• The Entrepreneur Forum | Startups | Entrepreneurship | Starting a Business | Motivation | Success
  1. Trading Life for a Weekend?
    Build a Business, Build a Life.
    Unleash the Dream.

    Become an UNSCRIPTED™Entrepreneur.

    UNSCRIPTED™ Entrepreneurship is an entrepreneurial business strategy that unleashes dreams, frees time, and explodes your incomes—but more importantly, it changes your life forever. Make no mistake; this isn’t something you “try”— this is something you live.

    FREE Download (Unscripted) FREE Download (Millionaire Fastlane)  Register
    Registering for the forum removes this block!

GOLD Ask Me Anything About SaaS ( I'm building my 7th )

Discussion in 'Business Models, Niches, Industries' started by eliquid, Oct 14, 2017.

FORUM SEARCH


Need answers about entrepreneurship? Search FIRST! More than likely your entrepreneurial topic has already been discussed.
Powered by Google Site Search.
Tags:
  1. eliquid
    Offline

    eliquid Platinum Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    828
    Likes Received:
    3,009
    Joined:
    May 29, 2013
    Gender:
    Male
    Location:
    Louisville - Kentucky
    Rep Bank:
    $24,623
    Hello everyone.

    A lot of you know me for my digital marketing Gold Thread here on the FLF ( see sig ), as well as other posts in the forum.

    What you might not know is that I've successfully built 6 profitable SaaS programs either solely by myself, or with a single partner ( and with no other employees other than VA's ), and that I am working on building my 7th SaaS

    I've learned a few things in SaaS over the last 7 years that span:
    • Handling competitors
    • Reducing churn
    • Fraud reduction
    • Technology for SaaS needs
    • Big Data ( billions of new data points daily over years )
    • Increasing LTV
    • APIs
    • Pricing
    • Onboarding
    • Project management
    • Marketing
    • Customer service
    • Partnerships
    • Customer demos/profiles, MVPs, UVP's, ahHa moments, etc
    • many many more things

    I've focused all my SaaS programs in the digital marketing space, but I have some ideas for new SaaS programs expanding outside of that for the future.

    I can't answer questions related to:
    • Legal - please seek an attorney
    • Specific finance questions about my current or past SaaS programs - I'm not going to divulge other than generalities to the public. I can verify for a mod though if needed.
    • Info on verticals outside of digital marketing - meaning if you have a SaaS for doctors and you ask me a specific medical question, I won't know it if it pertains to doctors or medical
    • LLC vs Scorp Vs etc - This is legal
    • How long is a piece of string type questions

    And before anyone asks.. NO, not all 6 SaaS are currently running right now. I closed down the first 5 over the years at different times due to either partner problems or interest died off for me and I rolled into the next SaaS combining ideas to make something new.

    For clarification, I am running 1 active SaaS right now and building another ( the 7th ) that is not public atm since it is not finished.

    Ask away!


    P.S. - Listen, I'm a different type of person. I have very unique views that don't always fit the norm you might have heard elsewhere. What I tell you is what has worked for me and the way I see things from my own personal experience. There are many ways to skin a cat. If you don't agree, that's cool but always think things over for yourself and what will work for you.

    .
     
    Last edited: Oct 14, 2017
  2. inputchip
    Offline

    inputchip Bronze Contributor Read Millionaire Fastlane I've Read UNSCRIPTED Speedway Pass

    Messages:
    105
    Likes Received:
    224
    Joined:
    Mar 26, 2017
    Gender:
    Male
    Location:
    Alberta, Canada
    Rep Bank:
    $1,259
    Oh I've been waiting for this one! Thanks for doing this @eliquid, we all have a lot to learn from you i'm sure!

    1.) What is your preferred programming language for doing SaaS products? I have been hearing Ruby on Rails is usually the way to go.

    2.) Would you recommend, for someone like myself who is somewhat familiar with the basics of other programming languages (python, fortran, javascript, html, css) to learn another language to undertake the programming myself, or should this be left to the experts?

    3.) How have your partnerships been structured in the past? I'm currently in negotiation for a partnership. My partner wants a larger percentage as we are pursuing his idea and it is within his industry, however I have a much more technical background and will likely be doing more leg work.

    4.) Have you presold any of your previous products? Or any other form of validation? Or did you go straight into building an MVP?
     
  3. Denim Chicken
    Offline

    Denim Chicken Silver Contributor Read Millionaire Fastlane FASTLANE INSIDER Speedway Pass

    Messages:
    403
    Likes Received:
    646
    Joined:
    Jun 5, 2010
    Location:
    California
    Rep Bank:
    $3,670
    Hey Eliquid,

    I've been wanting to start my own Saas for a long time.

    Do you code yourself and handle the technical stuff hands on (or at least initially like @James Fend) or do you always try to partner with someone and you handle the business and sales?

    My biggest problem is the limit on time and energy to learn coding to get to the functional level. But no matter how I slice and dice it, unless I trust the dev completely, it's something I need to do myself.

    Any advice when it comes to gauging demand and competition before setting out for a build?
     
  4. c4n
    Offline

    c4n Full throttle Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER

    Messages:
    54
    Likes Received:
    88
    Joined:
    May 30, 2017
    Gender:
    Male
    Location:
    Europe
    Rep Bank:
    $458
    I for one would absolutely love to hear your input on ALL of these points. Like a quick summary/lessons learned guide maybe?
     
