42
68 Comments

We bootstrapped our SAAS to $500k ARR in less than 2 years after failing two times, AMA!

Hi everyone,

I'm Pierre de Wulf, and with my old friend Kevin Sahin we built ScrapingBee, a web scraping API.

Following people on IndieHackers was always an inspiration for us and today I'm happy to share our story.

Today ScrapingBee is doing well, we managed to reach more than $500k ARR in less than 2 years and are still growing strong.

But it was not always the case.

We began our bootstrapping journey by making lots of mistakes, building useful stuff making no money, and also useless stuff ... making no money.

But we also learned a lot along the way.

I'd love to spend a little time today to answer any questions this community would have!

  1. 5

    You mentioned failing building stuff that was useful. I think there's a misconception amongst creators that if you build something that's incredibly useful, you'll have a successful business. In contrast, I've learned that a successful SaaS business is mostly about having something somewhat useful & figuring out how to distribute it.

    What lessons along the way have you learned about marketing & distribution? I mostly mean acquiring users / customers. If you knew what you know now, do you think you could have been successful in any of your past endeavors where you built useful products?

    1. 6

      I totally agree, first-time founder / builder tend to greatly underestimate the distribution part.

      The funny thing is that we knew that distribution was hard. We had read lots of books / podcast / video about entrepreneurship and knew that "build it and they will come" was mostly bs.

      What we did not know was how hard it was.

      I remember thinking "yeah, we'll just have to do this and we'll get hundreds of customers". Never happened.

      Knowing what I know now, I probably wouldn't have built those products without having figured out distribution first.

      1. 3

        Follow-up question: what worked well for you to get enough users for the 10k/100k/500k ARR in terms of distribution?

        Are there all different channels/methods or you've found a single golden one?

        1. 4

          From $0 to $500k we always did the one thing that worked: SEO.

          That is our main acquisition channel, and we didn't wasted time trying something else ever since.

          1. 1

            Hey Daolf, thanks for sharing your story, and congratulations!
            Could also please share some lessons about how to do SEO efficiently? Any tools?
            I think some IH developers are also learning about SEO :-)

            1. 4

              Hi there, the Ahref course about blogging for business is a gold mine:

          2. 1

            Impressive stats.

            I would be careful to just have one distribution channel, especially one that's as fickle as SEO, plenty of stories out there where a change to the black box google made people go out of business.

            Redundancy is always good :)

            Just some food for thought 🤙

            1. 3

              You are correct there is always a risk. But when you do SEO correctly you won't lose all your traffic overnight.

              You might lose 20% or 30% after some core update, but more is unlikely if you have enough content.

              We will start working on another channel in a few months, but being a team of 2 it would have been impossible to do that sooner.

  2. 2

    Great work Pierre!

    I read in the other comments that SEO has been your main channel for new customers. I've also been trying to grow my readership on my blog for my social scheduling platform (https://pallyy.com) - and while it's grown to about 60k readers per month it's just not converting into customers.

    Maybe I'm not focusing on the right topics, but would love to hear your tips on creating nice content that converts well!

    Cheers.

    1. 2

      Congratulations, 60k organic visitors per month is no small thing!

      Of course the "intent" should also be taken into consideration.

      According to Ahrefs, your most popular blog post is "How to Add More Than One Photo to a Instagram Story", you should ask yourself is people looking for this kind of information would be interested in your product.

      Without thinking too much about it, for Pally, those are the keyword / phrase I'd try to rank on:

      • when to post on Instagram / Twitter etc [Agency probably want the best ROI for their POST]
      • tutorial on what makes a good post on Insta etc
      • how to schedule on Insta etc
      • what is the perfect frequency for Insta etc

      You could also write a big guide with everything you know about social marketing and try to distribute it / launch it on PH.

      Just my 2 cts

      1. 1

        Yeah, I've got a lot of random learning how to use Instagram posts - but they just don't work.

        Scheduling & agency related are good ideas though, and I do have a few but I'll focus more on those. It's just harder to rank for those keywords because of the competiton.

        Good idea on the guide, actually I've already written one on how to build a social media agency so maybe I could use that.

        Thanks!

    2. 1

      Hi Tim, congrats on 60k monthly readers! Plus, it looks like you have a solid product that's quite affordable. A couple thoughts came to mind as I read your post, so I hope you don't me jumping in and sharing them.

      1. Unlike ScrapingBee's audience, there's a good chance your target audience isn't coming to search with the intent to buy.

