We've bootstrapped to $1.6M ARR in a crowded market. AMA.

Hey all,

I'm Tom from EmailOctopus, an email marketing platform which along with Jonathan we have grown to $1.6m in ARR, which puts us at $3.6m in total revenue.

The business has been going now for around 7 years, with Jonathan and I both full-time for 4 of those. We're now a remote team of 9.

We've built this product in a crowded space, without a pre-existing audience and without putting any of our own money in besides the initial domain name purchase and the first logo which was done on Fiverr.

Here's a very quick timeline of what's happened over the past few years to give some context:

November 2014: Go live as a completely free product.
December 2014 Feb 2015: Launch post made on Reddit.
Jul 2015: Added paid plans alongside the free plan
Mar 2017: Both went full-time on the project ($13k MRR)
Jan 2018: Featured on the IH Podcast
November 2019: Reached a cumulative £1m ($1.3m) in revenue.
Jul 2020: 10bn emails sent through the platform.
May 2021: Finally launched on Product Hunt


And that's us. We're here to answer any questions you might have about EmailOctopus, growing the business, or more broad email questions.

  1. 8

    This is amazing! Didn't know EmailOctopus was an IH project.

    What have been the best growth channels? Did the initial reddit post drive any traffic to the project?

    Best of luck in the future!

    1. 14

      Thank you! I think we may have launched before Indie Hackers, but have been on the website for a long-time now and are semi-regular attendees at the London/Vancouver meetups.

      I've covered this briefly below, but the initial Reddit posts did introduce short term spikes. The graph here shows the short-term spike from it, where we went from an ave of ~10 sign-ups per day to our most successful launch day where we saw 25.

      My favourite growth channels have always been the free ones :) In the early days we did a lot on Quora. They used to have something called 'Answer Wikis' where each question had a Wiki below it, which anyone could edit without needing to answer the question and reach the top through the algorithm. We saw a lot of success in the early days by adding EmailOctopus to those answer wikis. Another source of good quality sign-ups was through our free templates promotion, where we open-sourced our email templates and allowed anyone to download them either direct through Github or via our website with an email address. With those email addresses we then sent a drip campaign encouraging them to try out our core product – if they needed templates, presumably they already had an email marketing program, and may would need an email marketing platform either then or in the future.

      But in terms of the most consistent, and best in numbers of new customers, the best growth channels have been our paid efforts using Google PPC particularly on longer tail competitor keywords in emerging territories. Although it's incredibly challenging to do it profitably and requires a lot of close management from Hollie, our marketing manager. For context though, we didn't do any paid advertising until we were at $22,000 MRR in October 2017.

      1. 5

        What an amazingly thorough answer, I really appreciate it.

  2. 3

    Hi Tom,

    Congrats on the success so far.

    Have a couple questions:

    • What gave you the confidence to launch in such a crowded market?
    • What are the biggest segment of your users?
    • How much do you attribute your success to technical/product brilliance, savvy marketing, staying the course, massive market ?
    • What next for Email Octopus? Do you want to eventually sell/get acquired or run this for as long as possible?
    1. 2

      So there wasn’t confidence needed, per se. EmailOctopus wasn’t launched with the expectation of it being a big business employing people, it was just a side project to bring in some income for Jonathan — there was confidence it would make some money but not to extent it has; over time it snowballed into something bigger.

      Our biggest segment are people with large audiences who don’t make a lot of attributable revenue from email, like ecommerce would do. So it’s authors, musicians, nightclubs, events and those kind of creators. We also have a decent number of newsletters, some of them IH’ers like Zero to Marketing and No CS Degree.

      Much of our success has been a sharp focus on the right things, I think. We aren’t the best developers, nor am I a fantastic marketer — but we’ve done enough and have done enough consistently. Knowing when a feature is good enough and what matters to our users and prospective users. That’s been learned through work at previous companies and reading books like The Lean Startup.

      In terms of plans, the beauty of SaaS is the continued growth of the business so we’re not in any rush to be acquired or to head off and do something else. Having your own business which you can decide on your ethics, responsibilities and strategy is great fun — that said, we would be stupid to rule out an acquisition for a life changing sum.

      1. 2

        We aren’t the best developers, nor am I a fantastic marketer

        I love this and people should take heed; when you have a great product you don't need to be the best engineers using the latest stack, or the greatest marketer in the world.

  3. 3

    First of all, congrats! I'm curious as to what you see as your inflection point, where you saw real traction? Did you get a ton of engagement with your Reddit post? Did you see big uptick after the IH podcast?

