| Readwise URL | https://read.readwise.io/read/01kt2ksc4pnqq9n0kfkp79v038 |
|---|---|
| Readwise ID | 01kt2ksc4pnqq9n0kfkp79v038 |
| Date | 2026-03-04 |
| Author | youtube.com |
| Category | video |
\n\nSource: https://www.youtube.com/watch?v=E_rX4JJrYkY
This is how every indie hacker shouldgrow their apps.>> This is Mickey, a guy from Spain wholaunched his app just seven months ago.And today, he’s already making over40,000 MR in less than a year. This oneis going to be fun. Let’s dive in. I’mPat Walls, and this is Starter Story.All right, real quick before we dive in.Mickey is about to break down how heBroke 40K a month using Google search.
But if you’re watching this andthinking, I want to build something, butI don’t have an idea yet. Well, I’ve gota free template that will help you findone. Link is right down there in thedescription. But more on that later. Allright, Mickey. Welcome to the channel.Tell me about who you are, what youbuilt, and what’s your story.
Hey, Pat, I’m Mickey. I’m the founder ofLate Social Media API that helps brandsscale and, um, make content acrossmultiple platforms. We went from zero to30K MR in just 7 months by focusing onGoogle search ads.
All right, cool. Well, I’m reallyexcited to talk about it. We’re going totalk a lot about Google today, which Ilove and I’m very excited about. Butfirst, before we do that, I want tounderstand what did you build? What’syour product? What’s the pricing model?How does it work?
So, Late is a social media API. What weDo is wrap around the official APIs,Facebook, Instagram, Twitter, LinkedIn.
We go through all the platform permissions. We, uh, basically pay for thehighest plans so that you have access toall the features available, and, uh,we hand it to you in a single API thatbasically has access to everything, anduh, without limits. We charge our usersfor connected accounts. Basically, wehave three plans. We have the bill, sothe users can just start building theirintegrations. We go through acceleratefor more accounts. Then, our unlimitedplan allows you to connect as manyaccounts as you want. So, currently, wejust crossed 40k MR. We started around 7months ago. So, we have been growingpretty fast, and, uh, well, the firstclients started landing around Augustlast year, and today, we just crossed 4KMR. So, our total signups are 50K, uh,since launch. We have more than 700paying users for now, and our churn rateis less than 10%.
That is pretty amazing. You were able todo 40k MR in really less than a year.That’s really, really impressive. I’mexcited to talk all about Google and howyou did that. But before we do, I wantto understand how you got here. Whatwere you doing before? Why did youdecide to build a SAS business? How doyou get here?>> Yeah, beforehand, I spent over 5years experimenting with startups. Istarted an online tech company calledAcademy here in Spain. We raised venturecapital, and today it has more than 100employees. So, uh, around a year ago, Igot deeper into building SAS productsand recognized a gap in the market.
There was no simple, affordable API forsocial media automation, and well, mosttools were either too expensive, toocomplicated, or lacked good documentation,which is essential for developers. Sowell, this led to the idea of buildinglate, an API-first platform designed tobe developer-friendly. I startedspending my time on proven channels likeSEO and paid ads, which is basically whathas allowed us to hit 40k MR in 7months.
Okay, cool. So, we’re not even going totalk today much about how you built oranything like that cuz I don’t thinkit’s all that important. You have areally cool approach here where youbasically said, “I can build this. All Iwant to do is focus on distribution.Focus on how this thing can grow.” Andit did. It got to 40k MR in less than ayear. What was a high-level strategy thatyou had when you went into this, and youdecided to build this? What was yourplan to grow it quickly?
So, basically, we saw that social mediais noisy. It’s oversaturated. And whileorganic reach is dead for mostbusinesses, and it’s very, very difficultto work on it. There are people who areactively searching for social mediascavenging APIs or Twitter automation tools,that they have a 10x higher intent toBuy. So uh we are actually herecapturing the intent that red existsrather than are trying to create itthrough viral content. We focus 100%high in channels like SEO and go searchads because they have predictable uniqueeconomics and uh don’t depend on lack ortrends. We use data from bothchannels to optimize landing pagesmessaging and we reinvest all ourrevenue into scaling what works.
Mickey’s story is just insane. 0 to$40,000 a month in just 7 months iscrazy impressive. But here’s whathappens when you get to this sort oflevel. Once you start having to closebigger customers, things start to gettricky. It’s not just about whether yourapp works or not. These customers wantto know if your SOCK 2 or ISO 2701compliant. And if you can’t say yes, thedeal is probably dead. This is why wepartnered with Vanta on this video.Vanta automates the painful parts ofcompliance so you can keep scaling.