  5. TheSilverSpoon
    Offline

    TheSilverSpoon Bronze Contributor FASTLANE INSIDER Speedway Pass Summit Attendee

    Messages:
    144
    Likes Received:
    405
    Joined:
    Nov 4, 2013
    Rep Bank:
    $2,182
    Perhaps this is a little open ended - but from a high level, what does your early marketing strategy look like? What do you do to get the first, say, 50 users on board?

    I've built and managed the builds for a number of SaaS products now and have a strong grasp of the tech side, but still seem to struggle with the marketing.
     
  6. eliquid
    Offline

    eliquid Platinum Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    828
    Likes Received:
    3,009
    Joined:
    May 29, 2013
    Gender:
    Male
    Location:
    Louisville - Kentucky
    Rep Bank:
    $24,623
    My preferred is PHP, but only because I know it like the back of my hand. I can get to a concept quickly, validate, show a working model to others, and start the next task ( which might be to hire someone to code it in another language, but more than likely is to start making money ). PHP was built for the web, so it's my goto for starting out. However, as your needs grow you will need parts done in other languages.

    Each of those areas have their speciality. For example, does your SaaS deal with Machine Learning AI? Python has excellent libs for this. Does your SaaS have critical areas that need the fastest processing possible, then C or C++ backed with Redis or SSDB is excellent for those tasks. Designing the interface.. then HTML, CSS, PHP would be choices on that area.

    Personally I like to make everything in PHP, get it making money, and work the kinks out later with new code. Most coders will tell you NOT to approach this the same way, they will tell you to code it right from the beginning. Neither of us is wrong, but I code to make money, not to make "pretty code" that follows some academic standard. Without the validation of money, nothing else matters.

    Both of you are valuable. His idea and his domain along with possibly his contacts; your legwork and sweat and also more than likely your upgrading of code and having to hit deadlines. I wouldn't do anything less than 50% each for ownership with an agreement that includes how decisions are made like maybe you are the tie breaker for code issues, he is for features added. This isn't needed because you both should be able to come together, but just in case you might need some help in who has power in what area or speciality to move things faster possibly.

    This is both a yes and no.

    I haven't purposely done this ( like with a coming soon landing page, kickstarter, etc ). However, because all of my SaaS up to this point have been in digital marketing, it was possible for me to use existing lists, authority, and name to sell it in the same fashion. People found out I had product X, and they just wanted in on it because of their relationship with me.

    The first 5 SaaS programs I had were pretty much products that were already existing, but I made them 1000x better. The 6th was something similar to other products, but entirely different and approached solutions differently. The 7th is sorta like the 6th.

    Validation wasn't needed with the first 5, but as people learned about the 6th it really caught wind so there was no validation for it until I built it with my partner. However, it personally scratched my own itch so I knew I could pull it off. Now that we make some good MRR and our competitors copy us left and right daily, I feel validated.

    Since I can code myself, I generally go right to MVP. If it doesn't work, I've lost my time but not my money ( in paying another person ) and I build it out in sections of MVPs. Meaning, I don't built the full product as a MVP.. but maybe a one function of the product as the MVP. If that one function catches on, I build the next one until I have my beta product, which is a MVP of MVPs.

    If you code something and it doesn't catch on, you can always try to sell it on Flippa.

    .
     
    Last edited: Oct 14, 2017
    Kingmaker, 100k, James Fend and 6 others like this.
  7. eliquid
    Offline

    eliquid Platinum Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    828
    Likes Received:
    3,009
    Joined:
    May 29, 2013
    Gender:
    Male
    Location:
    Louisville - Kentucky
    Rep Bank:
    $24,623
    I've always coded my SaaS programs first by myself. Some of them ended up completed/finished by myself, or finished by a partner. However, I always generally have coded the "idea" first, then either finished on my own or with a partner co-coding... or them finishing it for me.

    I feel the same way a lot. However, you have to let this feeling go if you want to grow and have a partner. Trust in your partner is Paramount. Play to your strengths though.. as in can you market well? If so, let someone do the coding while you market. If you try to learn coding and you have gotten nowhere because of time, it will always be this way and you will get nowhere for a long time and potentially miss your opportunity when someone else builds it for themselves.

    There are lots of ways to do this, but the easiest and simplest is to improve another product or idea and apply your twist to it.

    Past that, it's knowing your market really well. There is a reason all of my SaaS programs are in a domain I know extremely well ( digital marketing ). I know the pulse, I am the market because Im the consumer too that has been in the market for 2 decades.

    Past that, I would look for your early adopters and talk to them a lot. My early adopters hang out on forums and Reddit. They are also in certain Skype and Slack groups and we go to certain conferences. My early adopters will tell me if they want this product and then with their help, they join and spread the word for me which allows others who are not early adopters to come in. Where are your early adopters?

    [​IMG]


    When you combine all of these, it's hard to miss the mark
    • A domain you know extremely well, where you are the customer too
    • Programs you have competitors in, where you copy and improve their product but also add a twist to it that provides you an "advantage moat"
    • Listening to your early adopters
    .
     
  8. eliquid
    Offline

    eliquid Platinum Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    828
    Likes Received:
    3,009
    Joined:
    May 29, 2013
    Gender:
    Male
    Location:
    Louisville - Kentucky
    Rep Bank:
    $24,623
    This would be hard. I love to share value and talk a lot.

    I also love to overshare to make sure people really understand what I mean.

    I might have to push your answer down some and circle back later as it will take some time to do it right. If you don't see an answer in a few days, I'm still working on it internally, but also some of them might get answered in other posts naturally too.

