14
26 Comments

My startup went from $0 to $890/month in Q1 2021

I incorporated SurveySays in early January.

Don't get me wrong -- the site had been live for a few months prior to that. I had been building, testing, and tweaking since quarantine began last spring.

But official launch came in January, and the first month went...well, not so good. No sales.

It's one of my weaknesses to get discouraged quickly, so I started wondering if all my work was for nothing. This was a great idea...in my mind. I'd tested it a million times...myself. I was absolutely convinced this product was something people would LOVE who were...just like me.

Maybe others didn't feel the same way?

But regardless, I kept at it. Didn't change course for two more months. Decided to trust myself and my idea, and see how things would pan out after two more months.

As builders, we're all about monthly stats. But is one month long enough to build metrics for a brand new startup?

Hardly.

So another month goes by. Some traction.

Then by March? Holy moly! (see picture)

What did I do differently? Nothing! My users are more than 50% likely to come back and use the product again, so that's natural/organic growth. And almost everyone who's used the tool learned about it from someone else who used it -- great word-of-mouth marketing.

Snowball effect, anyone?

MORAL OF THE STORY: Give your plan some time. If you're sure about your product and you've done your homework, don't doubt yourself! One month, two months, or even three months really isn't long enough to get a sense of how your product resonates with your target market. It may be that your target users just haven't found it yet.

(My product is at surveysays.app. I'm offering a BIG discount to IndieHacker users with coupon code FEEDBACKNOW.)

  1. 3

    Impressive stuff there, Nick! Congrats.

    What are you doing for the past 3 months to attract traffic to your site, which contributed to the growth? Cheers!

  2. 3

    I'm pumped that it's working out for you! But I think Indie Hackers should generally do the opposite. In rare cases, sticking with stuff can work out. But more often than not, you end up burning resources as you become increasingly invested in the success of your product. (Sunk cost fallacy + tying your identity to it.)

    One always needs to choose between Type I and Type II errors. Since bootstrappers (and entrepreneurs in general) are biased to be short-term optimistic anyway ("this is gonna revolutionize the world and I'll be swimming in cash") we should do the exact opposite.

    Namely, kill stuff quickly.[¹] Does that mean you're going to be killing ideas prematurely that would've worked out had you done things differently (Type II error, false negative)? Yes.

    But that downside is trivial. Plus, the ideas that do make it past a stringent filter mechanism are that much more likely to succeed.

    Now take the inverse. I.e. the status quo.

    You work on stuff because you believe in it, hoping someday it'll start generating revenue. (Type 1 error, false positive).

    Now the downside is essentially unconstrained. If your idea is a dog, the probability of which increases as a function of time, you've burned through resources.

    The "upside" is that as duration increases, your level of certainty that this combination of product, market, and timing, is bad.

    Well, that's good if you're planning on publishing research in a scientific journal. But us founders don't give a shit about what's not working.

    We're in the–make stuff and get paid for it–business.

    So under those constraints being too aggressive with killing makes the most sense.

    But again, I'm pumped it worked out for you, and good luck!

    ## REFERENCES

    I wrote a piece on this, Paradigm Shift: Drastically Increase The Odds of Success.

    ## NOTES
    [¹] In a perfect world, you'd be precisely in the middle. But that's an "In theory there's no difference between theory and practice. In practice, there is." kind of situation. You're always going to be leaning in one of these directions and if you don't make the choice consciously, the choice will be made for you.

    1. 2

      This is actually incredibly insightful. If I were to push back a little, I think the outlined Type 1 and Type 2 errors actually manifest as a spectrum implicitly present in most entrepreneur's minds, while unsure where exactly to draw the line (mostly in terms of either duration or effort) when to halt a project. Do you have any thoughts or hints how to approach that?

      1. 2

        Cheers for the comment Wouter.

        Interesting conjecture. You may be right but I'm struggling to see them as anything but discrete. Think of the order of the null hypothesis (there is no relation) vs the alternative hypothesis (there is a relation).

        You can either use this scientific approach or the inverse. In this former situation, your default is I'm wrong and if you can reject that you increase your certainty. In the latter, your default is I'm right and the tiniest bit of confirmation can be used as validation. This is an example of (at least in my mind) a discrete set of options.

        But it's possible that this isn't quite what you had in mind and that I misunderstood you.

        As for your question apposite to duration, I use the framework my friend @noahkagan created; 3 paying customers, within 48 hours, without creating a product.

        Now, many people will point out that this stringent constraint will erroneously filter out good ideas (false negative), but the pros outweigh the cons, which was the topic of the previous comment.

        At this stage in my research, I'm not quite sure if I can say that it's always a mistake not to use this approach (i.e. are there exceptions?), but I'm willing to state that for 99.9% of founders its a mistake. Especially the type of audience that's on IH since they're optimizing for lifestyle, not a globally disruptive unicorn. ¹

        NOTES
        ¹ Although even in this case, I strongly suspect it's still the right approach. See this hour-long essay series on that, The Dumbest Startup That Ever Worked  — What You Can Learn From Airbnb.

    2. 1

      Good points.

      A lot of this just depends where you are in the process, I suppose. I had been building and thinking about this product for almost a year, and so I probably should have had more faith at the outset that I'd gain some traction.