      Or

      1. Your target audience already has a social media scheduling tool that they're happy with.

      If #1 is true, you might benefit from focusing on lower funnel search terms such as "social media scheduling software" or "how to choose the best social media scheduling software". If that doesn't work either, I'd go back to the drawing board and add other distribution channels that are more aligned with how your target buyer chooses social scheduling software.

      If #2 is true, I think a few conversations with social media managers / agencies / freelancers will tell you what you're up against and what it would take for them to switch to Pally. For example, you might need some comparison content such as "Pally vs. MeetEdgar" (illustrative example). Speaking with your target audience might also help with unlocking #1 above (i.e., finding distribution channels that are more aligned with how they make buying decisions)

      I just wanted to jot down those thoughts in case they're helpful. Keep at it and good luck!

      1. 1

        Thanks for taking the time to give me that feedback mate!

        I think you're right in #1, the audience isn't the right type for the most part. I'm working on this, and have started adding "vs." pages as you mentioned in #2 to possible help generate some traction that way.

        Thanks again!

  3. 2

    Thanks for doing this Pierre! Couple of questions. I want to apply to Tinyseed but we've only just got a product and no revenue yet.. did you have revenue when you applied?
    And related - it seems that everyone on Tinyseed loves it.. has it been good for you?

    1. 2

      Yes, we had revenue when we applied, I believe $3k MRR.

      I also believe that companies got accepted with much less MRR than that.

      TinySeed was a game-changer for us in many ways. The money, the mentorship, the community, everything was and still is awesome!

      I think they now run 2 or 3 batches per year, so I'd advise you to apply now and then later if you, unfortunately, get rejected.

      This year one company got accepted on their third try ;)

      1. 2

        This year one company got accepted on their third try ;)

        Yeah, I know who that is :)
        There's another round in 6 months so aiming for that with some revenue!

        1. 2

          Good luck with this! 🤞

    2. 1

      If I remember correctly tiny seeds suggest having at least 500 MRR before applying.

  4. 1

    @Daolf i have one question here.
    I am working on my web scraper for getting lead data but for that I need multiple concurrent connections.

    How can I make it possible? I will highly appreciate your support on this. 😊

  5. 1

    @Daolf what do you use for monitoring or E2E tests to ensure your apis don't have bugs / regressions?

    Curious about tools used by api-focused startups.

  6. 1

    Hey Pierre,

    No questions here. I already got a good answer from one of the questions you answered here. I just want to say Big congratulations to you and Kevin and a big thank you for all the helpful articles and tips you have shared in the past. I saw this AMA when scrolling through indiehackers and thought web scraping, I know I read a post about a web scraping tool before I started indie hacking.

    Then I went to search on google with some scrappy keywords and found the article again this one https://www.freecodecamp.org/news/12-months-3-products-some-mrr-and-one-irrigation-pivot/ and I was so happy because the name on the article was the same and because that was the article that made me aware of the indie hackers community which gave me a huge boost to try this out one day. I remembered vividly about you guys making your first $1k and going to work on your father in law's farm that was basically one of the keywords I used to find the article, lol. Currently taking a leaf from all you did and indie hacking my way at the moment.

    Thank you so much again for all you put out there and wish you continued success on scrapingBee and all your projects.

    Cheers

    1. 2

      Hi there.

      You've made my day, sincerely.

      I've never been so happy and touched by a comment on IH!

      You never know who will read your own story when you share it and my goal was to show that:

      • it was hard
      • but it was worth it

      And if reading that article helped you find some motivation, then 😍😍😍😍

      Good luck with your journey, and thanks again for the kind words.

      Pierre

      1. 1

        Thank you for responding Pierre,

        Your response just made my day too!

        Thank you so much for your encouraging words.

  7. 1

    What do you think are the reasons for products not making money? What were your learnings from them?

    1. 2

      The main things I learned was that:

      • it takes a lot of time to find and develop a successful acquisition channel
      • finding a problem to solve is important, but not enough, you have to find a problem that people actually want to solve.

      We overlooked the later point so much that we ended up building a "nice to have" not a "pay to get".

      1. 1

        I see. Thanks for the insights 👍🏼

  8. 1

    Damn! ScrapingBee is goood!
    A couple of parallels with what I'm working on, BeeTrendy

    • 'Bee' in the name
    • a lot of scraping involved to find the trending info on most of our sources.