    Will try to do something for you though in some way, at some time too in this thread.
     
    Last edited: Oct 14, 2017
    c4n and Andy Black like this.
  9. eliquid
    Offline

    eliquid Platinum Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    828
    Likes Received:
    3,009
    Joined:
    May 29, 2013
    Gender:
    Male
    Location:
    Louisville - Kentucky
    Rep Bank:
    $24,623
    The first 50/100/200 have been easy for me and it feels a little like cheating.

    I've touched a bit above on why, but to put it another way/angle:
    • I have a "small brand" / personality in the digital marketing world, especially with a core group of digital marketers ( SEO's and Affiliates ). So does my current partner. Past partners have too somewhat. This "brand" helps. If Kim Kardashian had a line of makeup come out today, how easy would it be for her to get her first 50 customers? Extremely easy.

    • I've ( up to this point ) worked exclusively in the digital marketing/web vertical for 2 decades myself. My SaaS programs have exclusively been in this too for the past 7 years. By nature I know all the places people like me hang out, our lingo, who the key influencers are, etc. If I didn't have my "brand" name, I easily know where I could get 50 customers quickly because of this domain knowledge. It also helps I'm a marketer too, instead of say an accountant.

    • If I didn't have both of the above, I'd have to talk to key influencers and get them in the door of my product. Then I'd have to keep close relationships with them and improve on their feedback. In the early days of one of my SaaS programs, we had people sign up and use our product that were speakers at digital marketing conferences. We didn't seek these people out, but they found us and used the product and LOVED it. They went on and spoke about us in front of thousands of digital marketers at conferences and shows and we only found out when the tweets, slide decks, and recordings came out from the event. Talk about a lot of free, valuable, targeted advertising that was the best kind possible... word of mouth ( social validation ). This has happened several times for us.

    • Early marketing pushes have also included a lot of content. On our own blog, interviews, podcasts, guest postings, in-depth training guides, etc.

    • If you can make yourself an "expert" ( look at what @Andy Black did here at this forum ), you will have an easier time. What can you do to make yourself the expert starting today?

      .
     
  10. ChrisJHarrington
    Offline

    ChrisJHarrington Bronze Contributor Read Millionaire Fastlane Speedway Pass

    Messages:
    282
    Likes Received:
    445
    Joined:
    Nov 27, 2014
    Gender:
    Male
    Rep Bank:
    $2,742
    What's the highest ARR you've scaled a SAAS company to?
     
  11. eliquid
    Offline

    eliquid Platinum Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    828
    Likes Received:
    3,009
    Joined:
    May 29, 2013
    Gender:
    Male
    Location:
    Louisville - Kentucky
    Rep Bank:
    $24,623
    For @c4n - Handling competitors:

    I never thought I would need to handle competitors. At least not in the way I do now.

    How I was raised, I was stuck between the Baby Boomers and Generation Y. I was raised with the thinking "working hard long term" was all I needed, but living in a reality that was actually different as it unfolded in front of me growing up today.

    At some point I believed that if I simply worked hard on my product ( or career ), people would just want to hire me and pay me well. Either for my knowledge or my product. I really thought people would love my products so much that competitors would just hand me money for it and buy me out. It took a lot of "real-life" events and FME to change my ingrained teachings.

    Instead, my competitors steal and rip me off and don't even give a reach around. It's blatant and rampant. It happens so much I get numb to it.

    While this hasn't always been me, this is me today when it comes to competitors..

    I'm coming after you and I don't care what I have to do, I will carve my name in your forehead and take everything I can from you. There isn't room for all of us, so you're going down. If you're drowning in a lake and expect me to help you up, I'm sticking a water hose in your mouth at full blast.

    If you watch the "Walking Dead", I use to be the early Rick. Today I am the Negan.

    Playing nice just doesn't work. Getting along just doesn't work. I've tried multiple ways and different tactics. When you create something new and someone steals it and rips it off and you're the one having to constantly change an industry and others benefit in the billions of dollars while you are struggling to come up in the world, it stings.

    If you're my competitor, I feel sorry for you. I got 99 problems but a competitor ain't one.

    I don't have hate for my competitors, they are just in my way. Sure I would get salty each time they stole a feature from me, but I've realized with tremendous help from my partner that we are always 12-24 months ahead of each of our competitors and while they spend their time playing catch up, they can't innovate on new features or benefits.

    With that thinking, I have been able to grow a lot.

    Our competitors can not pivot as quickly. They are in "reactive mode" instead of "innovation mode". We are taking their customers away from them every day. We get looked to as the leaders, while they scramble around. We build moats, while they incorrectly copy features they don't really understand the "why" behind.

    I thank my competitors for showing me a profitable niche, but I drown them as quickly as I can afterwards with no mercy whatsoever. It took me a long time to get to this point, but "things are a changin" now.

    I focus on innovation now and I try not to look at what they are doing. It's best for mental and physical health. Focus on your projects and not so much what they are releasing. Listen to YOUR customers, not theirs. If you built your product correctly, your product will be different from theirs even though it might be the same similar technology and functionality.

    If your competitors are stealing from you and finally trying to upgrade their product after a decade of not doing so, you're doing something right.

    .
     
  12. eliquid
    Offline

    eliquid Platinum Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    828
    Likes Received:
    3,009
    Joined:
    May 29, 2013
    Gender:
    Male
    Location:
    Louisville - Kentucky
    Rep Bank:
    $24,623
    For @c4n - Reducing churn:

    I've been working this a lot more recently.