      If you've actually done your research and sought honest feedback from the right people, I think I'd always suggest giving things some time before changing course. But on principle, you're right that this may not be the right approach for everyone.

      1. 2

        Yeah, I get where you're coming from. Ultimately, it's your company and you should run it however you see fit.

        I try to make very concrete statements because I think the "it depends" is more harmful than helpful for Indie Hackers. While it's true that things aren't black and white, there are significant patterns. And when we hide behind ambiguity we make it appear as if that isn't the case.

        (Not accusing you of anything, just illuminating the reasoning behind my posts.)

  3. 3

    Pretty nice but reading your website my #1 question is: Who are the people answering and why are they answering?

    1. 1

      Good question. That's covered a bit on our homepage, but in more detail here: surveysays.app/data-quality. We don't own these respondents, but have rather chosen to partner with firms that do a much better job than we could ever do when it comes to sourcing and cleaning these lists of consumers.

  4. 3

    Question regarding your product: How did you get people to answer the surveys (even on "same-day")? And do you compensate them?

    1. 1

      That's covered here: surveysays.app/data-quality. The bottom line is we partner with firms that are very good at sourcing and cleaning these lists of consumers.

  5. 2

    Congrats and thank you for posting this, it's like you are talking about me right now lol but it's assuring to read about others experiences, sometimes you are just about to give up on the idea even if like me I worked on it for more than 4 years (crazy I know), have you done anything to attract customers somehow? how people started to know about you in the beginning

    1. 1

      Well, for one, I talk about it obsessively. When I launched, I posted to my LinkedIn, Twitter, Reddit and other channels that I use regularly. I do have a large email list I've gathered over the years that I plan to leverage, but I'm going to wait until I feel the product is even better, as this is a very valuable list to me.

      Just be very public about everything you're doing. Have you heard of #buildinpublic?

      1. 1

        Hi @NickFreiling, sorry for the late reply, I do heard about #buildinpublic but not very sure what it is, I think people tweet about what they are doing, what feature they added etc and they put the hashtag #buildinpublic

        1. 1

          Yes, it just means being open about your progress while launching a new product/startup. I'm trying to do that here.

  6. 2

    Celebrating the positives with you, Nick! Way to go! I hear you on the "get discouraged quickly." Me too. Good job pressing through and seeing some results!

  7. 2

    "What did I do differently? Nothing!" Love this. It's the conceptual antithesis of "10 tricks entrepreneurs hate - find out how to grow your business INSTANTLY!".

    I'm currently trying to find just the right channels to grow our business, and even though it's been moving slowly, recently we've seen our first uptick after a long period of intense and unpaid work. Naturally, we're hoping for a similar compounding trajectory and this is just the type of motivation one needs every once in a while.

  8. 2

    Congratulations! I can especially relate to your statement "It's one of my weaknesses to get discouraged quickly". I've been spending the greater part of this year building my SaaS Adflow.ai and I have been getting positive feedback from the people using our free tier. I just implemented paid plans and – in all honesty – I fear I might get discouraged if the initial traction is low, so your post is very inspiring 💪.

    I'm curious, did you change anything in your growth tactics to generate momentum in March?

    1. 2

      Hey there @lorenzosignoretti, just check out Adflow.ai and I really like your landing page design, love how clean it is.

      What are you doing right now to promote your product / attract traffic?

      1. 2

        Thanks for checking it out! We are fairly active on a number of relevant communities, to tailor our product to marketers needs!

        In addition, we launched a relevant marketing stunt on Product Hunt last week which generated some initial traction (the free open beta has only been live for a couple of weeks).

        1. 2

          Love this marketing stunt on PH!

          How was the result(s) from this?

          1. 2

            It was well received and generated some initial users signups for Adflow. Unfortunately, we had a small bug in the first two hours which jeopardized the launch a little bit – here is the breakdown from my buddy @signorettif on Twitter if you want to read more.

            Overall we got over 100 upvotes and we stayed on the frontpage for most of the launch day, only to be kicked out of it on the next day.

            8/10 would do it again 😃

    2. 1

      The only thing I did was A/B test a few items on the website. No clear and obvious reasons there, though, that would have caused this jump. I think it just takes a few weeks for search engines to pick up on anything at all, and then, of course, word-of-mouth marketing generates a snowball effect, but takes some time to get going.

      1. 2

        Super useful insights! I've been on the fence whether investing in content marketing would be worth it for us at this stage. I have been getting more and more signals that maybe I should.

        1. 1

          It's the only thing that will always work, IMO! Might not generate overnight success, but there's nothing I'm happier about having done five years ago than spend days and days writing a bunch of high-quality blog posts about my new company (a different brand). Those pieces yielded more than $1M revenue over the following five years, but yielded nothing for the first six months.

  9. 2

    Interesting! What was happening to your SEO etc during that time, were you seeing traction in click-throughs etc?

Trending on Indie Hackers
Let's discuss databases? :D I would like to know about your database and your journey in using it 36 comments I'm 14 year old. I built my side project in 4 hrs and sold it for $2K within 4 days of Launch. Here is my Journey 🎉 23 comments List of places to submit your startup (for free!) 18 comments I'm 20 years old and launched an app that went #1 on the App Store. AMA. 17 comments I'm launching a new app and looking for feedback on the sales page 7 comments 💔 Y-Combinator rejection to new SAAS launch 🚀 5 comments