying which is also something to understand about you know the way we build our tool and many other Chrome extensions out there. It's a premium model they call it. — It's majority free users and then it's almost like you use those free users to make the product amazing and justify why would you pay for this? — Yeah. Do you have any sort of numbers for like what's a reasonable expectation for free users versus paid users? — So, I've been researching a lot of this from other tools out there beyond like extensions like mass market tools and — and I think anything above a 3% conversion rate is — so incredible. And of course, this is just like big numbers doesn't mean that is right for every tool. So like please don't quote me but it is really good if you get a tool that converts between 3 to 5%. We are actually just about two. So not far but also not there you know. — So we are still working on this and that's why you need to test your pricing your annual your middle plan your social proof all these different things. — It's just a big kind of experiment you're doing and you're constantly experimenting and see okay what dials and knobs work when you tweak them. And I think also something really important is the nonscalable part of making your product monetiz to be able to make money from your product. Yeah, sorry I mumbled my words. Doesn't matter. We are live here by the way. Uh and I think that the good thing is if you see an email of someone from that signed up, — yeah, — send them a message. Send them an email. Send them a, you know, a Reddit message. send them a LinkedIn message, wherever your audience is, hang out where your audience is. And I think this is kind of like a really underappreciated thing. But if you send them a message and you make sure that they know it's you and not like, you know, we don't have any automations in messages yet, we are small enough to be able to do it. — I think that's a really important um important point in the sense probably you probably don't want to rely too much on automation or AI reply. — Be yourself. — Yeah. simply from the user side, they could probably um they could probably recognize an AI email, but also from your side that you start to like we know users almost from their emails. — I know exactly what a specific user will ask for because I've been seeing, you know, their messages beforehand. — You've already seen like the comment they left on Reddit and the web store review they left and then you're like, "Oh, it's great to meet you over email. " And like a week ago, I saw that someone hit the free limit and I sent a message to Charlie. It's like upgrade is coming. — Oh, nice. — Upgrade actually did come. — It was a bit uh Yeah, it was almost like magic. But yeah, having that kind of almost first name basis, you understand your users, I think, is really important — and also it gives you an idea of what different type of users want from your product and which ones you should focus on. You don't build everything for everybody. Yeah. And that's actually the super hard thing is not how to add features. Like we can add features every day all day apart from when we're doing a podcast. — Yeah. But — that's a — it's really hard to know what not to build. — That Yeah, that is quite hard. Like uh when to say no is really important. Um — but be honest to your customers. That's also part of making money from your app, you know, telling them we are not working on this right now, but I really appreciate what you're saying and I'll let you know if we build it down when. — Yeah. The I guess the road map can always change. It might not be a priority now, but it might be a priority six or 12 months down the line. — Exactly. Yes. — And I think we originally saw a lot of kind of like at Gmail emails that were signing up. And then lately in the le in the last two months, we started to see a lot of more kind of like at company something. com kind of like joining and signing up and trying the tool. — A lot of them are on easy to find because they are real people. it's not some you know um whatever email. So you also start to learn what different requirements different again type of users need. So do these people at companies need the same that me myself at home? — Yeah. — What do they use the app for? Do they need something for us to add so that they can use it as a team in a company? And that unlocks a whole new way of looking at a Chrome extension because — and how do you think about sort of that opportunity space between sort of consumer and enterprise because I think the wants are very different. Uh does one feel like a bigger opportunity? Is that product dependent? And like do you

### [35:00](https://www.youtube.com/watch?v=7i2N7Hqyqyg&t=2100s) Segment 8 (35:00 - 38:00)

just pick one and focus on it or do you try and split yourself between the two? — Yeah. So, so I I think maybe starting off going, "Oh, we're going to go to B to C or B2B might I think maybe that might be a mistake in the sense, okay, have you built something useful first? " Yeah. — And then find out where — where does it fit? — Yeah. So, so right now like uh you could class us maybe as a consumer app, but really they're using it for business purposes. — So, it's what we call a proumer. — Yeah. Proumer. Um and Yeah. And then the idea is if they really like it and they find it's really useful for their business, they should expand to the rest of their team, business because then that would make them all more productive. — Um, and I think at least internally we always think about growing this exponentially and that's what we want to do with prepro because we see where at least the opportunity is going internally. Yeah, — but I think even though we want to get every company in every, you know, organization and every enterprise with a B2B deal with pretty prompt, it takes time. — Mhm. — Things are not magical. It's a lot of like unscalable, you know, things. And I think we would never unlock that as a company, as a product, not just us, but anybody out there also. until you unlock the user for a single user. Unlock it for a person before you brand because a brand is a bunch of people together. — So like — I really like Branches, the co-founder and CEO of Airbnb. He says like it's much better to have a 100 people that love you — Yeah. — than a million people that sort of like you. — And that's always stuck with us and it's always in our mind. Get a hundred people that really love what you build is extremely difficult to do. If you do that, — then you can scale to the millions and to everything else. But first, make it and we didn't talk about paying. Talk about loving. If you make someone love what you do, it's much easier. The payment is just like a small step. — Yeah. And I don't think we could have wrapped that up more perfectly. That's such a poetic way to end the episode. Thank you so much. I'm curious if users want to sort of give your extension a try and to learn more, where should they go? — So, they can go to prettyprompt. com. That's pretty-prompt. com and uh they'll be directed to the Chrome web store and you can install it. And like I said, you don't even need to sign up to uh to use it. Once you click on it, it will take you directly to one of your AI providers and you can start using it. — And any questions or any ideas, any thoughts, just, you know, hit us up online. We will talk with us. — We'll reply in 15 minutes. — That's a guarant seconds. — This was kind of a different style of episode asking the team here some questions, but I think it was really nice and we really got to talk about monetization in detail. If you have further questions or other ideas of what you want to see, then feel free to leave them as a comment below. Otherwise, thanks for watching. Thank you both so much for joining us and we'll see you next time. —

---
*Источник: https://ekstraktznaniy.ru/video/38012*