Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: How did you grow from 100 to 1,000 users?
209 points by sherm8n on April 29, 2017 | hide | past | favorite | 65 comments

I looked at the other Ask HN for 100 users.

Only 1 person mentioned influencers. I'm going to recommend it again, but with more detail.

An influencer has a particular niche/segment in the market that they are passionate about. They may have various channels in which to communicate to their audience, these may be facebook page/groups, youtube channels, instagram.

These influencers have from 100 to 100k audience members. There's nothing you need to do, apart from approach them and to find whether it's a good fit to pitch your product. However, be aware that the influencer will be looking for a % of sales.

What is great about an influencer, is that in some cases they did all the hard work for you. They already created a product that the audience members consumed and are generally happy with it. So if you have an addon product, then you are simply preaching to the choir at this point.

Where to find influencers? Youtube, Facebook Groups, Instagram. [1] Is a place to go, if you want to be laser targeted and have stats.

[1] http://www.hyprbrands.com


Now for anecdotal experience. I'm in the affiliate marketing arena. I know many many different influencers. They either have products developed for them or they cross promote each others products.

These guys regularly generate from around $10k to $100k a month. I've met few that have done $1m a month in sales.

Done correctly, you can position yourself to flip a switch and have over 1m unique users read or hear about your product.

I know there is a segment on HN that generally frowns upon affiliate marketing, but many companies such as Netflix, Groupon, AirBNB, Apple, Amazon used them to great effect to grow exponentially.

How do you track and pay commissions? Any tool recommendations?

Have you checked out GrowSumo.com? They were YC S15. Luke and Bryn are awesome guys. They got us connected with a handful of influencers that promoted my first startup – www.goodaudience.com. Then their software tracked monthly commissions and payouts.

You need to mention that they charge $300/month minimum. Their feature set seems pretty low for the high price...

Thanks sherm8n. Happy to chat to anyone about best practices, but our platforms lets you set commission and make payouts.

back in the days when we were making ios games we would send out a lot of promo codes to various youtube reviewers with followers between 100-10k. it was getting popular back then so we didnt paid for nor offered money for the review although few replied back and asked for it. well more like a paid service than affiliate. a few days later we experiences a spike that originated from a reviewer we never heard back.

How to get in touch with you ? I would be very interested in learning more about your approach.

I work on the tech side of imfluencer marketing and am available to answer most questions through email. No bs or strings.

Keep in mind that no product or budget is too small for influencer marketing!

How could I go about getting in that position to flip a switch and get 1m or 100K unique?

what industries or niches does this strategy work for? which ones do they not work as well?

If your product is related to SEO, social media, ppc this can work really well.

We recently launched a side project in SEO vertical with couple other partners and did over $200k in sales revenue over a week through our affiliate channel.

If you are looking particularly where affiliates are promoting for products then check out voluum, link assistant etc.

If your product is more of a tech side and starts with less than $20-$30 then chances are less people are going to promote that.

Thanks for this! Would you mind diving a bit deeper on how you were able to drive $200K in sales? That is impressive and congrats!

Thank you, most of the sales were generated through affiliates, our profits after the launch were about 95k. If you check out jvnotifypro or jvzoo.com, you will get an idea.

To drive $200k was not simple, we partnered up with a person who had a record of 6 figure launches and had a good relationship with JVs.

To be honest, most of the tech savvy people would be hating jvzoo type audience but the reality is if you have good following you can generate good revenue from your side projects and feed that revenue into your main Saas startups for bootstrapping. This is something that we are doing without involving any VC for our core product.

One another example is a shopping cart product is Samcart.com, the co-founder Brian also captured the top Jvs for his launch including Jeff Walker, Todd Gross etc. These Jvs I have mentioned actually have the similar type of audience like Jvzoo.

Also, on the Saas side, one of my buddy Thomas from pagemodo (previous co-founder) did a similar thing while running the startup. And later he sold his venture for $114 million.

And the Voluum founder Robert gryn used the same approach to scale his product.

"If your product is more of a tech side and starts with less than $20-$30 then chances are less people are going to promote that." Why is that ? I didn't really get it.

Not OP, but I think it's because devs are slightly harder to sell to compared to non-devs. Also, the commission for a product less than $20 will be too low for anyone to put in any effort to promoting it.

