28
42 Comments

I got 1400% traffic, 950% sign-ups, Product Hunt #2 and only 1 SALE! What do I do now?

Hello Indie Hackers :)

I am Stan, a UX/UI designer and SaaS founder from Russia! A year ago, with my fellow @tereshchenko_kn, I started to build my first SaaS side-project — MetaSurvey. A light and fast survey tool.

We have launched two times on ProductHunt: in 2020 and in 2021. We even acquired Product #2 of the day, an enormous 1400% traffic, and a 950% Sign-Ups boost after the second launch. MetaSurvey gained some early users and sales. We are updating the Landing page based on feedback — conversion to Sign-Ups is pretty high (7,64 %)

Sadly, Pro sales are growing very slow! Besides, Pro users solve their tasks and return to Free plan or leave.

Growth advice needed!

Dear IndieHackers community — I encountered a growth/traction obstacle. I will be grateful for your help! What do you think is the growth problem? Is it a poor presentation of UVPs/USPs on Laning page or incorrect Pricing and Plans? Maybe weak pitch or low traction?

I don't want to shut down the project *( Please, help it live.


How it all started:

The idea was born in Fall 2019 when I have joined a team of 11 talented IT guys on a 2-day hackathon weekend in Saint-Petersburg, Russia. It was held by IndieHackers community leader @alexanderisora, Founder of Unicorn Platform

We validated new ideas for a hackathon and were trying to make a poll. Surprisingly, we found that most Online Survey tools have boring and bland Forms and poor mobile adaptation (the UI was often stretched from a desktop design). Also, the survey-builders Dashboards UX was often clumsy and overloaded with panels!

So we decided to build a fun alternative to SurveyMonkey and TypeForm with light and friendly Dashboard. We focused on beautiful, fun to take surveys UX, which would improve the Response Rate.

After 48 hours, an MVP was born. It had only 1 survey option— Swipe cards. But they were fun to take and cool. We created a team and, after 2 months, MetaSurvey 1.0 was built. It was ready for a ProductHunt launch!

On D-Day product got Product #2 of the day on PH, becoming viral and drawing community attention. The founders' and indie hackers' community feedback was highly positive! We thought that "fortune smiled upon" our team and the market fit is found: MVPs and ideas validation for Early Startups and Founders.

I started Twitter and IndieHackers activity, wrote posts did networking. Soon enough, we got early users, sales, and feedback. In the first month after launch on PH, our product acquired 3 Pro users and dozens of Free!

We were celebrating. But the true D-Day was ahead...

12-months of development

Soon after PH launch, 9 of 11 founders left for their projects or lack of time — only 2 makers left: @standiz and @tereshchenko_kn. We decided to "take a chance" and continued building and improving MetaSurvey. We loved it. We were adding features (mostly not requested) and fixing UX/UI or core bugs. We took one day off on Sunday.

For 12 months, we worked in "heads-down mode" 4-6 hours a day, before and after our desk jobs. We even had to change our daily regimen. My co-founder was waking up at 6AM — and working before his desk job. I was building UI till 3AM after my main job. We had no personal life.

We were obsessed with building a unique and cool Survey creator with super higher Response Rates (RR)! A tool that would help founders to validate their ideas, saving time. We also wanted to re-invent a feedback gathering via fun Tinder-like and gamified UX.

Yes, we used our surveys to improve MetaSurvey! We created 2 surveys and collected over 150+ responses. The Response Rate is 88%.

We found that the Online Survey market was full of competitors. We analyzed 45 competitors in detail — only 7% of them had a user-friendly builder UI. Mostly, people hated taking and creating forms.

Core values /USPs:

  • Focus on simple and "human-friendly" UI
  • Hand-picked the most fun to take questions, up to 95% RR
  • Survey Builder UX was made clean — to create surveys in minutes
  • Surveys were specially adapted for mobiles
  • 3-5 min to create
  • 15-20 sec to take a survey

Launched v2.0 on ProductHunt

On 1-st of May 2021, after spending 1600+ hours, we finally released v2.0 and launched it on PH. I wrote an article about the launch on IH. We were dreaming of boosted up visitors, high conversions, and hundreds of sales!

Actually, we got stats even better than expected: 1400% visitors, 950% Sign-ups in the first 3-4 days after PH launch, and 1300% Goals reached on our Landing! That was the highest metrics spike since the product launch in 2019. We got 700+ visitors and 55 Sign-ups, more than for the past 5 months!