    Here is what's funny. If you search the web, you will find a ton of people talking about this and every single one of them have a different way of calculating it and what's an acceptable number for customer churn.

    We us Chart Mogul since it integrates with Stripe and I go off their numbers for our customer churn. I look at our numbers and ask myself how do I improve these numbers, not "this is where XYZ person told me I need to be".

    Our churn numbers were all over the place. It was hard to really pinpoint why.

    You could come up with 1001 reasons why, but only a few actually matter and can be impacted by you.

    I started sending out an exit survey when people left. Very very few people would actually fill it out, so it took awhile to get good data I could trust.

    I looked in our customer chat logs from Intercom ( another service we use ) and looked at common questions and complaints.

    I talked to my partner a lot about conversations he would have with others in Slack or Skype. I would also talk to customers on the phone and email. We combined all this data to spot trends.

    While I could list out a 1001 reasons ( hey, Im an INTJ ), it really boiled to some common themes that we could actually impact and wrap our heads around:
    • People could not afford us - Really? Is this a generic nice excuse with a different problem behind it? At $XX.95 a month I couldn't figure out why they couldn't afford us. We had cheaper plans before ( several ), but all the problem clients came from those tiers. My first thought was if you can not afford $XX.95 a month, we just were not the solution for you since we had a lot of troublesome clients from those tiers. Later I realized that maybe we simply need to add more value so that when you canceled it wasn't that you the customer were saving $XX.95 a month, it was you were going to miss out on XYZ that was a great value that would pain you not to have anymore.

    • People didn't know what to do - So our tool breaks new ground. We are almost that tool that is 10 years ahead of it's time almost. Because of that, a lot of people need to be "retrained" in how they look at SEO. A lot of advanced people get it, a lot of beginners or people who just go through the motions don't get it. This is a challenge. When they sign up and get in, they are almost overwhelmed. Overwhelm creates panic and paralysis by analysis. People don't get their ahHa moment, people get frustrated, people then cancel. I've learned that even if you dumb down things for users and give them a ton of videos, they still need hand holding every step of the process and they won't watch the videos anyways most times. Force them to take certain actions if you have to within the design of your program, send them multiple emails on 1 topic if you have to so you can get them to notice and take action. Restrict functions to them until they do X actions, etc.

    • Alienation - I heard a lot of "this looks like a tool for an agency, not someone like me" and various similar type of talk. Not true at all. I wasn't speaking my demo's language, even though I was the demo. As an advanced digital marketer, I made the mistake of talking like one. I worked with my partner to identify our core 3 users. Anymore than 3 and you just can't market well overall. After we decided who our 3 users were, we created customer profiles around them. We named them, gave them job titles, what their boss expected of them daily, what they needed in their job, etc. After some time, we were able to developer onboarding messages ( along with marketing messages such as blog posts, emails, ads, etc ) that were able to touch on all these points. People now felt we really knew and understood them and that we were a fit for them once we showed them why their "type" needed us.

    • People were not having their ahHa moment with us - So what is your ahHa moment for your customer? You know, that point in time when they say, "Oh shit, I get this now.. I see why this product was invented now". You really need to define this and it might be different per customer profile ( see above ). Now, what are the minimum steps needed to get to this moment? For us one step needed was to create a project within our interface.. your first project that is. Lot's of people just were not doing it. You can't have your ahHa with us if you do not create a project, so we creating onboarding to focus on making a project. We also forced our users to create a project as soon as they sign up. They can't do anything else until that project is made. Other people weren't logging in enough their first 30/60/90 days, so we developed onboarding and triggers for emails that addressed that to get them back in and using. I saw people unsubscribing from email communications with us which meant they wouldn't get our blog posts, but more importantly our onboarding and triggered emails, so I created in-app messaging to overcome that and still convey value to them and get them to take action. All in all, you have to almost push certain people into having their ahHa moment sometimes.

    • Creating more value - Yeah, you're X. Your tool does X. However, if that's the only value you have, some people are going to leave once they get their Y from your X. Maybe they will come back in 9 months when they need X again for a side project. You're really good at X, but the user only needs X for 2-3 months for whatever reason. When you add more value, like say training or information.. now the user looks at you as a source of knowledge they can't live without ( if you do it properly ). For example, I might offer a PAID ONLY benefit that paid members get a monthly PDF of info that free users ( the general public ) don't get from us. If my SaaS helps their SEO, my PDF this month might teach them how to get new clients for their SEO agency. See, my SaaS helps them with SEO on a technical level, but my extra value for paid members teaches them the business of SEO which helps them make more money. If you cancel your sub, you don't get these guides which by themselves are worth a lot of money on their own.

      Reducing churn just a small percent means massive LTV gains, which equals more MRR and ARR. The chart below shows LTV recently.

    [​IMG]

    In the blue box, I started working on churn as my sole focus. Since that month the churn has gone down 3 months in a row. Before it was up and down, up and down.

    Notice that as churn goes down, LTV goes up.

    A customer that was worth $697 before, is now worth over $1900 simply because we lowered churn.

    Notice that in Sept to Oct, the churn only moved around 1%, but that created $600+ more revenue PER customer. You only have to move churn a small % to make massive gains.

    I know the math on this.. If I can get down to 1.2%, the LTV per customer is in the mid $3k range. I know I can get to 1.2% because I've been there in the past before it moved up the next month. My goal is to get right at 2.00% and keep the churn stable month from month.