5% of $20 is $1, meaning you won't be able to cover operation costs even if you sell 100 of it every month. If the product/course/service is $2997, and your commission is 5%, you are looking at $10k+ net profit every month if you sell 100 of that.

In addition, because the author/creator/founder of the $20 doesn't have a high profit margin to begin with, the commission rate will be very low compared to premium products/services. E.g., 5% for $20 product vs. 20-30% for $2997 product.

Influencer and affiliate marketing channel works in markets where there are many potential influencers... that's it.

For advertising and marketing related tools(SEO,PPC,social media, marketing automation,etc) works better because their main clients are marketers, many have an audience and know how to market a product.

However, it also works for sales tools and CRMs for ex., there are lots of influencers in this space.

Works amazingly well for e-commerce, which is where influencer marketing is most used, everyone with a good following on Instagram can become an influencer.

If your space doesn't have too many influencers, start a referral program.

I did that with the cPanel Varnish plugin [1]. It's Varnish Cache integration for cPanel WHM (duh!). I signed up on several web hosting forums and would volunteer to help when someone posted looking for help with traffic or website performance.

But I'd have to login to each box, spend a good amount of time on the server, understand the business requirements, and then deploy Varnish. It was hugely time consuming but I loved doing this. It was less about bean-counting but rather more about helping. Also progress seemed slow at first because I would gain one user every 2 or 3 days. But word-of-mouth buzz from people I've helped was the thing that moved the chains for me.

The cPanel WHM platform has been moving quite fast over last few years. And so did several things under the hood. I also wanted to build new features that just couldn't be done on cPanel. I felt the project had potential to expand and reach a much larger subset. And that's when I built Cachoid [2].

[1] https://www.unixy.net/varnish

[2] https://www.cachoid.com/

yes, i recall your plugin several years back. think over 7 years ago! although i had a hard time with dynamic content especially with sessions and logged in vs logged out views. but the performance boost from varnish is pretty mind blowing when configured correctly.

It's crazy how time flies! Caching dynamic content is definitely tricky. Subsequent releases of the plugin have had WordPress specific VCLs among others. Cachoid has WordPress specific VCL too with a WP plugin so post edits/changes are reflected with immediate purges and cache warm ups via the Cachoid API. But the benefits of caching pages far outweigh the drawbacks. At least for more sites.

wow that's pure hustle! you logged int about 900 boxes then? :)

Quite a bit more. This is from my jump box:

$ wc -l ~/.ssh/known_hosts 1240 /root/.ssh/known_hosts

I did it with Pavlok, which is a wearable device for waking up early and changing habits. The first 100 was done through writing blog posts, getting email subscribers.

When we were ready with a basic prototype, we did a webinar presentation to our email list to pre-sell it (6 months before our indiegogo). We ended up selling about 200-300 prototypes before the webinar, and 300-400 pre-orders.

When we launched our Indiegogo, we built up a ton of press, media, bloggers, email and social influencers, etc to announce our campaign all at the same time to make our indiegogo sell ~3000 units pretty quickly.

Obviously it's a little different than software, but I find the launch methodology we used works in software launches too.

[1] http://pavlok.com [2] https://www.indiegogo.com/projects/pavlok-breaks-bad-habits [3] https://medium.com/@maneeshsethi/kickstarter-is-sorta-debt-a... -- this is an article that talks about how we built a hardware physical product without a real VC round.

Maneesh, I found your journey truly inspiring in the way you automated a slap in the face. I had actually heard about the product before seeing the Shark Tank episode, and was aghast by their rudeness! You're clearly a smart guy and I think you made the right decision at the end.

Would you mind commenting on the value that TV exposure on Shark Tank brought to Pavlok?

I posted this on a similar thread a few days ago. For Standard Notes [0], here's what I did:

1. Comment on privacy related HN posts about a privacy-focused notes app. That would have gotten me 40-50 users.

2. Write articles [1] on encryption/privacy/webdev. Some of them made it to frontpage HN, some didn't. That might have gotten me to 500 users.

3. Repeat. Tirelessly. Painstakingly. Depressingly. Just keep going doing small things every day. Eventually they start to compound.