The sad part — we got only 2 sales after the v2 launch (1 of which was a random Free user that converted to Pro after an E-mail with a discount). Although we had 0 users from the first PH launch ;)

The insight followed: something is totally wrong with the USP, Market fit, or the Pricing.

Our MRR started to go up after the v 2.0 launch, but the growth is slow! The conversion from Sign-Ups to Pro is 5,2% [MRR]
MRR

Where to go next?

Dear IndieHackers, I will be grateful if you advise. What is a growth problem? Is it a poor presentation of UVPs/USPs on Laning page or incorrect Pricing and Plans? Maybe weak pitch or low traction?

Sources to check:


Thanks for reading!

Which Growth Hacking helped to BOOSTyour Starup?
  1. Launch on PH / Appsumo / AngelList
  2. Reviews on G2-like / AlternativeTo
  3. Roast my Landing page-like
  4. Boosting Twitter / Reddit
  5. Boosting LinkedIn
  6. Applying to YC
  7. Goggle-ads tuned
  8. E-mail marketing tuned
  9. LTD / Referral platforms
  10. Discounts and Promo-codes
Vote
  1. 10

    You have an intriguing product, but I don't get the sense that you understand who your target customer is yet.

    Your target customer is not "anyone who wants an easy to use survey" (that's simply a restatement of the product itself). You need to be able to explain who your target customer is and what problem they have without describing the product.

    For example, I also offer a survey-type product, but it is offered as solution to a specific problem in a very specific niche -- self-serve SaaS companies who are trying to improve new user onboarding during free trials.

    Your main unique product feature seems to be the Tinder-like swipable answer format. It is a nice "cool" feature, but it is obviously not much use if I need multiple choice responses or need someone to fill in their name and address. So I would try to identify a niche customer/problem that would places a high value on this unique feature, and focus your efforts there.

    If the benefit of this feature is that it improves completion rates, then you might find a hot niche in online marketers who use survey funnels on their landing pages for lead gen, and who really care about conversion rates.

    Or maybe retailers who want to survey past customers to find out which products to offer.

    But your general survey user - e.g. somebody who wants to survey their wedding guests to find out whether they want fish or steak - is not likely to care about their completion rates enough to highly value this feature.

    On the other hand, if you are committed to a broad approach, then you are most likey going to be competing on price - e.g. "Typeform but cheaper."

    If that is the case, then you should be going directly after people who are already paying $35-$70 for Typeform/SurveyMonkey and trying to get them to switch to your lower priced offering.

    1. 1

      Thank you for the feedback, Mo! We changed offers and USP on the Landing dozens of times! We thought that our customers are Indiehackers and Startups, but we were wrong... They are super-curious but have Low conversion to Paying.

      Our main USP is: now taking a survey is super-fun and not boring. Which results in higher than average Response Rates >80% (based on 10KK of surveys)

      ... Tinder-like swipeable answer format. It is a nice "cool" feature, but it is not much use if I need multiple choice

      Wrong! We have 7 different question types - including Multiple answers and NPS!

      Now we are searching for our market fit. Thanks for great insights!

  2. 5

    Hi Stan, congrats of the PH launch success, don’t be disheartened by the lack of take up. That volume of traffic interest suggests that with some tweaks you may be able to really grow this.

    Ok, feedback:

    Main points:
    #1 Your free plan is the key to future growth. It’s a way for people to get used to your platform, when they need to scale they will choose you. Right now, your free plan isn’t nailed, nor are the things you scale up across memberships.

    #2 optimise your landing page for free account signups. To me that means editing your landing page down by 30-40%. Improve the grammar on the page. Talk more about benefits for users over listing features.

    #3 offer unlimited surveys on the free plan. Keep the limit of 10 questions and 100 people, maybe push it up to 150 people so it’s more than Typeform offer. This alone will make people sign up. No one will ever make so many small surveys that you could go out of business. You could clear up surveys older than 3 months on the free plan, if you need to keep storage costs down?!

    #4 introduce a pay-as-you-go mechanism. People might want to pay for an additional 500 people for one survey.. set up a charge per X responses. Your customers can then compare paying that as a one off vs upgrading their plan longer term.

    #5 offer yearly and monthly pricing. Give a discount for buying yearly.

    Landing page:

    • the grammar isn’t correct in the first sentence. I’m learning a second language right now, so I respect that English may not be your first language and how hard it can be to nail the nuances. Get help with this, it can be a big turn off for people.