    I only wish I get even a fraction of your success. Read all the comments below, your story is inspiring 😁

    1. 1

      Hi there and thanks for the kind words!

      Good luck with BeeTrendy 🐝-friend!

  9. 1

    Hi Pierre,
    Thanks for sharing.
    What about churn rate and activation ? I feel like after acquisition this is the most difficult challenge when building a product (when you are between 50 and 500k ARR).
    What about you ? How was it for scrappingBee ?
    How did you manage to overcome it in your early days ?
    What was your custommer service process and so on ?

    1. 1

      Hi there,

      You're correct that acquisition is just one step amongst many others.

      For our previous company, our churn rate was awful.

      It was because the product was s*** and targeted the wrong person.

      For ScrapingBee it's a bit different, being an API the number one cancellation reason we hear is: "I don't need this product anymore".

      We don't really have a customer service process, we just try our best to provide the best customer service we can.

  10. 1

    hi Pierre De Wulf,

    Congrats on hitting 500k ARR! that's indeed a huge milestone!

    Curious about your first startup - universal wishlist. it seemed like a problem that everyone would have, any thoughts on what was missing there / why it didn't work as expected? What would you have done different there?

    1. 1

      Hi there and thank you very much.

      So we noticed that in order to succeed in this space we needed absolutely:

      • a great UI/UX
      • an iPhone / Android app
      • lots of share functionality
      • a way to acquire users for free

      Those were all things we did know how to do and were particularly interested to learn.

      And while the first three could have been solved by throwing some money around to pay some good developers, I have still absolutely no idea how we'd acquire users even with a perfect app.

      I still sure there is a big market and need for this thing, and if anyone ever makes it work, I'd love to read about their acquisition strategy.

  11. 1

    What is Trader Joe's like as a customer?

  12. 1

    Hi, Pierre

    Alex here, congrats for your success:

    1. what channels did you use to grow to $500k?
    2. how long it took the dev of app before you got first paying customer?
    1. 1

      Thank you Alex:

      1. 99% SEO
      2. 3 weeks :)
  13. 1

    Great achievement!
    Can you share some info on the technical side ?
    What cloud provider you are using ?
    How your stack looks like ?
    Do you using third side proxy providers ? or you build your own ?
    Thanks

    1. 1

      Hi there,

      We're using several different cloud providers from AWS to DO and use Python a lot.

      We use a combination of both our own proxy and some very particular cloud providers.

  14. 1

    Congrats on the success, Pierre and Kevin! ScrapingBee looks awesome! I'd love to give it a try!

    • What would you say was the most challenging technical part of building a good scrapper?

    PS. By any chance, do you guys offer a smaller, more "indiehackers-friendly" starter plan (even if only for a couple of months)? I wouldn't likely need 1M API calls in the beginning :)

    1. 1

      Hi there and thank you very much!

      The most difficult way was to build a scraper that can scale. That was challenging.

      We had to remove the $29 plan because the support overload was too much to handle. We're currently experimenting with the new pricing and might add back a $50 plan in a few weeks.

      If you only need a couple of credits to try our API please let me know in the support chat and I'll help ;)

  15. 1

    Are you investing? I'm still pre-launch/pre-revenue but with a solid understanding of the market and innovative AI/ NLP technology.

    I've also failed before, have ton of experience with startups and would be ideal to have someone with your experience to back the project.

    The V1 of the project is at branding.gq and I can send the business plan by email 📨

    1. 2

      Hi there,

      I wish I had this kind of money but I don't.

      Good luck with branding.gq, it seems that those kind of tools are doing well lately.

      1. 1

        Jeje, cool thanks for the words of encouragement.
        I also hope you have that kind of money soon.

        Good luck with your other two projects.
        Hint: I'm a big reader, and the thing I struggle with the most is keeping updated with Machine Learning research and papers. That might be a profitable vertical to explore a $5 USD subscription?

  16. 1

    Thanks for doing this!

    When you sold your previous company to your competitor, did you cash out anything?

    Pricingspy was more of a niche use case of scraping I believe. How did you leave the product, but not the market? What gave you the confidence? :)

    And congrats on all the progress! :)

    1. 2

      Thanks a lot for the kind words.

      When we sold PricingBot we managed to cash out enough to stay alive for around 8 months.

      So not much, but since at that time we were nearly bankrupt it was a big deal.

      When using PricingBot we were using a web scraping API and thought that we could do better. We also had a lot of experience in web scraping.