    1. 2

      There hasn't been a single inflection point. One issue without having an audience is that you don't have a massive group of people who you can launch to and get that AHA moment. Instead we had to rely on other 3rd party audiences, where we quickly disappear from the spotlight so it's a case of consistently showing up.

      If we look at launches such as a Reddit launch have had short term spikes in user sign-ups. These we can confidently attribute to the launches though once launched it's not been a step change in sign-ups. The impact of the launches simply gets lost in the noise of everything else going on and long-term it's just impossible to attribute. But at the same time, we're confident that they have helped.

      Let's get some real data though and take a look at our most successful Reddit post, which is this one. This was just before we had paid plans, it received 70+ upvotes and was top of this subreddit for a day. In terms of sign-ups there's a clear spike perhaps a gentle daily increase in sign-ups but at the numbers we were talking back it hard to say it's a significant impact as a result of the launch.

      When we zoom out and look at user signups per day, over time. There have been a few notable spikes. November 2016 was the launch of our free open source email templates. October 2017 we started doing paid advertising. November 2017, Black Friday promotions.

      1. 1

        Thank you, and that Reddit post is a thing of beauty. Great pithy value prop!

  4. 2

    Congrats, that is an amazing journey. I have a couple of questions.

    1. Why did you wait that long to launch on Product Hunt?
    2. What was the best acquisition channel for you at the beginning?
    1. 2
      1. We didn't think Product Hunt would be the right audience for our core product, as it's not particularly anything innovative. So we didn't prioritise it. We did instead launch our side-projects, like our free pack of email templates – as we saw a trend of useful free products getting traction there.

      2. Best acquisition channel at the beginning, both in terms of quantity and reliability was Quora via their Answer Wikis.

  5. 2

    just wanted to say I'm a happy [paid] customer! RanchWork.com and VidaliaOnions.com use EO.

    1. 2

      So cool! Thank you – enjoy a cheaper bill (30%) for the next 3 months ;)

  6. 2

    Thanks for the AMA - I remember Justin Jackson saying all the biggest IH projects are generally on open protocols like email.

    How did you fund it in the early days?

    How easy or hard was it technically to create?

    1. 3

      So we didn’t do paid advertising until we were at $22k MRR so our only costs really were hosting and ourselves. The latter is very important, getting paid is part of the fun of working on a business!

      So we paid the hosting, this scaled as customers grew and took around 10% of revenue. That left 90% of income to pay ourselves, so in the very early days it was worked on during evenings only. As we reached about $3k in MRR we moved to work 2 days a week on EmailOctopus, with the remaining 3 spent contracting. Jonathan managed to negotiate his present job to allow him to do that arrangement, and I found a consulting role.

      Once we reached ~$13k MRR we both went full-time with the business providing enough money to cover our expenses (mortgage/rent) and a reasonable standard of life, although it was a pay cut for us both.

      1. 1

        nice one! I note you didn't go all in on day 1 which is smart.

        Seems like you had a similar journey to Veed where you gradually did less contracting.

        1. 1

          Yep, not too dissimilar. We have been pretty risk averse, which I'm sure has stunted growth to some extent – but at the same time it's allowed us to really enjoy running it and do the other things outside of EmailOctopus which we love.

          I did miss your 'How easy or hard was it technically to create?' question. The initial product was relatively simple. You uploaded a list of email addresses and then begun sending to them. At the low-volumes we saw initially this wasn't too difficult to do.

          Our main technical challenges have come from scaling and handling the sheer number of events that an email can generate. Our largest customers have upwards of 3m contacts and emails generate a lot of activity in their first few minutes of sending. Being able to send those out at a fast rate, then being able to receive, store, process and then query at a later date all that data is a real technical challenge.

          Ben, our Senior Dev, looked at our email event storage bucket today. It's now over 21 terabytes. Much of that needs to be queryable for segmentation/automation.

          1. 1

            wow, 3 m contacts is a lot!

            Nice one - yeah, I'm just not a fan of the whole ''you gotta quit your job even when there's no traction" argument in startups so it's good to see a more realistic path for bootstrappers

  7. 2

    Wow! Congrats Evans. What have been some early signs and learnings that you think somebody who is just starting to create a project should know?

    1. 5

      So the early signs are always people using your product. People using and paying for the product gives you the energy and momentum needed. So how do you get there?