    I've thought about writing more and more on reducing churn and helping SaaS companies improve their churn rates. Again, just a little adds a lot to the bottom line.

    .
     
    Last edited: Nov 1, 2017
  13. eliquid
    Offline

    eliquid Platinum Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    828
    Likes Received:
    3,009
    Joined:
    May 29, 2013
    Gender:
    Male
    Location:
    Louisville - Kentucky
    Rep Bank:
    $24,623
    For @c4n - Fraud reduction:

    This one I won't talk too much about.

    Mainly because, I might actually develop what I know into a future SaaS. Actually, I've already semi-done it because what I know I built into an API and my current SaaS uses it.

    Therefor, I can't give away the secret sauce here.

    But I can say this. This was developed before many of the current fraud SaaS companies came onto the scene. This was before processors like Stripe had a fraud feature ( theirs is called Radar ). I don't want to talk too much about how mine works because as you see I have competitors who are validating the market for me and I want to see their mistakes before making mine public and taking my slice.

    What I can tell you is this... fraudsters are very smart.

    Most of the fraud I find happens when someone is "testing" a credit card. This just might be my personal experience in SaaS. Remember, I don't run ecommerce for myself, so you ecom guys will have a different experience.

    But people are not using fraudulent cards to actually use my SaaS. They get a card, sign up to my SaaS and then never use it.. ever.

    I suspect they do this to a lot of companies, testing the card "safely" on a 3rd party to see if the charge goes through.. then using it in a situation to buy from ecommerce or as validation for selling the card to another person. We don't get many of these types of transactions, maybe one every few months but its always this pattern of purchasing and not using the SaaS.

    So, if you see someone come in and purchase something, but never use it.. you might have a fraudster on your hands.

    Another type of fraud I see is "friendlier" where someone signs up and uses our product, but 6 months later stops and tells their card company they don't recognize the charges and they do a charge back for several months at once, after they used our product for months.

    This is a bit more serious to me, so we keep data on what every user does whenever they are in our SaaS. I mean everything.

    We send this data on to Stripe and we generally win all our disputes. We have a few issues with companies like American Express, but generally we win most of our cases. You need to keep excellent records to battle disputes because losing too many disputes can get your MIDs and merchant processing canceled. You want to win, not lose these cases.

    .
     
  14. eliquid
    Offline

    eliquid Platinum Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    828
    Likes Received:
    3,009
    Joined:
    May 29, 2013
    Gender:
    Male
    Location:
    Louisville - Kentucky
    Rep Bank:
    $24,623
    For @c4n - Technology for SaaS needs

    This is a case of where you have to use, what fits your needs. Therefor this post might be a bit vague as everyone will have a different need.

    I like to build MVPs. I might build 10 MVPs and only 1 or 2 actually make money. The ones that make money I then push live and tweak as needed.

    Therefor I am not properly planning out tech beforehand for each MVP. Why do all that work and spend months on it if it doesn't make money?

    But once I get to a point where something is public, tech is very important.

    And the truth is, even highly experienced programmers won't know what tech is needed for each part of your SaaS until you hit a problem. Sure they can say they know and quote "Best Practices", but they don't know either until you hit a roadblock, or the path to a roadblock is so clear you see it months before it happens. Even then, months might not be enough time to act.

    I'll give you an example.

    One of my SaaS products was never meant/intended to handle big data. It just wasn't expected. We had a plan and a system to prevent having big data.

    A couple new features roll out, we got a massive amount of customers, new pivots came in, and now we are processing billions of data points daily and storing them for longer for many more customers. In a series of events that took place in less than 30 days, we outgrew our tech in less than 45 days.

    It situations like these you can't plan for or know about until they happen. Which is why I don't spend months ahead of time prepping and getting nowhere. Money first, always. Then react if need be because reacting with no money coming in will lead you nowhere.

    At that time, we had to change plans.

    • SQL needs sharding now, and what's our shard plan?
    • The processing of data now takes forever, we need a quicker solution. Some code changed from PHP to C++
    • MySQL on any server is getting bombarded, lets offload to Redis certain features which also makes them faster to process
    • As we write more code, more needs to be managed. Here comes git or a git like solution
    • We are spending a shit ton on servers. What data do we keep HOT and which can we keep COLD in something like Amazon Glacier?
    • With so many servers, how to we keep up with updates, upgrades, and security measures?
    • PHP 5 is at EOL now? Fuck. PHP 7 is faster and maintained, however certain libs have to be redone and resintalled too.
    • We do more writes than reads on our data, which is the best solution for that.. that also has less overhead?
    • Do we keep everything in 1 DC, or global? Whats the cost impact?
    • System daemons, cron jobs, or Supervisord? Queues or rolling reserves for data processing? Can ElasticSearch helps us or should be go Hadoop and Apache Spark?
    • Should we cram all it in Amazon Web Services or build our own?
    • Look at this sexy new Go or Rust lib, we should incorporate this even though no one knows Go or Rust.
    • Which feature should be coded next? What has priorities?
    • With some of our customer data on 3rd party platforms, how do we ensure that data is safe? Safer than safe?
    • Sometimes the newest version of something, is the one that breaks the most. I like to stay 1 version behind unless it's a major security issue or the newer version has a benefit I desperately need or want ( like faster processing )

    A lot of times it will realistically boil down to what you know and what you are comfortable with. Also, what's actually available to you as well at the time you need it.

    Don't worry about what's the "best". Worry about what works for you and you can keep up with.