      One key thing was that Kevin had also written a successful book about web scraping in Java and was running a web scraping blog with some traffic. This alone validated the fact that we would know how to attract developper to our product.

      Something we never really nailed with PricingBot.

      1. 1

        Great! Another reason why building an audience before the product makes sense!

  17. 1

    Awesome work!

    My main question is, how do you get new customers?

    I noticed a referral program where you give affiliates recurring revenue and thought that was interesting - how helpful has that been for getting customers?

    I'm just generally curious about your customer acquisition funnel (blog, referrals, Twitter, etc.).

    1. 2

      Thank you very much!

      1. Mainly SEO and now words of mouth since ScrapingBee is now 2 years old.

      2. The referral program was an experiment. It now only accounts for a very small part of our revenue, but that is something we might focus on again later this year.

      3. Our funnel is quite simple, people search information about web scraping, they read our blog post, they then need some web scraping tools and think about us ;)

  18. 1

    Hi Pierre,

    Thanks for taking the time to do this AMA.
    Couple Questions

    1. I've noticed that you've tried to build multiple softwares before you became successful, so I'm curious, were all of these problems that you had that you had an itch to resolve or was it more that you saw other people had this issue and it looked like you could solve it?
    2. Did you build any sort of community around ScrapingBee at the beginning through build in public or post on forums letting people know you can resolve their issue like some founders do?
    3. Did you make use of any lifetime deals on sites like AppSumo, etc to help your cashflow at the beginning?

    Also, if you have time in the future, would love to have you come on our podcast to talk about it more.

    1. 2

      Hi there,

      1. For the universal wishlist, I was actually trying to build a solution for a problem my girlfriend had. I talk about it to Kevin and we both thought that it would be a fun little project.

      For PricingBot, we did it because we actually noticed that people who used the extension were using it to spy on their competitor's price.

      And for ScrapingBee, we did it this time to solve the problem we had building PricingBot.

      All in all, a quite "organic" problem-solution discovery.

      1. No community at all, we thought about it but decided to focus on something else (SEO). Part of this decision was motivated by the fact that a few people are passionate about web scraping. This is a useful tool but rather a means than an end.

      2. Not at all. Running a web scraping API is quite expensive so Lifetime Deal would not be sustainable I think. One important thing we did was to join TinySeed! This helped a lot

      Sure, why not, feel free to DM me on Twitter ;)

  19. 1

    Hey Pierre, love your journey and the focus on SEO.

    I just came across your alternative pages (https://www.scrapingbee.com/crawlera-alternative/) and I absolutely love them, as a marketer.

    Can you elaborate on the thought process for creating these pages?

    1. 1

      Hi there and thank you very much.

      Those pages were just a quick marketing experiment and unfortunately a failed one :(.

      The idea was to try to attract people looking for some much more expensive alternative, but we never really managed to make it work and decided to focus on something else.

  20. 1

    I just want to say Pierre answered a question instantly which I sent via his website. Great experience. Most websites have a form-to-mail and you wait a day or two without knowing if anyone got the message, or will reply. ScrapingBee is doing this better.

    (now if only people could video call or schedule an appointment 😍)

    1. 1

      Thanks a lot for these Jesse!

      We used to have a more accessible schedule link but spots quickly got filled with a lot of noise.

      People who really just wanted to have a quick chat about general web scraping and this was not sustainable for us.

      We're still working on finding a way to bring them back.

      1. 1

        One possibility is for you guys to have a paid or free online course about web scraping in general, as a lead magnet. People who complete the course (especially if they pay anything over $0) should get an email DRIP that leads to signups over time.

        Another possibility is to mark the "Book an Appointment" as something where people pay say $99 for a 30 minute initial appointment. Tools like Calendly let you require payment upfront before the appointment. That would likely reduce the noise. You can even sweeten the deal by saying some or all of the dollars are applied to their 1st month subscription if they become a subscriber.

        Anyway - congrats on all the things you're doing right!

        1. 1

          Those are very good ideas!

          The course thing is really something I'd like to try, but don't have the time to do it ourselves and we need to be sure to work with the right person for this.

          Maybe later this year.

          A paid appointment is something I never heard of but I fear that it would also make our biggest leads go away. A developer in a big company doesn't necessarily have quick access to a CC card, and that would add a lot of friction to the initial contact, we have already closed $500/mo deal on the first demo call 🤷‍♂️

  21. 1

    Firstly, congratulations on rebound from a failure, you're a great inspiration to new founders and aspiring founders.