      • Launch early. As soon as your product solves one persons problem, launch it.
      • Deliver quality. While you should launch early, that means launching without all the additional features and complexity not compromising on quality. Users will return to a nice product that didn’t quite do what they wanted, but they won’t return to something which has caused them to lose time or work.
      • Once the initial product is built, focus on marketing. It’s easy to revert to building features as the outcomes are predictable (writing code = a feature), however until you have customers and users you won’t know if you’re building the right thing.
      • It takes time. All the numbers you see here on Indie Hackers are after years of work, either on their product or on picking up the skills to build their overnight success. The beauty of SaaS though is that it multiplies and after 2 years of seemingly slow growth, you’ll have a strong business
      • Be honest about your size and who you are. Nobody has ever been upset to receive customer support from a founder. Don’t try to fool people into thinking you’re bigger than you are, everyone (besides Enterprise) loves an underdog. Embrace it!
  8. 1

    Saw your advise in the newsletter. Loved it. Succint and easy-to-follow.

  9. 1

    Awesome AMA, Tom! Including in the next issue of the Indie Hackers Newsletter. Thanks for sharing!

  10. 1

    did you use Product Hunt to launch? & do you create tik tok content for the company as well as the other standard socials?

    1. 2

      No, we didn’t launch on Product Hunt for 6 years. It just wasn’t a priority at the time and I’m not sure our product is right for that audience

      We don’t use Tik Tok yet, although video is clearly something I’d like us to explore more.

      1. 1

        Totally get that & tik tok is a great place for organic reach, def worth the time! Could e-intro you to a community manager who's killing it on tik tok for some more insight if you'd be interested. - shoot me an email!

  11. 1

    You've been already 7(!!) years on the market. And it's amazing. What is not amazing, after 7 years, you still don't (!!) have a feature to send segmented emails when users sign up.

    1. 1

      Fair point, we definitely don't have all the features which others in the industry do and we would also never claim to!

      You are also correct is that one of those is the segmentation on automated emails. I appreciate you're really keen for it. We are a bootstrapped company, with a focus on growing sustainably, so it means we do have a smallish development resource. That means we have to ruthlessly prioritise things.

      When we launched EmailOctopus is simply allowed you to send an email to a list. The list didn't have first name, or any fields attached. That was our MVP which delivered value to quite a significant portion of users there is a huge segment of users don't need automation, personalisation or segmentation. They simply want to send an email out to their audience to announce their new book, or their next tour.

      To build on that MVP we have identified areas where we can deliver value to users beyond that and we prioritise these by speaking to our users and prospective users. Simple welcome emails were first in 2017. Followed by custom fields and drag and drop editor which we launched in 2018. It wasn't until 2020 that we added segmentation to the core product. Later this year (or maybe early next) advanced journey style automation with segmentation will come.

      Ultimately, we take time over our features ensuring they're reliable, easy-to-use and well polished. I also think it's important we don't overwork our team and expect that they start/finish at the contracted time they're supposed to. These things, combined with the scale at which we're working (21tb worth of data!) does mean our approach can seem slow and cautious.

      There are plenty of email marketing platforms our there though, and if our approach doesn't suit that's absolutely fine.

      1. 1

        So, you are trying to say this is feature is very unimportant to your users? This is simply not true. It's very important. Sorry, I don't know how to prove it but it is. Having 9 people in your team, it's definitely possible to implement it. It's a pity that all you have to say is "if you don't like us, go away" instead of listening to your users :((((

        1. 1

          No. I think you may be putting words in my mouth there.

          I am explaining there were a number of steps we needed to go through before getting to this stage, delivering the most amount of value, to the most amount of users across all those steps while delivering high quality features.

          I am certainly not doubting it's important. If it wasn't important we wouldn't work on it, I have also said it's on it's way.

          1. 1

            I am certainly not doubting it's important. If it wasn't important we wouldn't work on it, I have also said it's on it's way.

            Great, thanks! Looking forward to trying it out.

  12. 1

    Congrats Evans! So how did you guys manage building & talking to users in the first 2 years while you were working part time?

    1. 2

      We would build on the days we weren’t working on our other roles.

      The talking to users was just done in an asynchronous format - we’d talk to those using us over email, do Typeform style surveys and then massively research our competitors and pain points people were having with them by reviewing their online profiles on places like Capterra.

      We clearly couldn’t offer 24/7 support but we cleared down our emails in the mornings/evenings we weren’t working which meant that even for urgent things there was really only ever a 4hr reply delay. We were small at the time too, so only getting a handful of emails per day - very manageable.

  13. 1

    Whoa, dude I think I saw your company years ago!

  14. 1

    Thanks for the AMA.

    1. What measures are taken to prevent spammers from signing up for free and paid (via fraudulent transactions etc.) accounts?

    2. For spamming activity that does arise, how do you minimize and resolve the repeated IP blacklisting and other penalties that inevitably occur as a result?

    Thanks in advance!

    1. 1
      1. This is the biggest challenge of any email marketing platform. We verify every single user who signs up to the platform with a combination of some automated checks and some manual ones. Our acceptance rate here is below 50%. Once an account has been verified their metrics and content are continually monitored with regular a dedicated #compliance Slack channel in which concerning metrics or content is flagged up. Our No.1 concern is phishing, Our no.2 is spam.

      2. Spam will always creep through, even very legitimate businesses send spam. So this is a case of actively monitoring emails being sent from the platform, as well as our IP addresses/domains against real-time blacklists. We have a number of IPs which we auto-assign users to based on their sender reputation, this ensures good senders always see good deliverability – and those who send a spam campaign by accident (for example to an old list) have an opportunity to rebuild their reputation. We of course also terminate a number of accounts who clearly breach our ToS.

      While it is an 'ask me anything' I can't sadly be too publicly candid about the manual/automated checks in place, nor the metrics which really matter; it may encourage people to circumvent them.

      What I can say is a legitimate user who sends emails to people who have explicitly opted in (therefore seeing low bounce/complaint rates) will see excellent deliverability.

  15. 1

    Hi Tom, I've been using EmailOctopus with Songbox for a while and I love it. Best decision I made was leaving Mailchimp for you guys.

    Here's my question. At $13MRR what salary did you take when you first went full time in the business, and what were you and your partner's main function in the business then?

    1. 1

      At $13k MRR, that was about £10.5k per month in revenue back then as the pound had crashed post Brexit.

      Around £2k of that went towards hosting costs. So our take home from memory was £30k per year each. All of that was PAYE as we were heavily involved in R&D so we could claim much of the tax/pension/salary through the R&D tax credit scheme. We wanted to hire an additional developer, plus also used a contract developer so any additional profit was used there.

      Our role split back then was quite clear. Jonathan was solely on development, and I picked up the customer support, marketing, forecasting and speaking to customers. As we've grown we've hired a team – we've got 2 developers, 1 marketing manager and 4 people in customer success. So mine and Jonathan's role is probably 50% strategy (both product and business) with 50% management/hiring/doing bits in our respective areas.

  16. 1

    That's amazing. I have 1 technical question - what are you using for sending emails? Are you using solutions like SendGrid or you built your own SMTP server?

    1. 1

      We are set-up to be able to send through a number of third party services. Users can either use our own infrastructure held with those services, or connect their Amazon SES account.

      At present we are using our own dedicated infrastructure run by Sparkpost. We send around 1bn emails per year through our servers there, and around 2bn per year through our customers own Amazon SES accounts.

      That set-up is quite common in the email marketing world, for example Convertkit and Substack use Sendgrid infrastructure. Intercom and Hubspot use Sparkpost. You can see which service each ESP uses by looking at their IP's and the location, some use multiple providers.

      1. 1

        Why Sparkpost dosn't show the pricing ...?

        1. 1

          https://www.sparkpost.com/pricing/ this should be it...I guess @td_evans managed to negotiate pricing different from what is shown?

          1. 1

            thank you very much , can you explain why you didnt use aws SNS
            as your main SMTP provider ?

          2. 1

            We send way in excess of 100m emails per month, so there is room to negotiate. Our rates are significantly below SES pricing.

        2. 1

          It’s an Enterprise platform usually, the prices are very negotiable at the volumes we send.

      2. 1

        Now this is a solution...and how do you balance between profits and costs, as you're giving away a free option...meaning, a lot of users can be eligible for the free plan and will continue using it for a long time, this means you're going to pay a lot of money to the third party services

        1. 1

          Our free plan is limited to 10,000 emails per month. Which based on published Amazon SES prices (we pay significantly less than this) a free plan costs $1 to run in terms of external services, if they max-out their usage.

          So that's the max cost per free account, but the majority of active free plans do not exhaust their usage quota. Our free plans, for example, cost us $1,400 last month to run. This is essentially a marketing cost which we seek to recover through upgrades and virality.

          I personally wouldn't want to launch a SaaS product without some way to try it out without paying. It's too much of a barrier.