    • Header describes the product and what it does. You should be speaking to the benefit / the need it meets for the user.

    Ok.. so, you’ve focused on time- 10 second surveys. Is that what matters to people? I’ll hypothesise a few compelling headers that might speak to your market’s needs as food for thought:

    • higher response rates than Typeform
    • create high performing surveys in minutes that your users will love
    • join 600 companies and make higher response surveys in minutes
    • optimise your surveys with actionable stats and insights

    Back to landing page feedback:

    • I would move the block about the 600 companies using it to be the second block on the page. This is very compelling.

    • remove the words “fun” and “cool”. It makes it sound like a toy, reducing the sense of quality and an impression of longevity. Don’t go too far the other way and make it too corporate sounding either.

    ### Some reasons people may not have signed up:

    1. Most people only need a survey when they need one. Those people won’t pay a membership “just in case”.

    2. For the people who do always need a survey tool ready, you’re pushing against how invested they are with that tool. I suggest you put up some content marketing (you can hire someone to write these cheaply) about MetaSuvey vs Typeform - do it for each one. Ghost.io has some good examples of this content since they launched their memberships features.

    3. Your free plan says “for personal use”. Why does it matter if it’s business or personal? You just want someone to use your product to try it out. Better that they ARE a business! The free plan on Typeform is just a small plan, limited in the number of responses and surveys I can take. You haven’t said “you can use this for business” but it’s implied by calling it personal.

    4. I think you’re making a big mistake by limiting people to 3 surveys. Typeform offers unlimited surveys in their free plan, but only 10 questions per survey and 100 responses per month. I’m a Typeform customer because of this. I use it for small things, I know it well now. When I need to scale to a bigger audience, i will go with what I know - Typeform!

    You’re listing too many features eg on the analytics section.. people don’t want to “track survey statistics and comments”- that’s not getting a job done! Go to the next level- tell people what they will do better or how they will be better because of the stats and comments :) Nikon doesn’t make better cameras, they make people better photographers!

    There is a lot of content on your landing page. I don’t feel like it’s in the right order either. Suggestion: print out your website on paper, cut them up and lay them out on a table/floor. Then say out loud what customer you are and what you’re looking and narrate your journey out loud to yourself / your co-founder. Record it on your iPhone perhaps to play back so you’re free to “think out load”.

    Doing this on paper will help move things around. Take things out, play with it.

    • save some landing page content for a separate page. Let people “learn more” if they want to, this will help cut down that content.

    • the landing page doesn’t feel like a sales pitch, it feels like a list of nice features. I don’t know how to compare it to Typeform for example. You should make that easy for me.

    I really think you have something. It’s definitely time to step away from your keyboards. Get into your customer’s heads and really look at your site. Take turns being a type of user, talk it through out loud- role play it. Then go talk to your customers and find out how well you know them.

    Definitely A/B test any major changes you make from here on if possible.

    Good luck, congrats. I hope this helps. Happy to talk more if it helps you.

    1. 1

      Hey Paul! That is the most outstanding and complete feedback I have ever seen!

      I will implement 98% of these tips! Moreover, we are holding a MetaSurvey team meetup now, concerning all comments from the IH community! (and of course, we will drink wine)

      #1 We will unlock all features for all plans. Also, we will limit Pricing plans to the number of responses/questions per survey. We are thinking about limiting Free to 150 responses total / 10 questions per survey. Branding will stay only on the Free plan. Tier 1 and Tier 2 will have no branding.

      #2 We will revise all Landing page copy texts and clean those logical and grammar mistakes. We already connected with a native-English bug fellow from Finland and asked him to review all Landing copy text. We will talk more about benefits for users.

      #3 Again, we will revise a Free plan. Mb removing the number of surveys number limit. We will try to adapt the Free plan to beat Typeform`s limits. We are thinking of limiting the Free plan to 150 responses total and 10 questions / per survey. Clearing up surveys older than 3 months is a good idea too.

      #4 We will introduce a pay-as-you-go mechanism. but in long-term feature-request — it is time-costly to build. We don`t treat this as a high-priority feature, although we were thinking about releasing it in future MetaSurvey updates. Like get +500 responses for $1.20, get +1500 responses for $2.80 etc.

      Go to the next level- tell people what they will do better or how they will be better because of the stats and comments :)

      We will re-write copy and headings to say people what they will get using MetaSurvey tools!

      There is a lot of content on your landing page. I don’t feel like it’s in the right order either. Suggestion: print out your website on paper, cut them up, and lay them out on a table/floor.

      Maybe we will do that. We will make content more user- and problem solving-oriented. Printing Landing on paper seems to be a great idea!

      Thank you Paul for this amazing, insightful feedback! I believe that implementing all of the ideas will lead us to $250 MRR and more!

      1. 1

        I'm happy you found the feedback helpful. Let me know when your next big updates are shipped, I'd really like to see how you progress.

        I believe you have something and you can grow this considerably. Your passion is inspiring, so thank you for sharing your product and story :)

  3. 5

    hi stan your product have huge potential. dont be dissapointed of ph launch. actually your audience resides outside PH. most of the folks on ph are there to check out new stuff & not necesarily buy it. i am seeing more & more people creating google forms during this pandemic.

    1. 4

      improve your pricing model. it looks overcomplicated complicated. give all features in free plan. In early stage try to grow it as freemium. look at pricing plan of google's firebase.

      1. 3

        I basically agree with alok - give all the features at all the levels but just limit the levels by the # of surveys and questions - rename the levels as starter, pro, and maybe Pro plus (get rid of bot and droid because you are already having to explain those names by putting (pro) next to it anyway) this cuts the pricing matrix down sharply and underneath it say all the features everyone gets.

        1. 1

          We have renamed the Plans as you suggested. Also, we think about making all features available for all plans, (except branding removal on the Free tier).

      2. 1

        Thanks for feedback about Pricing model! We are reworking this page for constantlyand I thought it was clear(

        Can you please tell why do you find it compliceted?

      3. 1

        This comment was deleted a month ago.

    2. 1

      That`s a great tip, Alok! Only after the second launch on PH, we found out, that this community is curious about everything new! That is why Sign-ups are growing without Sales)

  4. 4

    Hi Stan,
    I'm originally from Russia/SPb too :-))
    As for your product... I have a feeling you do not understand real problems of your customers and probably didn't specify your TA clearly.

    Surprisingly, we found that most Online Survey tools have boring and bland Forms and poor mobile adaptation (the UI was often stretched from a desktop design). Also, the survey-builders Dashboards UX was often clumsy and overloaded with panels!

    But does it really matter? I conducted surveys several times and always used Google forms - it's not perfect, the UI may look clumsy too but I've never cared. Now, I even can use Twitter polls and even it allows to add only 4 options and don't have any UI at all, it works in many cases.

    To my, your UI looks even a bit complicated to me. And this feature can't make me pay because I don't have any problem with UI.

    I'd suggest you to do 2 things:

    1). Try to find a niche. It could be very narrow and specific (like airspace) but if they use surveys it's okay. You'd create a specific landing page for this niche - it may be a simple subpage - but it'd address their pains and expectations not everyone's.

    1. After you figure out which niche you would focus on first, conduct interviews. Not a survey, not a poll, not a question on forum but live video-interviews. This website https://zamesin.me/books/custdev/intro is super-helpful (I also recommend to watch his videos on Youtube), I used it for my own interviews (read here how I did it: https://seacat.blog/how-im-looking-for-a-good-idea/). So, the point of those interviews is to find a real pain point and sometimes you find that there is no pain point all (what is hopefully not your case).

    That's what I think, good luck!

  5. 4

    The product looks very nice! I know how it feels like when you don’t get all those customers you were expecting.

    I wonder how often you look for websites that are already using any form from your competitors. You could do the experiment of recreating the form with MetaSurvey and see how better the final result is. If you get a better result you could just contact them, showcasing the new form, with your price.

    I know it would be an “unsolicited email” but it could work.

    1. 1

      We are glad you like the product!

      That's a great idea to look for dull, bad-UI forms and re-create them with MetaSurvey, then sending the upgraded forms to site/e-shop owners!

  6. 3

    Learn how your competitors are getting customers and try that. I would also try to focus more on SEO, your current website is lacking good SEO content.

    1. 1

      Most competitors use Google ads. We launched our first 2-week campaign, spent 200$, and got 1 customer*( All competitors have a very poor presence on IH, PH, Angelist. Anyway - they are making money and selling Pro plans.

      1. 2

        IH, PH, Angelist are just websites, you don't need to be present on those websites to make a business. Most online businesses are not on those sites.