Instead of drowning in paperwork, ithooks right into your stack, runscontinuous checks, and keeps all theevidence clean for the auditors. Vantahandles it so you can stay in growthmode. Over 14,000 companies like Rampand Writer already use Vanta to provesecurity in real time. And right now,Vanta is offering $1,000 off for StarterStory viewers. So, if you’re scalingfast and want to land customers withoutthe compliance headache, check out Vantaat the first link in the description.
Huge thank you to Vanta for thispartnership. Let’s get back to theepisode. Okay. Yeah, and that’s a reallyreally good insight. A lot of peoplethink when they get started, I need tofigure out 10 different channels. Igoing to have to do Reddit. I got to doorganic social. I got to do a waitinglist. I got to do all these things. Ithink smart founders like you pick onechannel, then maybe pick a couple ideasof what could work, but then ifSomething sort of starts working, theygo all in on one channel, they get it to40K MR, and then like you said, you getit to a million ARR. That, I think, is theright way of thinking about channels.
So, let’s talk about the channel thattruly worked for you, which was Google.I know there was a little bit oforganic, but then also you were runningpaid ads as well. Let’s do a deep diveinto this one channel that worked foryou.
So, for Google, basically, we createcontent on our website that is usefulfor our users or our target users. Andsome of this gets on Google and ranksorganically, and some of it doesn’trank. So, we run ads for it. Sobasically, it’s the same content that isuseful for both purposes. Obviously, weprefer it if the pages show up for theorganic Google searches, but sometimesit’s not possible. That’s when we runads on it.
Writing one article or lots of articlesthat can organically rank, and then youcan run paid against. So, you only reallyhave to do one sort of action. Let’sdive into the organic stuff. Can youbreak down what works in on the organicside? Basically, for the organic side, wejust focus on very bottom of the funnelkeywords. We use tools like hrefs tofind keywords with 300–800 monthlysearches that have low competition, butthat they are very, very high intent. Wejust target terms like social media APIscheduling API for Twitter and thingslike that. Then, we create content. Wewrite in-depth guides that answer theuser’s intent. For example, how toschedule Instagram posts automatically.
And, we also try and maintain a cleansite structure, fast load times, andmobile optimization, proper metadata. Wealso do link building. We’ve built backlinks through tools and also by reachingout manually to other people and payingthem. And, we track everything, all ourRankings with hrefs and Google searchconsole and basically constantlyoptimize our CTR based on the results.
This has already produced more than 8KMR in organic revenue.>> Okay. So you write these articles, theyessentially rank for free throughGoogle’s organic search. So thisgenerates around 8,000 in new customersper month. What about the rest of therevenue? How’s that happening?
So they happen through Google search adswhich are our guaranteed revenue engine.So while IO compounds over time, paidsearch gives us immediate results andfeedback. So here’s our strategy. Webeat on high intent commercial keywordsthe same that we are trying to rankorganically for. We create landing pagesfor each one of them. So each ad grouphas its custom landing page tailored tothe keyword and the intent. For example,Instagram API has a specific page forInstagram. We automate everythingthrough uh Google ads. We use automatedBidding, which sets a target CPA at $120,which is our average catch threshold.
Google’s algorithm optimizes bits to hitthis target. And on the ad copy, webasically focus on value. For example,scale 30 days of social media in 1 hour,or reach 100k followers fast with ourAPI. On the testing side, we docontinuous AB testing of headlines,descriptions, and landing pages.
Everything tracked through Post Hook.Currently, we spend around 2.5 in revenue for everydollar spent. Monthly revenue from paidsearch is around 15K MR.
I love what Mickey is showing us rightnow. He just found one channel thatworks and he’s going all in. But here’sthe thing. Before you get into growth,you still need to build something. Andif you’re watching this and stillfeeling stuck on that step, well, I havesomething for you. I put together ourfree business idea template, and itwalks you through seven proven methods.
To find painful problems that peopleactually want solved. In under one hour,you will map out 21 problems worthsolving and then pick the one that youshould go build. Sometimes the bestideas are hiding in plain sight. Andthis template gives you a simple systemto spot them. So head to the first linkin the description, grab the templatefor free, and potentially find your nextidea. All right, let’s get back into theepisode.