[0]: https://standardnotes.org

[1]: https://journal.standardnotes.org

Similar thread: https://news.ycombinator.com/item?id=14191161

Seems like the answer is always to grind it out. I commend anyone who has put in the work to succeed.

We operate in the public sector SaaS space. We tried all sorts of focused marketing, but by far the best results were obtained by 1) Picking up a list of municipalities 2) Calling the main switchboard of each one and asking who makes decisions relevant to our business 3) Calling said person, setting up a live demo.

There's still immense power in the human-human interaction, even if it's not webscale.

What's your average customer LTV?

High thousands to low tens of thousands per year, lifetime is hard to guestimate as there's very little churn in our business. We still retain practicly all the customers we've ever made.

Thanks for sharing. I figured that kind of revenue justifies the cold calling. Congrats on finding the niche!

Would you mind sharing a link to your product?

These are the top 3 things that worked for us at https://emailoctopus.com:

1) Posted on niche subreddits and HN. Commented and gave a gentle plug on related posts. 2) Launched a free side project on Product Hunt: https://www.producthunt.com/posts/templates-by-emailoctopus 3) Ran a crazy-cheap Black Friday promotion for all our leads.

I've seen my current product go from 0 users to thousands. It is a business workflow tool where most users log several hours per day, so I understand that it's very unusual compared to other services.

The big thing for us was just a strong sales and marketing presence backed by a product team (my team) that listens for the user and the marketplace and responds quickly. It takes a little time but snowballed by just sticking to that formula. Sales is in the DNA of my work, and so they are very effective at all aspects of it.

Sounds like weak advice, but it worked for us.

Sure, but sales is expensive. Is there a trick or insight to keep sales costs low ?

It almost baits me to pitch the company I work for. I would think we keep sales cheaper because of efficiency.

We were very very good at setting up a very efficient pipeline of demo setter (sdr) to demo performer and closer (ae). I believe this is one of the more cost effective ways to scale because the specialization allows for getting really good at that one part of the pipeline.

Agreed (coming from a similar spot it sounds like, in my case daily operations tools for SMBs): build a good product, and a sales team to sell it. Then sell it, all day, every day. Just make sure the product can back up sales!

Keeping churn in check is very important. High churn means something is wrong with the product.... Or maybe not? Implementation and support are important parts of the equation too. Product can't be garbage though.

You get your first 100 users already so I think you went through the hardest part. Unless it is all friends, you can analyze those first few users and see common characteristics such as age, gender, location, interests, etc. These first users are your micro-micro universe and you want to get to know them. Once you found those common traits, you can go out and look for those very similar potential customers and reach out to them. How? It depends on your product and service. You can try paid ads via Facebook/Twitter/Instagram/etc, you can try influencers to feature your product, reach out to bloggers in your field, talk to people face to face, etc. etc. You have to find what works for you and your app. Make sure that you analyze everything you try. Some things might not work out totally as you expected, but might give you some good lessons how to try next time. I suggest you check out this list: http://blog.linkody.com/seo/growth-hacks-list

Posted this in the 0-100 thread, but posting here as well:

I did the following for http://www.oppsdaily.com

1. Posted landing page in a slack chat (first ~10 users)

2. Posted on the indiehackers forum (+30 users)

3. Got mentioned in the indiehackers weekly newsletter (+150)

4. Posted to hacker news (+?)

5. Posted to product hunt (did not go so great)(+?)

6. Started posting weekly metrics to HN - made the frontpage (+1500-2000)

Hey cool idea, I've subscribed.

One feedback - when you press Subscribe it opens a new tab to take you through the Mailchimp flow. Considering there's not much value on the landing page, I think it's better to stay in that tab.

The first few dosen users were from an iosdevforum. Then someone recommended a nice feature - at first I wanted to tell them it's impossible. But as I was writing an explanation, I realised it's actually possible :)

Drank a lot of red bull for a week and pushed the feature to production. Then I emailed a reporter who covered our competition in TC weeks before, and put a post on HN.

It picked up steam on HN, and the TC article appeared a day later:


Notice the crappy logo and make sure to watch the video - recorded in my living room after a few sleepless nights.

While working on the improvements, I invented another feature (guessing competitors keywords), which is now a standard across the ASO apps.