    2. 1

      Thanks for the advice! We have launched a Blog recently, with articles on how to improve response rates and make great surveys. Isn`t that SEO?

      1. 2

        it's SEO, but SEO takes time and you need hundreds if not thousands of blog posts/landing pages and at least 1 year to start getting good traffic.

  7. 3

    Haven't had a chance to read through the whole blog, but I can say this, PH !== Sign ups

    PH is a lot of people who are super curious so they will sign up to try but from exp CX is V low

    However it works v well for products

    1. 1

      Thanks for your advice Vaibhav! We had a huge 1400% traffic spike from PH! So we were waiting for at least 5-10 Pro sales from it... *(

      Yep, as you said, we got around 80 Sign-Ups total from curious people, but they didn`t convert to sales. Although, we got tons of feedback!

  8. 3

    Hey, I'm not sure what to do in your situation, but maybe it worth a read https://www.indiehackers.com/post/i-built-a-50k-following-in-9-months-and-it-changed-my-game-completely-this-is-how-i-did-it-7e51c338a0

    Though, this guy made an app for Twitter and grow audience on Twitter. But if you had 40K followers, than you'd probably get something out of it

    But also, there is always a classical way of using paid ads. Reddit/Google ads/Facebook ads etc. So you can spend some amount of money for each channel, and check the conversion. If something worked well, then you're good.

    1. 1

      Thanks for the advice, Viktor! Image FeedHive was successful because his target audience was ALREADY in the channel he was boosting. (Twitter) Everybody in wants to raise Followers, so it`s a win-win strategy)

      We tried to use Google Ads, but we have little knowledge about campaign optimization! We got only 1 sale after spending around 150$ for 2 week Ads.

  9. 2

    Be sure to read @typeofgraphic's reply.

    Based on it, you could narrow down your USP:
    "Get more surveys filled" / "Get most surveys filled" / "Get 88% surveys filled"

    That's actually a pain point of survey creators.
    It doesn't matter how many surveys are sent
    if nobody fills 'em.

    ... or maybe reversed - it's pain point of end users.
    Many times when i've seen a survey and just clicked away.
    Even if i really wanted to help.
    They are just too boring to do.

    Also, don't compare your product to Typeform or others, your clients shouldn't know
    who they are.

    1. 2

      Thank you for your feedback Marky! I wrote to @typeofgraphic's a huge reply and with the co-founder, we are completely remaking and narrowing the USP and Landing page copy.

      "Get more surveys filled" / "Get most surveys filled" / "Get 88% surveys filled"

      We are thinking in that direction. We are also narrowing our audience from everybody to e-commerce, e-shops, and market researchers.

      Not being boring and high response rate — are the main 2 USPs we are trying to project to people!

  10. 2

    Now go to that 1 sale and ask why him/her bought your product.

    Perhaps Growth isn’t your priority now. You should figure out why people bounce, you get a lot of traffic doesn’t mean people are interested in your product. Maybe there’re just curious, don’t obsess with vanity metrics.

    1. 1

      Thanks! We also have previous Paying customers — there are 10 in total. Now my co-founder and I are trying to connect and CustDev those Pro users.

      I agree that counting just total traffic is nothing. You need quality traffic of your potential customers and great Conversion from Sign-ups to Paying to win!

  11. 2

    The UX of the tinder-like survey is not great. When you swipe, the UI flickers and it takes time to load the next picture.

    1. 1

      Thanks for report! We know about this UX problem. Will fix the Swipe card glitch in the nearest update.

  12. 2

    This is not advice, but more of a request. What is the product you are using in that screenshot to track retention, signups, etc.?

    1. 2

      Hello. It is a Russian solution called Yandex.Metrika - they have EN version. It has light and slick UI/UX:
      https://metrika.yandex.ru/

  13. 2

    The site looks great and getting #2 on PH will result in solid SEO since you get a follow link.

    As others said, PH might not be your target audience, and same with HackerNews which is known for not being great for paid conversions.

    Your pricing page has some good "conversion encouragements" such as branding. However, I would consider removing the branding for the Pro tier - people generally don't like to pay and still have branding.

    If you really want to try to drive traffic to paid, lower the Free to 1 survey and 5 questions and make pro $4.99. If you see paid on pro then can experiment with higher pricing.

    Also, should cap JPEG and PNG.

    1. 1

      Thank you for appreciation) We put numerous hours in polishing Landing! You`re 100% right. With PH we made good traffic and solid SEO. Organic traffic from Google/Baidu is constantly increasing!