    PHP works for me and I can keep up with it. My partner can work with and keep up with C++. Our data can mingle and interact via SQL and Redis. Neither one of us slows the other down trying to fit a round peg in a square hole.

    Issues and problems come up, we work and code around it to fix it. You'll have the EXACT same problems and issues if you spend 13 months planning too. Therefor, get it launched as best you can and as quickly as you can and see if it makes money and what your customers want and what issues you actually face. Then code to fix and improve.

    Many times you might have something break or an issue comes up and you realize, it's not actually a problem or it's a problem that can wait 3 months to address safely still while you do more important things.

    Generally the simplest solution is actually the right solution for a lot of issues and choices.

    .
     
    Last edited: Oct 14, 2017
  15. Andy Black
    Online

    Andy Black Any colour, as long as it's red. Staff Member Read Millionaire Fastlane FASTLANE INSIDER Speedway Pass LEGENDARY CONTRIBUTOR

    Messages:
    5,372
    Likes Received:
    21,943
    Joined:
    May 20, 2014
    Gender:
    Male
    Location:
    Ireland
    Rep Bank:
    $126,352
    (Shudder) Reminds me of my DBA days. x100!!!

    Fair play to you both managing all of that while keeping the service up.

    I was just about to say the simplest solution is often best, and you finished on that note.

    Simplicity is elegant and beautiful, and allows you to move faster.
     
    BrooklynHustle and eliquid like this.
  16. eliquid
    Offline

    eliquid Platinum Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    828
    Likes Received:
    3,009
    Joined:
    May 29, 2013
    Gender:
    Male
    Location:
    Louisville - Kentucky
    Rep Bank:
    $24,623
    For @c4n - Big Data

    This is still a learning process daily.

    At some point you realize that you have this ever growing monster that will live on forever and it needs to be cared for. Every day it gets bigger and stronger and you have to plan ahead somewhat on how you will manage it at X phase and XX phase.

    The beauty is, it also becomes an advantage moat for you against possible competitors.

    My competitors were not leveraging their big data except for the most simple and direct purpose before my SaaS came along. After our SaaS was up and running with our unique spin on EVERYTHING, our competitors starting diving into their big data and trying to push the same things. For a few years though, we had the first mover advantage for those data sets which was a huge help in growing and establishing our SaaS.

    There is no right or wrong answer on how you handle big data and what you can do with it.

    Our setup is very very simple. If you want to know why, read the above posts I made already.

    We started on one server and upgraded it vertically ( in size and power ) as needed. Soon we found ourselves not able to manage that simply ( plus we were outgrowing physical limits that I was comfortable with ).

    We then started growing horizontally by sharding with new servers that we could keep light and small, just plug in new ones as needed and spread the data around.

    We sharded our data based on a data key that is ever growing. It's basically the ID of a project.

    If I really wanted to spend the time and energy, I might have went with a customer shard key, where the ID is the customer. This creates it's own problems though, so nothing is actually "perfect" or one size fits all. You're going to have issues no matter what you do that you will end up coding around. In a customer shard situation, I would have to recode the backend so certain things could be shared. More work which I don't want really.

    Our competitors can not handle their big data. If you want to see results for something, you might wait days or even a full week to get an update. They look at their data as a burden that has to complete over days and weeks maybe. We look at our big data as "how can we ensure we provide updates daily, no matter what it takes". Just because your data is big, doesn't mean your customers or results ( or processing ) needs to be slow.

    Our big data has to be fast. Therefor we trade off certain things in order to achieve that:
    • We store our data in 1 datacenter so we can use private IPs for fast and free data transfer. If we stored all over, 1 failure could still bust us so why bother? The cost trade off would be too much if we did redundancy in multiple dcs too which would hurt our customers. We do a backup to another DC and provider, but this isn't live data for day to day use... just backup.

    • We monitor how much CPU and RAM our servers are using and this is coded into our scripts. If CPU and RAM is high ( meaning maybe customers are using the system a lot ) we might pause processing our big data for a time when CPU is lower.

    • You need checks in your scripts to ensure data integrity. Last thing you want is to have gone a month collecting a ton of data only to realize it's sitting in JSON files with errors or missing data because of some update elsewhere on your servers or collection. Thats a ton of data to lose or change at once.

    • The data is a gift. Throw an unsupervised AI platform on it and see what it can dig up that your competitors haven't realized. If you don't know how to do this, spend several days a month thinking of how you can use the data and what you might dig up and then try to find it out.

    • You more than likely won't need all the big data, all the time. Figure out what you need all the time and what you can store safely in long term storage that might be slower to get at, but cheaper to store.

    • Sometimes multiple copies of something are fine. Duplication can be fine. However, does it achieve your goals or are you being sloppy? For instance, some of our data is duplicated in Redis, some in JSON text files. Some of this data is already in SQL though. However that SQL server might be at high CPU right now, or slow. The copy in Redis will be fine to use right now and the backup in JSON will too if need be. You need to factor in cost, but sometimes having duplication is perfectly fine IF it solves a need for you.

    • Security is huge concern. Our servers only talk within the internal IP network and none of them have libs or software they don't need. For example, our data servers don't need a webserver or a copy of Python. We lock down the software, libs, and connections. We can't have snoops or security issues.