    My questions are:

    #1 Was there ever a particular SEO strategy that you felt was beneficial to your SaaS, or was it the all-round performance of a combination of factors?
    #2 What particular resources would you recommend for someone looking into the marketing side of SaaS.
    #3 Did your customer acquisition strategies and marketing ever change as revenue increased (i.e. from 10k to 100k, was this strategy different from getting from 100k to 500k in revenue).
    #4 I have always been fascinated with the opportunity of joining accelerators, would you say the mentorship and community by itself provides valuable insight, and was of value to you?

    Thanks for your time, and contributions to this site :)

    1. 2

      Thank you!

      1. We haven't really applied strategy knowingly. In retrospect it seems that what we did is called the "skyscraper" technique. Writing the best possible content on one topic hoping to gain credibility.

      2. I'd go on Twitter and follow the journey of every bootstrapper one or two steps ahead of you. I learned way more following people at $10k MRR at the beginning than by reading books. The Microconf youtube channel is also a treasure trove.

      3. Nope, still the boring same things, although in the next few months we might finally try something new.

      4. Totally of value! 100% recommend I've talked about it extensively there

      1. 2

        Good luck in the future, and thanks for the response! I'll be sure to log on to Twitter!

        1. 1

          Thank you very much!

  22. 1

    Hey and Congratulations!

    Here are my questions:

    1. How did you that it was the right time to switch to another idea/project?

    2. How did you know that for example the "price monitoring SaaS" you've built will never be successful and profitable?

    3. How do you know it's the right time to switch to something new and stop wasting time with your current project?

    Thanks in advance and keep doing this great work!

    Regards,
    Radoslav

    1. 1

      Thank you!

      1. That is a very good question.

      We gave up for 2 main reasons:

      • we weren't happy anymore, almost depressed. With PricingBot, the last 3 months were dreadful. It was like anything we did was useless. We were not able to move a single metric, up or down, in 12 weeks.
      • we knew we were running out of time (read: out of money) and that we needed to work on a plan B otherwise we would have to go back searching for a job
      1. We understood that we did not know our audience and that it was a problem that would take a lot of time to solve. Time we did not have.

      2. I'd say it really depends but if I had to give one rule of thumb, is that if after 1 year you still haven't found a way to regularly acquire customers, you'd have to seriously consider the idea that you should quit.

      I'm not saying that you should reach $10k or $20k MRR after one year. But if you haven't found one repeatable way to acquire customers after 12 months, I'd say this is alarming. But those are just my 2 cents, I know there are counter examples out there.

  23. 1

    Hi, thanks for doing the AMA. Your killing it!

    I imagine a tool like yours is always fighting anti-scraping and anti-ddos measures, is that the case ? And if so, have you had any times when your accuracy has been severely impacted. How have you dealt with that from a customer communication and expectation point of view.

    I've worked on apps with a similar accuracy problem before and struggled with that.

    1. 2

      It's true that anti-scrapers are an issue, the truth is that sometimes we managed to build the correct anti-scraping thing and when we can't, well, we can't.

      We're always working on improving this part but fortunately for us, this is not as much of a problem as it could be.

  24. 1

    This comment was deleted 2 years ago.

    1. 1

      Hi there,

      No fundraise is planned.

      We're currently a team of 3.

  25. 2

    This comment was deleted 2 years ago.

  26. 1

    This comment was deleted 2 years ago.

    1. 2
      1. We were using a web scraping API while working on PricingBot. We thought we could do better.

      2. Growth hacking forum + we knew quite a bit about the web scraping industry.

      3. We didn't. We just knew that web scraping is done by almost every company to some extent. We also knew that some web scraping companies are doing more than $20m ARR (Zyte).

      4. Trial and error. Since we had no money we did not have much choice. Our past experience running some technical blog, (mainly Kevin and his web scraping blog + book), gave us confidence that we could do the same with ScrapingBee.

      1. 1

        This comment was deleted 2 years ago.

        1. 1

          Thank you very much, I appreciate it.

Trending on Indie Hackers
How I grew a side project to 100k Unique Visitors in 7 days with 0 audience 49 comments Competing with Product Hunt: a month later 33 comments Why do you hate marketing? 29 comments My Top 20 Free Tools That I Use Everyday as an Indie Hacker 16 comments $15k revenues in <4 months as a solopreneur 14 comments Use Your Product 13 comments