      The conversion to Sign-ups is 7,64%m to Sales — 5,2% But the sales are going super-slow and they are mostly random. We will think abour removing branding on tier 2 plan.

  14. 2

    Hi Stan! I'm your neighbour from Belarus) First - keep pushing. Your project is awesome.

    Look, I have an idea to raise your prices. Focus on b2b. Your rivals GetResponse for $15 bucks gives surveys, email marketing, online chats, landing pages, automations... sick

    Take a look at Ukrainian startup that now dominate the world of prospecting >> https://snov.io?_get=leads

    Use it to find marketing decision makers. Sell big via zoom. Otherwise I can't the the way to beat your competition

    1. 2

      Thanks for your support, Artur! We will keep pushing and will not step back!

      GetResponse is not our main competitor - It`s more like a marketing platform. It offers many features for 15$. But it can not create cool 10-sec surveys with high Response Rates of 86% avg.

      I Will surely check https://snov.io/ - it looks like a great Growth hacking tool!

      1. 2

        Your response rates are impressive. Wish you dominate the market.

        The most complex task right now is to break through the noise. Email for me is like to knock-knock the door and look into the office room to the prospect.

        Treat like a chess game. Don't put all your seven moves in one email.

        If you focus on B2B test higher checks. 2 options: Freemium and $97 for example.

        Seems like you already know what to do, so keep it up!

  15. 2

    I think some of your problem is in the very small details of your presentation.
    there are many small content writing errors and issues that begin to collect in my mind as I read them and they are eating away at trust. You need someone who is a native English-speaking content writer to go through it and refine the text - simple example: the part of the main page that starts with the header "Meta Survey panel2" I don't know why that's a header? I wondered where panel1 was, the line below it says "Lack of complex logic and settings. Start for free!" sounds like something is missing rather than a benefit and the three sample questions show on the page all have minor grammar issues.

    1. 1

      Thanks for the tips mate! We fixed those logical and content errors that you have reported. We are grateful for your time and report!

      Tomorrow we will send all Landing page "copy" and headers to our fellow native speaker - he is from Great Britain.

      P.s. Is Product #2 of the dau on PH improving your trust?)

      1. 2

        haha - being #2 on PH is a very good thing! - every single bit of feedback I give anyone, especially on here, is because I want them to have the maximum amount of success! I actually have to resist going into too much detail because I am working on my own project, like 3 side hustles and a day job !!

  16. 2

    Great article! And well done on recognising you need to pause and take a closer look at things.

    I've been searching for a suitable, attractive, and not ridiculously expensive survey tool for Libreture for months! And I think I'll at least try out your free survey asap.

    Your basic plan price is spot-on for peaople like myself (not the PH crowd, as @veebuv has said). I would also agree with @alok about the features - standardise the features across all plans, and use the Number of Responses as the pricing differentiator/tiers.

    Number of responses is the feature that costs you money, yes? Customers may already suspect that, so may not appreciate having to pay for redirects, adding their logos, etc. Look at how Plausible have split their pricing plans: page views. If you get more hits, you pay more - which ties into the idea that Plausble have to record more data, and if your site is successful, you may be happy (and more able) to pay more.

    I know that I wouldn't mind having a 'Powered by MetaSurvey' logo visible, if I could also brand it for Libreture and redirect respondents to my own Thank You page afterwards.

    Am definitely going to build and launch a survey with MetaSurvey, and would certainly consider a paid plan if I hit the respondents ceiling.

    Great work,

    Kevin

    1. 2

      Hey Kevin! Thanks for your time. Such huge community support is the most valuable for us right now.

      We are thinking about re-working all Plans and — making all features available for all plans, (except branding removal on the Free tier). We will use the Number of Responses / Surveys / Questions as the pricing differentiator for tiers

      1. 2

        You're welcome, and great to hear!

        Been testing MetaSurvey already and liking it a lot. Thanks for building a much-needed product.

Trending on Indie Hackers
Is my Landing page too plain? 18 comments Page 404. Why do we give a F$%k about details in design of our products? 10 comments I analyzed >1,000 indie SaaS projects and bought 6 for over $15M in total! AMA 7 comments Roast my idea: Partnerships (and cross-promotions) for SaaS Founders 6 comments Use cases of the Social Token Revolution 🔥 2 comments Who here use prediction markets, and for what? (linking PolyMarket, the platform with the highest volume) 1 comment