A few months later, users began telling us AppStore search algorithm changed. After a few weeks of crunching I published a slideshare explaining the recent changes. Another reporter picked it up:


They embeded our slides, which pushed them into top 3 slideshares of the week, or something like that. (after a few years the viewcount is 500k or so)

Users began coming in en masse (up until 800 paid or so), and people began sharing slides and so on.

I gave a few podcast interviews, which pushed the promotion even further. But it was mostly TC and HN that pushed it so far.

If people really like your app AND you have a way for them to invite others easily AND share about it on their social links easily AND you started off by going to your core group personally and battle tested it, it just happens.

Total nonsense. Most apps fail to gain enough users to experience the long tail effect of which you speak. http://2fb.me was an app that connected Twitter and Facebook through sharing. I built it believing this methodology to be true but it's not.

Hmm your app looks cool. Maybe you need to connect with your users better, maybe nobody knows about it? Maybe the message isn't getting across.

this is true only if you are in an ecosystem where people commonly share apps such as tech, food, marketing, and productivity and your app has a very useful freemium component.


Free or paying? Anyway, I think that is the size where you have to explore social media. Typically Twitter is easier, Facebook will work is your biz intrinsically spreads across friends, or you can explore others such as Instagram and Pinterest if you'd like.

It's also probably a good time to start experimenting with ads, though don't be fooled by the results and stay focused on product-market fit (I assume you're following the startup school classes).

For us, Theneeds, the 100-1000 was totally through Twitter. It was also the time where we started exploring apps outside purely web, first Pokki and later Chrome (but Chrome was 1k-10k).

You mean Twitter ads? I tried several times, was very underwhelmed. Curious what made it work for you.

No, no ads. Twitter meaning start a meaningful account, interact with users, new & old. It's a long term investment, like SEO, but it also pays back relatively soon in terms of signups.

Twitter recently has been shadowbanning new accounts which follow the particular pattern. Especially if most posts contain URLs. Might no longer be a good idea. If you post with popular hashtags and get no response probably Twitter shadowbanned you.

Couldn't you verify from a test account?

Remember that there's not just one answer. While other answers may work in different degrees, you should find your marketing mix.

While I agree with some strategies (particularly inbound marketing), I think that viral marketing has very high potential. Think about that: you want to grow from 100 to 1,000 users. How many hours did you spent on each one of those initial users? Now multiply it by 100. Seems scary, doesn't it?

Better harness the userbase you have, and let them spread your product. If each one of your users can bring one more user, you've effectively doubled your userbase, and then again, and by the third time you are 200 users away from your goal.

This seems so easy and magical on paper, but you have to design your product's virality: thinking about how it's easier for your existing users to bring more of them, and rewarding both, either with cash or with influence.

A lot of business have been made with this strategy in mind. Just look for the story of Hotmail. If you want to read more, check out Ideavirus by Seth Godin, and Viral Loop by Adam Penenberg.

For our freemium service , Product hunt gave the biggest boost in free users. Real paying users we get with Google Adwords


I grew my thing, a developer's productivity tool that runs in the browser and command line, by focusing on feature enhancements and code quality. I am obsessed about quality of performance and quality of output. This has its pros and cons:


My little tool can do things other more popular tools cannot do, tends to be more stable, and executes extremely quickly. From purely a technology perspective my tool is probably superior in many ways to other similar tools. This has earned me a super loyal following and allows me freedoms to solve problems, provide enhancements, and solve bugs very quickly.


I am so focused on the code and technology that I don't perform marketing or advertising of almost any kind. I am entirely reliant upon organic search results, word of mouth from my users, and occasional feature announcements. The challenge with this is that I have no control and often no awareness of my traffic.

Being obsessed with code quality and product quality the code is very terse and not friendly to many developers. This means there is less interest from users to submit pull requests to the project.

The extremeness of this approach makes the popularity of my tool hard to gauge. Last year it appeared I was on trajectory to become one of the most popular packages on NPM. I pulled out of NPM because they kept breaking my package, and so I have completely lost that traffic. I was about to exceed a million NPM downloads a month almost entirely from direct sources external to NPM. This is what it means to be obsessed with product quality.

The single greatest burst of traffic to the website came when I published a new diff algorithm here last month. I made the front page for a day and got about 50,000 visitors.