Okay. Well, I mean, I likethat. It’s cool. Like you can just dothe organic thing and do this 8K andyou’d have 8K MR, but you’re sort ofagain treating it like a math problem.You’re pouring a little bit of gasolineon the fire and when you see certainkeywords working in paid, then you’rechanneling maybe more articles or morekeywords around organic and then thosefeed them back. So, what I think ouraudience will love to hear from you,could you break down this strategy stepby step? Imagine you have a brand newSAS. It’s ready to go. You have an MVPready to go. It’s a decent product.
Maybe you got a few customers. Whatwould be your step-by-step process toget started on this Google strategy in2026? So, for step one, you justbasically find your core keywords thatare high intent and are very, very closeto your value proposition. You can useedge refs, you can use SEMrush. I meanthere are some very, very core keywordsthat you don’t have to search for. Youalready know that they are very relatedto your product. For example, for us,it’s social media API. It’s very, veryobvious. So, um, you know that thesekeywords are being searched, and youbasically create a spreadsheet with them,the monthly search volume that they have,and the difficulty, and you focus onthese keywords and nothing else, andbasically try and work towards rankingfor them within 3 to 6 months. For step
two, just focus on creating content forthe keywords that you’ve found. MakeSure that it’s high-quality content. Canbe a blog, a landing page, or just a freetool. And well, just add a big CTA toyour signing page so that you getsignups for this traffic. And basically,then you can just add internal linkinglink to other pieces of content on yoursite. And you try and do that through, uhthe first week. So you try and push 10to 15 high-quality articles in the firstweek to build momentum. It’s alsoimportant that you don’t create too muchcontent and also, Google doesn’t likethat. So it’s better to have qualitycontent. For step three, you just startlaunching your ads with the data and thecontent that you already have. Sobasically, set up a Google Ads account,
create some ad groups based on thesekeywords that you are also targetingwith organic content. You create threeto five variations of the copy. Justfocus on value propositions more thanjust stating the features. In step four,just optimize and scale based on theData that you have. You identify what’sworking. You identify the keywords thatare bringing new customers and you tryand optimize for this. You just can’tforget about the rest because the onesthat are converting are the ones thatreally matter to you. You can post theirperforming keywords and just double downon the high ROI keywords. And then youcan just start testing. Update theheadlines, update the copy and add CTAsbased on what converts best. Then justtry and increase the budget month over
month as long as the return on this adspend is working. In step five, justtrack and try and improve your cost ofacquisition. So now you can just setdashboard either in Google Sheets, datastudio, or any tool that you like. Youcan start tracking the cost ofacquisition, the return on ad spend, therevenue, and the traffic per channel. Andthen you can just start benchmarkingagainst what’s working. Good CAC is 20– 40% of the lifetime value. And your goalis basically to keep CA under 30% of thefirst-year revenue. Every month, you canjust AB test new keywords, add copylanding pages based on what worked lastmonth. So, that would be my five-stepplaybook if I would be starting overagain.
Okay. Well, thank you for sharing that.Great playbook. I love Google. This isawesome. Thanks for breaking it all downfor us. Let’s switch topics a little bitand talk about the product that youbuilt. Could you give us a demo? I justwant to see what this thing looks like,how it works, and yeah, could you showit?
Sure. So, our app is nothing fancyreally on the UI. We just have uh socialmedia. Basically, everything that youcan do through the API, you can also dothrough our UI. So, it’s very verysimple. Um, you can just from here createnew posts, and uh yeah, basically justselect the accounts that you want thiscontent to go to. We also have here theOption to connect all your accounts andhandle all your accounts from the UI.
Here you can also create new API keys.So, as a developer, you can just basicallyhandle all your APIs from here. We alsohave an inbox feature thatbasically allows you to handle all yourDMs and comments through the UI. Therest of the platforms you can set up webhooks. You get notifications when apost fails or when a post succeeds,when an account gets disconnected. You canalso check all the logs from yourpublishing activity to see if youhave errors or if everythingis working correctly. From here,you can just access the API docs, whichare very important becauseyou’ll be building an integration,and this is what we try to spend themost time on, so that it’s verystraightforward for you to learnand start using.
Okay, cool, and on a similar note, how didyou build this? What tech stack did youUse? What runs the business? How’d youbuild it?
So most important thing that we use iscloud code so we can code much faster.We also use data fast. It helps us trackwhat’s bringing revenue and what’s not.We also use Google ads spend, uh, up to200 per month. Andthis is essential for us for insightsmonitoring on the organic side.
Okay, cool. Well, thanks for sharingthat. Last question that we ask allfounders who come on to Starter Storywho built cool stuff like you. Whatwould be your advice to young Mickeybefore you got started or for anyonewatching that wants to get started andbuild cool SAS like this and learnGoogle? What would be your advice?
My advice would be to just stop trying to be everything to everyone. When Irealized that basically picking onegrowth channel, getting really good atit. In our case, it’s paid search. ItWas 100 times better than being mediocre at 10 channels. So, my advice would bejust pick your single best level forgrowth, get obsessed with it, andmeasure it relentlessly. And that’sgreat advice. As I kind of mentionedbefore, you can win with just onechannel, right? You can build amillion-dollar business. You can build a$10 million business, maybe even ahundred million dollar business withjust one channel. So, if anyone’swatching this and you take one thingaway from that, here’s proof of that.
Thank you, Mickey, for coming on andsharing. Awesome business. Hope to seeyou grow.
Thank you for having me.>> Okay, producer Gus, what’ you thinkabout that one?>> I thought it was really cool. I I thinkwe talked to someone else recently likethat was very similar but like it’s verysystematized right he has like a supersolid system I think for you knowWhatever finding the keywords, rankingthem, and then running paid search onthem. So I thought that was reallyawesome, just to hear kind of that systemand I feel like so many of the greatfounders we talked to like havesomething like that, where they have likethis methodical approach. Like, we don’tget lucky. Like, it’s all math, and Iwas like, wow, that’s awesome.
Yeah, that’s what I like about Google. Ithink a lot of people are talking aboutorganic TikTok and these sorts ofthings, or AI, and you know, all this, butat the end of the day, Google is like avery, very tried and true acquisitionchannel for a lot of businesses. At theend of the day, this is a greatdistribution strategy, or marketingstrategy for people who don’t want togo on camera and want to, you know,maybe they’re a little bit more of afinance type, and enjoy putting togetherspreadsheets of keywords, and seeingintent, and looking at the numbers. NotEveryone is meant to be on camera andand create viral content. This is a real solid plan that he puttogether and his playbook was especially
good. I think you can really do thiswith pretty much any SAS, specificallyB2B SAS more so, but I really think thatthis could work and is a great firstchannel. Yeah. And then the other thingI was thinking about was like onTwitter, you’ll see how many people aretalking about how like Google search isdead or like Google SEO is dead, kind ofthing. I think we’ve even done like somevideos on the Bill channel on thattopic. But I don’t know. I guess, what doyou think about that? Like, is he’s kindof showing that maybe it’s not this astrue, or what we’re like, what do youthink?
Here’s the truth: Google search is notdead, but informational queries are dead.So, if you search not dead butessentially dead if you search Googlefor how to, uh, create a great socialMedia post that you’re probably notgoing to rank. It’s going to be harder torank for that, and those people arecoming for information on how to createa good social media post. They’re justgoing to use ChatGPT for right to get amore customized, way better answer. Butif they’re searching for things likesocial media scheduler, they may putthat into more so into AI, but at theend of the day, it’s still the same.
Like they’re going to put that intoGoogle, they’re going to put that intothe AI. They’re going to click inthrough to your software and buy it.Yeah, there’s an argument that one dayyou’ll be able to put into AI and it’llwrite all the code and create the wholeapp for you. But right now, peoplestill need tools, and intent-basedthere’s formational queries, and thenmaybe you call it intent-based queries.People are still searching forthings like I need insurance, I need a socialmedia scheduling tool, I need someClothing or something like that. So theSEO is dead, is more around informationalbased queries.
Yeah, I mean, I find myself using likeChatGPT a lot more first search. Butsometimes, even when I’m stilllooking for a tool or something tobuy, I almost still go to Google insteadof chat, because I don’t know. I feel likeGoogle has something it’ll really showme versus chat, which will just list outwhatever the answer is without necessarilyproviding a link. Anyway, so yeah, Ithought that was interesting to see thatthat still works if you do it kind of inthis mathematical way. Okay, cool. Well,SEO and Google still works, but you need
to build something. So, I willshamelessly plug Starter Story Buildright here. If you want to find thatidea that you can eventually put intothe Google Playbook, check out StarterStory Build. I’ll put a link in thedescription for that. We will help youcome up with your idea, build it withJust AI tools. You don’t even need toknow how to code and get it in the handsof real customers. I’ll put that linkdown there. Thank you, guys, for watching.
We’ll see you in the next one. Peace.