    • With big data, you have to figure out how you access the data especially if using SQL. You can't have un-needed indexes and keys growing and costing you data storage. You also can't expect to change how you want to pull the data and pull on a non-existent key/index. If you have a lot of data and need to change the schema, you could be looking at hours per server to do this. If you have lots of servers, you're looking at days of work. Sometimes these servers puke and hang while processing.. can you wait it out or do you hard restart and risk losing the data ( well if you are smart you have a backup or course, but now you gotta in-line load that data back into a server, etc ). You could go the cost of replicating the server before hand and working on the clone, only to clone it back to the original when done ( what I personally do ). Again, you have to know WTF you're doing though BEFORE you change stuff.

    • People want to believe this hype that No-SQL is the answer for big data. It's not. SQL just does just fine if setup properly. In your case and situation, you might need No-SQL. Use it if you know for sure you have to have it. Otherwise don't buy into hype for hype sake.

    • Have a plan for offloading big data if you need to. For example, you might need to delete data from a customer that churned. You know how long and hard it is to delete from MySQL normally on big data even if you have the ID/KEY? It can be a long time and it can drag a server down if that server is also processing other things. However, this would also fall under bad design too. A solution would be knowing how to partition a SQL table so a deletion of 1 customers data is instant and almost no processing time. You could also go No-SQL too. Again it boils down to knowing what you need exactly and then knowing how to get it. You don't have to do something because of hype. Again, what do you know and comes easy and is simple for you? Do you know it backwards and forwards? Good, forget about learning something else new and use what's available to you and use it properly.

      Also, you could just spin up a server per customer and then when needed just delete that server. So many ways to skin a cat. Don't fall for hype though.

    • Costs. This can get out of hand quickly. Besides payroll, this might be your largest cost for a SaaS if you do big data. You gotta nail handling of this big data before you end up bankrupt on servers and redundancy and other things that comes with big data. I try to factor in a % myself. If my SaaS makes X revenue, then our hosting/server costs should stay at X% too. If we grow, then the server cost can grow. It should never be above X% though, if it is we have an issue. Maybe we can't avoid the issue, but its an alarm to me that something might be off and need to be rethought and a pivot is needed in that department. You can't have this data eating away profit each month and growing, because it will if left uncheck. Hell, it will if checked. However, you gotta have more revenue coming in to justify it unless its a special situation.



      .
     
    Last edited: Oct 14, 2017
  17. eliquid
    Offline

    eliquid Platinum Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    828
    Likes Received:
    3,009
    Joined:
    May 29, 2013
    Gender:
    Male
    Location:
    Louisville - Kentucky
    Rep Bank:
    $24,623
    For @c4n - Increasing LTV

    Some of this was touched on in the Reducing churn post, so please read that if skipped.

    However, I will touch on other points outside of the churn post here for LTV growth.

    I gotta hit lunch, so I'll edit and update here in this post when I get back

    -- Saving for later --

    -- back --

    Ok so increasing LTV is pretty simple in thought.. how do you get more money out of someone?

    Well, in a SaaS this happens a lot of times by simply getting the customer to stay longer in your program. Instead of an average of 3 months, get them stay 9 months. Boom, LTV has increased.

    How do you keep them on longer? Great question.

    • You gotta make it to where they will experience pain when they leave.

    • You gotta make your SaaS a habit for them in their daily lives.

    • You gotta provide excellent customer service that goes above and beyond your competitors.

    • You have to get them to their ahHa moment and you have to ensure they are not experiencing times where they are a regular member for 7 months but vanish for 40 days without logging in. If you see that, reach out.

    The other way is to upsell and cross sell them.

    Maybe your lower tiers restrict X and limit Y. Maybe they don't get Z.

    You can build in triggers to your software that monitor user behavior and if you notice a user is getting close to their limits, you send off an automated email that offer them an upgrade to a higher plan at a discount if they upgrade in the next 24 hours. Mention how you noticed they were running close to their limit and you would like to help them.

    Maybe their plan doesn't allow them to have Z feature, but you notice they have hit that page on your program a few times. Since they don't have it in their plan, all they see is grayed out page with a warning, or a non-working button. No worries, you send off an email asking them if they would like to try it for free for the next 14 days and once they say YES, you enable it for them as a trial.

    In 14 days you remind access is going away and if they would like to upgrade to the higher end plan to get the feature. If you have done it right, they will upgrade more than likely.

    When users ask to cancel, are you willingly going to let that money just walk away?

    For some SaaS offerings, it might make sense that you offer them a lower end plan that is normally not advertised on your site. If you're lowest end public plan is $49.95 and someone cancels stating they can't afford it, offer them a non-public plan that is a little more restricted than the plan they are on, but at a $29.95 price tag.

    If they still say no, remind them of what they will lose by cancelling. Maybe they signed up during a promo. If they cancel but come back later, they will lose that promo benefit they once had. You can't keep their data either, so if they sign back up they will need to set up projects again and start all over. If your SaaS falls into this camp, you can offer "paused" or frozen plans. For $9.95 we can freeze your plan and you just can't log back in and add new projects or edit projects. This keeps your subscription in good shape and protects your membership while also reducing your cost and allowing you to come back later.

    You might think this is dumb, but I've had people cancel and then come back 2-3 months later. Even 2-3 days later. It happens all the time. Instead of making $0, you could be making something.

    Don't believe me? If you are a member of Click Funnels, they do something just like this when you try to cancel. I have a "paused" account at CF for the last 9 months now. It works.

    Is this stuff sinking in guys?

    .
     