What I have learned is:

* If you are producing something worthy of demand in a way that is superior to everything else it will eventually get the attention it deserves. If you are willing to accomplish those things the competition is either incapable or unwilling to accomplish you can compete no matter their funding or popularity.

* If the code is harder for newbs to instantly jump into your contribution community will be small or non-existent.

* If you damage the availability of your product most of your users will happily transfer their investment into something inferior. For most people convenience is, to a point, more important than niche features or product quality.

* Without any marketing strategy the popularity of your product will grow exponentially (if it is worthy of growth). So expect to invest years of effort before seeing any return on investment until that special tipping point occurs.

Don't be shy, what's your tool?

Your tool is great has saved me countless times.

I think the web ui is a little weak, but it is miles better than anything else out there.

How do you got your first 100 users?

We run a community platform called Snapzu (http://snapzu.com) which is similar in nature to HN/Reddit/etc (which entirely depends on user submissions, voting, etc) so we right off the bat had to battle the notorious "chicken and egg" problem. It was really hard attracting users when the entire site depends on other users (of which obviously at first we had very little of) for content.

However, we did several (mostly automated) things to increase our user count to 26,000+ members, which is where we stand right now. The main goal was to make sure that each user is catered to and getting exactly (or close to) what they are looking for.

1. Split up the content into several categories (14 to be exact). This allowed us to focus on specific categories such as Science, Earth, Politics, etc. Each category has its own social media account (on Twitter, Tumblr, Medium, WP, etc.) where we share our highest voted RELEVANT user-submitted content automatically using IFTTT. Example: http://science.snapzu.com

2. We also used the same user-submitted content to build our newsletter and send out the top relevant posts of the week for the "tribes" (communities like sub-reddits) that they are subscribed to. This attempts to get as many people to come back and often, a tactic extremely important when the community is dependent on other users coming back and contributing/participating.

3. We create automatically generated "top list" posts for some of our top categories and share them on all our social channels, newsletters, certain relevant sub-reddits, etc. These again are dependent entirely on user-submitted content, and although I mentioned they are automated, they still require some work (for the intro quote, minor curating) but 95% of the work is automated. Each post takes approx 10-15 minutes to create, instead of several hours, and uses the past week's top submissions as the main content. Example: http://snapzu.com/teamsnapzu/weekly-roundup-earth-and-nature...

4. Obviously there were also several things completely out of our control. Reddit had their massive debacle a couple years ago when Victoria got fired (she hosted their AMAs) which created a massive Reddit "revolt". Over the course of a few weeks in July 2015 we had 40,000+ people come in from there, of which approximately 8000 signed up. But that obviously would not have happened if we were not "somewhat known" at this point. We were basically at the right place at the right time, as one of the few semi-known Reddit alternatives/competitors. Because the "Reddit revolt" was somewhat big news, we also got mentioned in several articles (Daily Dot, Inc, Moz, etc) which brought in a few more thousand users.

We're now using many of the things we learned from the entire 4 year process of growing our own community platform to help bloggers (and/or website/business owners) who are struggling with the same things we were (and still are). Many of the growth problems (and eventual tactics) involved are nearly identical mainly because of the chicken and egg problem I mentioned before. If you are curious it's at http://blogenhancement.com. It contains several tools we use ourselves and basically allows bloggers to start, run, and utilize their own communities to get more audience, engagement, content, and revenue. It also ties in beautifully to our platform that we've spent 4 years building and constantly improving on, so it's a natural win/win for both parties.

Hope this helps. Cheers!

I'm your typical NoScript user from the days of web pages that were faster loading than the links we retrieved them on. I'd normally complain about one of these sites but your science link [re-]loaded seemingly in an instant, looks great, and scrolls well. So, thanks for a better experience than looking at many other demos. :)

I really like the name so I clicked the link. Somewhere on that you still have flash running? (Hans, get the Firewerfer!)

Why do you have to request an invite in order to sign up?

hn and product hunt

If you have a $5k or so budget and your product appeals to a more general audience, you will definitely get 1000 users by sponsoring the Epic Privacy Browser...they have a high visibility sponsorship that reaches a few hundred thousand users...so any general service will get at least a few thousand people to try a service (actual numbers will depend on whether your service is general or niche, etc.).

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact