22
68 Comments

Convince me to use Analytics on my project.

I have a new Saas I made, already have 25 customers (early bird lifetime deal) and probably going to keep building it. Not convinced yet that I need google analytics or even a facebook pixel.. But also I'm not convinced I need any analytics for my site.

My main idea is that it's easier and simpler to build it without having to add in any cookie consent.

Am I missing something BIG?
Am I supposed to use analytics, of some kind? HotJar?

Convince me to install something... if anything.

posted to Icon for group All Things Analytics
All Things Analytics
on December 13, 2021
  1. 9

    You should install analytics on your site so that you have numbers to obsess over every day.

    It's like checking the stock market or the price of Bitcoin every morning: knowing the numbers won't change what you do, but it'll help convince you that you're in control.

    ;-)

    1. 1

      This is like the winning reply of the day for me. It convinces me I'm right in not even giving myself the chance to get distracted at this point in the project.

      1. 1

        You don't need to get distracted now but having a base level of data means you can detect changing trends later, when you do care about how your marketing activities affect visits.

  2. 2

    Are you convinced the other way around, that you don't need analytics? If so, why?

  3. 2

    Here's the answer: Analytics is just a tool for understanding your users. If you can have a conversation with everyone who visits your site, that's amazing (and I'd love to purchase any solution that let me do the same!). That's a much better tool than analytics.

    BUT, if you can't, analytics is a less-effective-but-still-nice way to learn some about your users. For when you want to learn about users on your site you can't talk to for whatever reason, analytics is there to fill some of the gap.

  4. 2

    How do you know which marketing channels to prioritize once you're not sure where people came from?

    1. 1

      Which marketing channels have you prioritized based on knowing where people came from?

      1. 2

        Was surprising to see that I put 3x less effort on LinkedIn and got 3x traffic from there.

  5. 2

    For me is more crucial go backwards.
    Start from money in your bank account, then payment system, then how to convince people to pay you, then number of people know your service.
    If you are able to define your chain of value, you can define KPI on that and understand impact of your changes.

    For example:

    • more visitors means more money in your back account?
    • a more quick/friction-less system related to payment system means more money in your back account ?
    • a better support system on payment problems means more money in your back account?

    and so on... you can do experiments and I think this is the right way to move, one step at time of course

    1. 1

      Can you give me an example of an experiment you've done? And you've seen in increase/decrease based on using some kind of analytics (I'll assume Google Analytics unless you say otherwise)

  6. 2

    In order to grow, and set your KPI's, to create a salesplan, to know your cashflow (thinking you are doing great but when you are growing rapidly, you can't expect to have a clear overview).
    It is great you are talking with your users directly, I think that is a huge bonus and helps you to grow best! But once you grow more and more, keeping the direct line with your users will eventually be harder to maintain.
    From personal experience, having my data in check just helps me make decisions in where I want to invest in my company, what is serving my users, and what is not. Where do we see a decline and where do we need to raise more awareness...

    I believe Data should be used to create as accurate as possible assumptions of the near future and not just to show off growth rates or just because the investor wants to see results.

    And I'm personally not a fan of Google Analytics, but definitely a YES to hotjar :)

    1. 1

      So you've used HotJar? How has some data you've gotten from HotJar helped your product?

      1. 2

        yes, it helped knowing where the user stops in our SaaS, or what they click on, and if it is something that more users do, it's good to shine a light on that part and "fix it". I also noticed the number of mobile users on our website (yay to data) and fixed the look and feel of scrolling to increase their time on our platform and website.

        1. 1

          Also noticed that a fair amount of visitors to the landing page are using a mobile device and that they didn't convert. Which made sense since I had ignored that.

          So +1 to your answer!

  7. 2

    You should have "analytics" on what matters - are you collecting the money you should be? Are trials (if applicable) converting? Are your customers solving their pain/finding joy/making widgets/whatever they do with your SaaS?

    If traditional website hits and similar stats don't matter to you or to your goals, don't collect them. But if one of your goals is to turn web traffic into sales, you should probably collect them so you know where you're at.

    None of this needs to be Google Analytics or Hotjar or whatever. Roll it yourself or find a tool that measures exactly what you care about.

    Lots of people are used to having the baseline level of data that Google Analytics provides, but in many cases that data isn't necessary for the business to run, people just like to look at it.

    1. 1

      I've never heard of rolling own analytics. I mean I have user stats of course. Which users use what and how frequently. But like web traffic stats? Is there open source installable something for that? Something other than GA?

      1. 2

        It's not hard to have a database table that has time, user ID, and url (or whatever fields would matter).

        Plausible analytics is open source if you want to self-host: https://github.com/plausible/analytics

  8. 2

    When you're ready to scale, you will need data and analytics to inform you on your growth strategy.

    Without analytics, you're blind and you will definitely make mistakes that you would otherwise avoid had you had proper insights.

    1. 1

      Gotchya, so when I'm ready to scale, then install something. I'm just building a product and people are using it.

      1. 3

        No. When you're ready to scale you will wish you had data to give you an understanding of HOW to scale.

    2. 1

      Without analytics, you're blind

      No. Analytics =/= eyes. Analytics =/= insights. Analytics is just a way to capture a certain kind of data. And it is not the only way to capture data nor the data it captures is the only type of data that matters.

      It is, however, the easiest, therefore everyone everywhere now installs that without giving it a second thought.

  9. 2

    How do you measure weekly active users?

    1. 1

      I don't measure weekly active users. Do you? why?

      1. 1

        Let me explain. WAU is an example of KPI. In many cases, revenue is constant multiple of WAU. In my case, my app should be used weekly not daily neither monthly. So I measure WAU with amplitude analytics (not the number of signing, but the number of users who use specific features). I don't have revenue yet. Have you set a KPI?

        1. 3

          There's a missing opportunity to define KPI -- Key Performance Indicator.

          There are all kinds of relevant KPIs that people use to keep track of the health and growth of their app or business.

          Most all these are tracked on a Day/Week/Month/Quarter/Year basis:
          Revenue
          Leads (if you collect non-paying leads)
          New Subscriptions
          Cancelled Subscriptions (Churn)
          AOV = Average Order Value
          LTV = Customer Lifetime Value

          If you do paid advertising, then you add in:
          CTR - Click Through Rate
          Conversion Percentage
          ROI / ROAS - Return on Investment / Ad Spend

          There are more, but those are the most common that people look at outside basic website traffic analytics. You may not care about any of those right now, but you might later.

          Disclosure: I'm senior developer for a marketing analytics platform SegMetrics, so this is something I live and breath every day.

        2. 1

          So you measure WAU. But you don't have any revenue. Therefore your WAU(an out-take level metric) is not connected to your revenue (impact level metric). So at this point, how is it not a vanity metric?

  10. 2

    Processing data in a correct and privacy friendly manner comes with lots of advantages.
    For example, our product appmetrix.com is privacy focused and cookie free. And js tracker code is less than 1 kb so it doesn't affect your site or visitors in anyway.

    1. 2

      but it affects me because I have to install it. I'm not so worried anymore about the privacy aspect as the distraction aspect.

  11. 2

    I am a very pro-data person but over the past couple of years, I have developed a very anti-analytics point of view. And most of the time it is because people think when they plugged in an analytics product to their whatever, they feel data-driven when they were only measuring novelty metrics.

    I do advise sorting out the fundamental stuff (minimum viable brand, minimum viable product, customer/client profiles, funnel/journey/flywheel, necessary assets and channels) and generating a strategy before implementing analytics. In my opinion, analytics is a tool for evaluation and control and without the fundamentals + strategy you don't have anything to evaluate and control, therefore it is unnecessary.

    Now, people would say analytics is the way to sort your fundamentals. I don't agree. In the early stages, the data you would collect via the analytics would be from a very limited number of people and therefore would be skewed. Without a significant sample size analytics will provide only problematic data. Instead of that, one-on-one user research would be much more valuable to understand a whole lot of things.

    1. 1

      Thanks for this assessment. It feels like up to this point most others have said things like "no-brainer".. or have completely assumed that analytics is needed. But without explaining why.

      At least with this project I've had a few people who share their screen with me as we zoom and I can see what they are doing. It's been amazing. I think that visceral experience is way better than aggregating numbers.

      So it seems from your POV that I pick analytics later, as a tool, when I have a strategy. Pretty early on now and just plugging away at adding features that people have asked for so far.

      1. 3

        This is my opinion. Everything else written here is everyone else's opinion. If they can explain, it is great. If they are just repeating it without explaining, it is a mantra. It is dogma. I don't like dogmatic stuff.

        In the end, it is your product, it is your business. You should assess your situation and decide.

  12. 2

    Had I not installed Splitbee on my personal website, I would never learn that someone had shared my blog post on Hacker News and it made it to the front page.

    Of course SaaS is different from a blog but you at least have a landing page and you want to know where people are coming from.

    1. 0

      So how did you know it was on the front page of HN change your business?

      if you know it or not, it still happened.

      1. 3

        It allowed me to jump into the conversation that was going on about my blog post right then and there.

        1. 1

          You could have created a google alert (or similar) with your site's URL and that would informed you even when there were zero clicks.

          But nonetheless, did jumping into the conversation help materialize any business impact? More revenues? More subscribers? More anything?

          1. 2

            I could have, but did I?

            I don't know if it helped directly, after all it was just a blog post, not a product landing page.

            I was able to address the questions people had regarding my story and I earned some reputation on HN, which I wouldn't otherwise get.

            And I got some signups on my product and some Twitter followers, but it's impossible to tell I'd get less of those if I hadn't responded.

            1. 2

              That's the thing, we all chase the dragon that is virality. And we get a certain high when we find it.

              But none of us has the actual attribution modelling in place to understand how it affects our results. And even if you can attribute the viral spike to actual results, most of the time you can't recreate it.

              1. 2

                Well, then there's the thing where without reputation, your HN comments won't be visible to others.

                I don't know how more practical this example can get.

                1. 2

                  It is only practical if you plan to use HN as an acquisition channel (which I am not sure if works if you are not a YC member).

                  Otherwise, it is a personal thing. And there is nothing wrong with it. You can practically boost your business by putting your personality behind it. Some people do that (Branson, Musk) but some people don't (Thiel). It is personal style.

                  That said, I think you can earn reputation on HN without engaging with your own content. So if you want to buff your karma on HN, just focus on that. Analytics does not really connect with that...

                  1. 3

                    All your points are valid.

                    The question was about whether or not analytics could be of any help, and I provided just one example how it helped me.

                    I am happy if this information is helpful to anyone, and I am not unhappy if it is not.

                    Glad I was (un)able to help :)

                    1. 2

                      I enjoyed our discussion nonetheless. Thank you for your time.

  13. 2

    On your point about cookies: I've made an analytics tool called Squeaky that doesn't use cookies, it's just as powerful as hotjar but way easier to use. In fact the premise when kicking off the project was Hotjar level functionality whilst allowing businesses to respect their users privacy. That said, it turns out most of our current customers don't care about that...but the functionality is all there! haha.

    I think everyone has their own unique reasons to use analytics tools, but fundamentally it's an efficient way to gain insights into your users behaviour without having to sit over each of their shoulders. These are the top reasons to use analytics tools amongst our current users: conversion rate optimisation, customer support (e.g. reviewing the customer session related to their feedback), monitoring and improvements to onboarding, gather customer feedback...but yeah theirs so many reasons.

    Anyway, I understand your skepticism, I'm not actually going to try and convince you of the value of it, just answering your main questions :)

    1. 1
      • " an efficient way to gain insights into your users behaviour without having to sit over each of their shoulders. "

      I like sitting over their shoulders though...
      I find it way more revealing than seeing stats.

      1. 2

        It's not either/or 🙂 You can and should be using both - in the last 2 weeks I've interviewed 4 users, but I couldn't schedule more than that, so I've also watched over 100 recordings of visitors to my product to observe where they've experienced friction during onboarding, or where they're dwelling longest on our (admittedly quite sparse) website.

        I think you're talking about aggregated analytics data and charts, and I'm thinking more in terms of that, but coupled with screen recordings and user feedback etc, so a bit richer.

        You also don't only want to talk to users that are sticking around, or in some captive scenario like an interview - you can learn a lot from observing the users who bounce or churn quickly who you'd never manage to get in a call or face to face follow-up.

  14. 2

    It really depends on your current goals, are you trying to improve acquisition and understand your marketing efforts? => Google Analytics will help

    Are you trying to improve the UX of your product and still not sure if it's intuitive and easy enough to use? => Go with Hotjar/FullStory, it will give a lot of insights

    Are you trying to understand how your customers are using your product, and if it brings value? => Go with Mixpanel / Amplitude / Heap with a proper implementation strategy (only measure what you need)

    My personal take is that product analytics should be a top priority, especially for an MVP.
    It allows to measure features engagement (Does it bring value? Is it too confusing? Do people understand the feature ?), but most importantly, it allows to measure current objectives: for an MVP it should be product-market fit.

    For example, I'm currently building a very opinionated habit-tracker, it's very different from the rest of the market.

    For me, it's a much better app than the rest because it helps build long-lasting habits.
    Will it be for my users? Maybe, but that's complete speculation.
    I need evidence.
    I need to measure success, a proxy to measure the product-market fit of my MVP.

    In my case, I've chosen the number of successful habits / total number of created habits.
    I have my own formula with selected parameters to determine if a habit is successful or not.
    This ratio will indicate across ALL my users, if they are successful or not using my app.
    If my app is truly bringing value to them and helps build long-lasting habits.

    With two/three other metrics related to my current goals, I can make better product decisions, because seeing the numbers and finding the causes will guide your next steps:

    "Hum most users churned and my app is not bringing value" => you look into details, which users bring the metric up, and what actions did they take.
    "Okay then, I need to promote those actions to all new users as quickly as possible"
    => it can be a better onboarding, a more intuitive UX, a fix to your main feature, a new feature, maybe you need to talk to those users to better understand the problems if it's still unclear (you have the list of them now !) ...

    You also need to look at the symptoms of users bringing the metric down and make your new users avoid those steps.

    A very simple example: you see that the most successful users used feature Y right after signing up, but the majority of users didn't used it and churned. Your hypothesis is that the feature is not accessible enough in the interface. You change the UX to make this feature easier to access, and your metric goes up the next week. Bingo

    You should build your product in terms of how much value you bring, not in terms of the number of features.
    And product analytics is a way to measure how much value you bring.

    Build (code), mesure (analytics & UR), iterate (decisions).

    1. 1

      How does GA help me understand my marketing efforts?

  15. 2

    You could use hockeystack.com without cookie consent

    At the very least, you should track simple goals and feature adoption.
    Tracking intermediate steps is optional but useful for marketing, sales, user experience etc.

    1. 1

      I can track feature adoption through user stats. the user made coupons, thumbs up. how is tracking intermediate steps useful for user experience?

      1. 1

        Time from signup to activation or signup to "value" is a big one, for example. Most people can not track that, but shortening that time is the single highest leverage thing you can do to improve trial to paid conversion rates. Also, tracking intermediate steps makes it incredibly easy to see where users get stuck. If users are stuck at a certain step of your coupon creation form, for example, you would know there is a problem there and would work on various fixes to get higher coupon creation rates. Though most analytics tools can not do this, and you are right that only using pre-made aggregate dashboards as an analytics tool is not very useful.

  16. 2

    You need it because it’s one of the few things that you can’t retroactively fix. Everyday that goes by you’re missing valuable data, and you’ll never be able to get it back.

    At the bare minimum you need to at least know page hits, and the source of your traffic.

    There will come a day where you want to change your h1 tag. If this day happens around a time that you loose traffic, you’ll be left thinking that your new copy is bad. When really you’re new copy is great, but a popular blog stopped linking to your site.

    Hotjar, MixPanel, and all of those fancy tools are probably over kill. But Google Analytics is a no brainer.

    For me I personally keep a simple spreadsheet that I update every month. It’s dead simple but provides crucial info to running my business. The columns are listed below.

    -New Visitors
    -Filled out form
    -Booked a sales call
    -Became a customer

    For each of these columns I divide it by the metric listed above, and I come up with a conversion rate. When you do this for many months the data becomes really useful.

    If new visitors is low, I know my marketing needs work.

    If filling out a form is low, my website copy needs work. OR marketing is driving bad traffic.

    Booking a sales call is dependent on my CTA language, and the copy of follow up emails.

    Becoming a customer is dependent on how well the sales rep performs.

    You probably won’t be able to make use of these metrics right away. But you will some day. And that’s why you need to know how many people are coming to your web page.

    1. 1

      I know the source of my traffic even if I don't have stats that tell me. And to be quite honest when I did know the source... in previous projects I never was able to take action.

      Seems like your product/site has a nice little engine. Market to traffic to sales to close deal.

      How has the data become useful for you after "many months" ?

  17. 1

    In the future, if you promote your product in different places around the internet and you see new users coming in - you would want to know which channel "worked". That's when having Analytics in place makes sense, as it will give you the answer.

    HotJar provides heatmaps & screencaps of your user sessions - the ethics behind this questionable, but it's really amazing and eye-opening to see where users get "stuck" on your website, so areas you want to improve on.

  18. 1

    You can try privacy-oriented analytics tools such as PoeticMetric. https://www.poeticmetric.com/ That way you won't need cookie consent and you can still make business decisions based on numbers.

  19. 1

    Hi Andrew, love your work and have learned from Better Sheets. I built BlockSurvey with a focus on privacy and didn't have analytics for a year's time. Recently started with Plausible to monitor conversion on ads published on newsletters and Reddit. Just the right amount on without being invasive help. It helps to improve the product as well. I felt I should have had it at the beginning since privacy-focused solutions are available.

  20. 1

    Andrew, you can avoid all privacy issues by doing server side log file analytics.

  21. 1

    Very simple. If your goal is to increase your traffic and certain actions done on your site (like purchasing your product), you are going to try different things to make that happen.

    How will you know if those things are successful?

    The only way to do that is to keep track of where people are coming from (to correlate it with off site actions you took), which pages they are visiting (to see how your site is structured and what pages are popular), how long they are staying (are your pages engaging?), etc.

    If you don't care about these things, then Analytics is not for you.

  22. 1

    You can't improve what you don't measure :)

  23. 1

    You can't manage what you're not measuring

    So you don't know if you're growth could be any better or not. There's five areas worth we need to understand better from day one to grow products (IMO). These all have "levers" you can pull, but not all have one direct data point, so it can be tricky but over time you can get a sense really quickly about where you're leaving money on the table.

    1. Landing page conversions
    2. Onboarding
    3. Activations
    4. Retention
    5. Pricing & value proposition

    So here's the analytics tools every startup needs from day 1 on the landing page (I suggest using GTM)

    Quantitative

    • Prisch or Plausible (privacy friendly, simpler than GA and integrate with Google Search Console)
    • Google analytics (because you will want to run ads and you need to collect data long before that)
    • Heap has a good set-and-forget system that means you won't miss anything later. Track some key features, track if people do the onboarding.

    Qualitative

    • Microsoft Clarity (it's free and I prefer it to Hotjar). It the best way to see how people use your product without actually talking to them

    Audience profiles / social:
    Defintely add pixels for Twitter, Meta and Linkedin. It helps to build an audience profile, when you advertise in future (which you most likely will) you can opt to target "lookalike" audiences. This can really help.

    An alternative here (and shameless plug) is that you could get feedback from an expert in the form of a targeted audit (free to all new clients with code "new21") or a full audit from me at the-poe-things.com :)

  24. 1

    You're missing GDPR compliant analytics. I'm the co-founder of Pirsch (https://pirsch.io) which helps you to figure out how visitors reach your website without requiring a consent banner.

    You can do a lot of optimizations from that data, you shouldn't go without it if you would like to optimize content, increase traffic, and convert visitors into customers.

  25. 1

    Even if you use Google Analytics, you might be losing out on up to 3 in 4 leads if you're not tracking your forms!

    Make sure to choose the right analytics solution to get the most out of your marketing efforts!

  26. 1

    This comment was deleted 2 years ago.

  27. 1

    This comment was deleted 4 years ago.

  28. 17

    This comment was deleted 3 years ago.

    1. 2

      Depends really. If the goal is to just survive as a Indie Hacker, you may not even care about those numbers as long as you get sales and money keeps coming in the bank. Cash is king baby. You may not care if you had increase in traffic necessarily as long as you are making enough sales to keep you going as an IndieHacker (spirit of this site, right??)

      If you however are a funded startup with VC breathing down your neck OR you have investors in general OR you have big goals to grow fast in general, you better have analytics etc to really understand your traffic/user numbers and keep it as a bible.

      1. 2

        "Cash is king" 👍

    2. 1

      Can you go, just before this basic level and explain why. I am stumped here... why should I know where people are coming to me from? I know. I talk to them. I'm not doing any SEO, I don't have any blog. I'm not doing any traditional marketing like advertising or content... yet.

      1. 3

        This comment was deleted 3 years ago.

        1. 1

          Then explain why that matters. Explain why that matters if there is no channel strategy, no growth strategy, no customer acquisition plan in place. Explain why analytics without strategy is not vanity metrics, please...

          1. 1

            This comment was deleted 3 years ago.

            1. 0

              Not necessarily.

              Analytics in the sense we are talking about is a very modern concept. Business exists since long before the advent of modern analytics. As I tried to note on another comment, analytics is just one stream of data and it is not the only one.

              Can it provide an edge, an advantage? Sure. If you know what you are doing. Otherwise, it is just a totem.

              By the way, a business can exist and even thrive without much of the above (channel strategy, growth strategy, customer acquisition plan). Look at the brick and mortar stores around where you live. These are all stuff the Silicon Valley companies invented to show their VCs that their company is valuable and on its way to billions. Most of the time, it is made up.

              I don't personally enjoy kool-aid.

              1. 1

                This comment was deleted 3 years ago.

                1. 1

                  You too, man. I enjoyed our discussion.

        2. 1

          If it's so basic, then can you give me an example of how knowing where people have come from helped you take action?

          I have stats on my gumroad page, Doesn't really drive me one way or another. I know where people are coming from I know how much they spend per view, I know how much they spend in total from that... but I can't recall an action I took. Maybe I'm being stupid here.

          1. 3

            This comment was deleted 3 years ago.

Trending on Indie Hackers
I spent $0 on marketing and got 1,200 website visitors - Here's my exact playbook User Avatar 58 comments Veo 3.1 vs Sora 2: AI Video Generation in 2025 🎬🤖 User Avatar 29 comments Codenhack Beta — Full Access + Referral User Avatar 21 comments I built eSIMKitStore — helping travelers stay online with instant QR-based eSIMs 🌍 User Avatar 20 comments 🚀 Get Your Brand Featured on FaceSeek User Avatar 18 comments Day 6 - Slow days as a solo founder User Avatar 16 comments