    Last edited: Oct 14, 2017
  18. PeterCastle
    Offline

    PeterCastle Bronze Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    181
    Likes Received:
    351
    Joined:
    Oct 1, 2013
    Gender:
    Male
    Location:
    NH
    Rep Bank:
    $1,817
    Thanks for the thread @eliquid! I have a SaaS that I'm currently trying to grow. The biggest challenge for me has been the marketing. Adwords/Bing have been too expensive. What has worked best for me so far are Apple Search Ads (as I also have an app for the SaaS) but even those are not really sustainable in the long term, as they are still too expensive.

    What other methods besides PPC have you used to market your products? My plan now is to write a short eBook that I will use to get people in the door, and to turn my site's blog into a valuable newsletter on its own (content marketing).
     
  19. c4n
    Offline

    c4n Full throttle Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER

    Messages:
    54
    Likes Received:
    88
    Joined:
    May 30, 2017
    Gender:
    Male
    Location:
    Europe
    Rep Bank:
    $458
    @eliquid

    Thank you so much for sharing your experience; you have way surpassed my expectations with your detailed replies. This thread could turn out into a great guide to anyone wanting to start a SaaS business.

    I look forward to reading more. Rep++
     
    eliquid and Andy Black like this.
  20. Readerly
    Offline

    Readerly Contributor Read Millionaire Fastlane I've Read UNSCRIPTED

    Messages:
    37
    Likes Received:
    55
    Joined:
    Sep 25, 2017
    Gender:
    Male
    Location:
    Portland, Oregon
    Rep Bank:
    $212
    Howdy @eliquid, I'm building a SaaS right now for a niche industry (that's still pretty darn big). I'm following this thread avidly--and will post a question when one comes to me. Great stuff!
     
    eliquid likes this.
  21. Steve F
    Offline

    Steve F Contributor Read Millionaire Fastlane I've Read UNSCRIPTED

    Messages:
    89
    Likes Received:
    96
    Joined:
    Sep 14, 2017
    Gender:
    Male
    Location:
    Canada
    Rep Bank:
    $695
    Fantastic post in a great thread, thanks very much for getting this out there.

    If someone needed to dive in and learn a set of technology in order to build, first, some MVPs to test out, would php be a good direction? What about frameworks? Or even crazier, what about building MVPs in wordpress?

    Just feeling my way here, wanting to settle on one technology direction and take off.
     
    Andy Black and eliquid like this.
  22. eliquid
    Offline

    eliquid Platinum Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    828
    Likes Received:
    3,009
    Joined:
    May 29, 2013
    Gender:
    Male
    Location:
    Louisville - Kentucky
    Rep Bank:
    $24,623
    Probably the most documented, noted, and easiest tech stack to start off with ( if you have little to no tech background in this stuff ) is LAMP, which is Linux, Apache, MySQL, and PHP.

    Is it the right one for you? Only you can answer that though.

    The LAMP stack is documented 10,000x more than any other, so learning and finding working examples blows the doors off anything else. Once you can handle this for an MVP, then you will need to pivot to something that satisfies your need.

    I can not think of a SaaS that would be a MVP from Wordpress, but then again it doesn't mean you couldn't do it. I just haven't thought of a way that Wordpress could handle the needs of a SaaS in the way I think about it.

    One way you could use it is to be your members area for your SaaS. Since Wordpress can have DAP, OptimizeMember, Wishlist Member, etc as membership plugins.. you could use that as your members and billing area and then build a SaaS around it.
     
  23. eliquid
    Offline

    eliquid Platinum Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    828
    Likes Received:
    3,009
    Joined:
    May 29, 2013
    Gender:
    Male
    Location:
    Louisville - Kentucky
    Rep Bank:
    $24,623
    I touched on this in the above posts in different ways but a lot comes down to:
    • Being a brand or authority in your niche
    • Influencers helping you
    • Affiliate program offering
    • PPC ( Big 3, plus smaller offerings like AdRoll, Twitter, Media Buys, etc )
    • Forum marketing ( see brand and authority )
    • SEO
    • Blog posts
    • Guest posting on other sites
    • Podcasts
    • Interviews
    • Skype and Slack channels
    • FB groups
    • Linkedin and Linkedin ads
    • Tradeshows and conferences ( speaker )
    • eBooks or lead magnets
    Think of it like writing a real book or an actor in a new movie. Right before and during the launch, they go on tours for radio and TV stations and they do it nonstop every day. You pretty much need to do the same thing except in the "online world".

    About the best thing you can do is to make yourself an authority/expert and build a small community/following around you. @Andy Black is one of the best examples on this forum that I can show. He is regarded as the Adwords Expert here.. ask yourself why that is and then try to figure out how you can do the same in your niche/vertical and get people following you.

    Those people will spread your message and be your first success stories that will give you validation.

    Like I mentioned before, where are your early adopters at?

    .
     
    Last edited: Oct 14, 2017
  24. eliquid
    Offline

    eliquid Platinum Contributor Read Millionaire Fastlane I've Read UNSCRIPTED FASTLANE INSIDER Speedway Pass

    Messages:
    828
    Likes Received:
    3,009
    Joined:
    May 29, 2013
    Gender:
    Male
    Location:
    Louisville - Kentucky
    Rep Bank:
    $24,623
    c4n likes this.
  25. TTG SS
    Offline

    TTG SS Bronze Contributor Read Millionaire Fastlane FASTLANE INSIDER Speedway Pass

    Messages:
    171
    Likes Received:
    263
    Joined:
    Sep 3, 2014
    Gender:
    Male
    Rep Bank:
    $2,043
    Subscribed.