  17. 1

    I love the EmailOctopus offering and always been a fan - I tried to become a customer (as a freelancer doing projects for others) many times but everyone kept forcing me (asking me nicely but insistently, to be honest) to use MailChimp - the SNS API key thing never got digested for anyone I talked to...can't wait to find that one person that will finally let me make the jump

    1. 1

      Thanks for trying though!

      Some good news, we no longer need an API Key or that technical set up. You can begin sending straight away :)

  18. 1

    Hey Thomas congratulations! Happy EO customer here 🙂

    I have a broad question. Are you seeing any impact at EO since the iOS 15 release? They preload images so I expect open rate to slowly go up as more people install it.

    1. 2

      Aww thanks so happy to have you as a customer!

      Yes, we are. But not as significant as you may think yet. iOS15 uptake has been slower than previous versions and it only preloads images on WiFi when plugged into a charger, so you can some spikes overnight.

      It’s one to monitor though and over time open rate will become less useful as a metric. We are fortunate that we don’t do too much with that data, besides open rate reporting - I know some platforms use it for AB Testing, Geo Location and AI-based send time optimisation. Sometimes it helps being a simpler product!

      I expect we will look to add some kind of engagement block to emails, to encourage clicks.

  19. 1

    Amazing !
    Do you still asking for aws SNS registration ?
    Or did you replaced the SMTP providers ?


    1. 1

      We stopped asking for this around 18 months ago. It was a huge blocker to new users getting activated (I.e sending their first email) and didn’t match with the easy-to-use nature of the rest of the product.

      Our core proposition remains the same though and we are incredibly focussed on being affordable.

  20. 1

    Congrats on your success! I briefly used Email Octopus for a (now-dead) startup in 2017-18 and found it very nice, though still a bit rough around the edges. I'm sure it's a much better and more mature product now.

    I've seen you guys do a lot of advertising on newsletters and founder-centric websites. Has that been the best paid acquisition channel of yours? What other paid channels do you use in this crowded market?

    1. 3

      Ahh Hollie, our marketing manager will be delighted you've spotted us in those newsletters!

      So our marketing spend in 2020 looked like this:

      Things that worked here:

      • Google ads, notably brand and longer-tail competitor campaigns. We haven't had much luck with generic email marketing phrases on Google.
      • Sponsorships. In 2020 we did some awesome sponsorships with the likes of @harrydry and Dense Discovery. But we also did some sponsorships which didn't work. Our best sponsorships seem to almost be partnerships, with people we know or who use the product and where their endorsement is very much genuine. This is something we're exploring more in 2021 and we've just agreed to sponsor @petecodes 's upcoming course. For any IH'ers reading we are also on the lookout for other similar partnership style sponsorships.

      Things that didn't work:

      • YouTube and Twitter I think have provided us a grand total of 0 paid sign-ups (though Hollie may correct me).
      • Facebook. This has always excited me and seems to be the generic advice, however we've never managed to make it profitable. We do sporadically use it for awareness type campaigns.

      We also spend money on Capterra/GetApp – these used to be great but they have recently changed their minimum bid which may make them unprofitable, remains to be seen.

      In respects of the product, thanks for giving us a go! I think you're right that we're a bit more mature now, and do take an iterative approach to the development.

      1. 1

        FWIW, seeing people like Harry and Andrea from Zero to Marketing were big influences for me. Social proof is very effective!

  21. 0

    Hey Tom, thanks again for sponsoring my course!

    I've just finished recording this second actually. I teach everything about growing an email list with free tools and anyone who buys the course gets 30% off paid plans for 3 months :)


  22. -2

    This comment has been voted down. Click to show.

    1. 4

      Motivation is an interesting one. I know there’s a lot mentioned around revenue levels, and in the grand scheme of things we aren’t particularly successful in revenue terms. But at the same time we have enough revenue to treat all our employees well and that it would be a life changing sum if we sold. At that stage of the business motivation can be harder to find, particularly when being Uber rich isn’t of interest.

      I do admire the entrepreneurs that continue to push, like Zuckerberg and at a smaller level in our industry Nathan Barry. I’m not sure that’s either Jonathan or I!

Trending on Indie Hackers
Share your product or landing page, and I'll give you product design advice 104 comments How do I transition from a wantrepreneur to an entrepreneur? 35 comments Copywriting Examples — The world's best copy. In one place. 13 comments Building a microsaas in public 12 comments App Stores are powerful search engines 12 comments Working towards an MVP 9 comments