Magic
Simplify Google Analytics.
I was pretty naive...
I posted a milestone earlier about how I was conducting user research. I've changed things up since then, and learned a small lesson, thanks to the feedback from this community!
BEFORE:
My waitlist signup form looked like:
- What's your email?
- What problem are you trying to solve with our product?
And then I would personally hand-craft an email to everyone who responded, with a pretty low response rate.
AFTER:
Now, the form looks like:
- What's your email?
- Do you use or pay for an analytics product/service outside of Google Analytics? (If so, which one?)
- Briefly describe your use case. (e.g. "I run an e-commerce store with 3 employees and I'd like to track metrics easier")
- Would you be interested in chatting about the analytics problems you're facing?
LESSON: Ask questions upfront
I realize now, asking more questions upfront during signup is much MUCH better than deferring it to personal email afterwards.
I originally thought "I hate long forms, so why force friction on new potential users? I'll just personally email everyone, and they will appreciate the personal feel to the product."
Sounds nice in theory, but it's actually best to just make the upfront.
Why?
Because they MOST feel the problem that you're trying to solve when they're signing up. They were reminded of the problem so strongly that they decided to fill out your form. Thus they're most likely to give you precious datapoints IN THAT EXACT MOMENT.
You spent so long crafting your landing page to point them towards the thorn in their side, you have to strike while that iron is hot!
When they check their email later that evening, they might have completely forgotten about that problem, AND your solution.
And yes, you might think, "I'm missing out on all those potential users that want to put an email quickly and leave! There might be people who quit the form halfway!"
And my response would be that especially early on, you aren't looking for those kinds of users. You're looking for the people with the most pain. They're the ones who will provide the most valuable feedback and not give you half-hearted compliments.
Hell, you're trying to get paid!
If a user isn't even willing to fill out a few questions on a form, there's no chance in hell they'd fill out your Stripe checkout!
Thought this was gonna be a small post... but rant it is :)
TLDR; I got 10 waitlist signups yesterday, personally emailed 8 of them, and only got 1 response so far. Would appreciate some thoughts about my process of validating this project and doing user research..
After my app got approved by Google, and users could safely use the live demo of my app, I posted here on IH and on Reddit.
My landing page has a link to a waitlist sign up, which asks two questions:
-
What's your email?
-
What problem are you trying to solve with our product?
Then I get notified when I get responses, and then I handcraft an email to everyone who signs up.
The email looks like:
Header: "magic: thanks for the interest!"
Body:
"""
Hello friend!
Jonathan here, building magic. Thanks for the interest in our simpler GA dashboard!
You mentioned your problem "getting simple analytics for less tech-savvy managers".
Does our live demo solve your problem to some extent?
Cheers!
Jonathan
"""
My goal was to make a small ask, for feedback on our live demo, to kick off a conversation, so that I could get to know the person and their needs better.
I felt like asking them for a 15 minute call was too much up front.
I got 10 signups. I've sent emails to 6 so far, and only got 1 response so far.
Although it's been less than a day, I am worried that I am doing this process of user research or problem validation wrong....
Am I?? Or am I just paranoid haha
I even just read "The Mom Test", so I feel like I should be better at this now 😂
-
8
Hey Jonathan 👋
Firstly, I really like your landing page. Its clean, simple (so reflects the nature of your product) and I totally got what it is your product does. And the live demo – what a great way to show what it is your product does 👌.
As to your question. Give it time. As you’ve mentioned, it’s been less than a day. See how things are after a week and then decide whether you need to change your approach.
Here are a few thoughts I did have about your approach so far:
-
I like that you’re asking about the user problem in the sign up. It’s a good way to get a high level idea of the problems your users have.
-
Your follow up question (“Does our live demo solve your problem…”) is a closed question, so this may limit the amount of feedback you get. It's also focused more on your solution, not their problem.
-
I think you’d get a lot more from a 15 minute chat than an email response and I don’t think it’s too much to ask up-front. Afterall, you're trying to help them. The worst they can do is say no. You could include a follow-up question in your typeform to see if they'd be interested in having a chat. I’d make sure you phrase it so it focuses on them and their problem though.
I'd be interested to know how it all goes, so definitely post an update!
-
1
Just thought about this:
If they're not really willing to respond to my emails, the problem that they're having isn't urgent enough, and maybe my product isn't solving an important enough product? That is a form of validation itself?
-
1
Thanks for the feedback! I'll be sure to post an update later.
Re #2, I was also worried about making it a closed question, but I really wanted them to try the live demo first, and it gives them something concrete to go off of, easier to respond to. You're right though, it does focus on my solution more than their problem, so maybe it deserves some editing.
Re #3, Yeah that makes sense. I guess I'm just thinking about it from my perspective, if I was a customer, I'd rather have a few email exchanges before I hop on a call.
-
-
6
Hey Jonathan, 1 out of 10 is normal, it's hard to get people interested.
I found another method that works better: find people on IH who are in the market for your product, and reach out to them directly (email or twitter is available on their profile). Then offer them to do a 30 minute call where you give them 15min feedback on their product, and they give you 15min on yours.
-
4
Yeah, I did a bit of outbound, but definitely could do more! This idea of exchanging 1on1 feedback is actually http://pheedback.co/! Excited to see how @janel and @pabloheredia are going to build it out!
-
1
looks great, I signed up. Thanks for the heads up.
-
-
-
3
If I got an email like that from something I signed up for, I'd likely also not reply, because by the time I got the email, I probably won't remember what was in the demo. If I even looked at the demo at all. (May be I just skimmed the landing page and thought some part of it was interesting enough to sign up).
I think a better email would be, to highlight some specific part of your product, e.g 'You entered the problem X. I think our product can do Y, Z, and A that can help you solve that: <link to screenshot or video demo>. What are your thoughts?'
That reduces the effort on the user's part quite a lot, so they'd be more likely to reply.
Note: This is just my intuition, I have yet to email my waiting list, so I'm not speaking from experience :P.
-
2
Ahh that is actually a good point I had not considered.
I had assumed that people that would put in the effort to sign up would have at least tried the demo, but the way you phrase it makes sense.
I think another point might be, if they're not really willing to respond to my emails, the problem that they're having isn't urgent enough, and maybe my product isn't solving an important enough product? That is a form of validation itself?
-
2
Yeah that could also be a reason for sure. Or may be something was a need at one point but it no longer is. It could also be that you were asking them to make too much effort (review demo, think about it intelligently, write an email) and haven't yet demonstrated that you could solve their need.
If I were you I'd wait a few days, then send a personalized email with the template I posted in my last message. If they still don't reply, that'd be a better indicator of how strong the need is, since you're eliminating the variable of 'this needs too much effort'.
-
3
Sounds like a plan. Thanks so much for thinking along with me.
-
-
-
-
2
This is a totally normal conversion rate for an email like that. Good even. Keep going.
-
2
You should find a different name, it's just not right calling a minimal dashboard design for google analytics "magic".
-
2
Well you’re one to be talking about names..
And yes, if my plan was just to stop at the dashboard design, I would have copped the domain “simple-ga-dashboard” and called it a day. But I am considering building more integrations (stripe, mail chimp) and features like easy email reports. I’d rather go with a name that conveys ease/simplicity.
-
-
2
Hey @jonathancai
It think you are missing out on an opportunity in your typeform to get an understanding of whether they would like to follow up with you.
The user maybe thinking that this is form is enough feedback and any
maybe...
- What's your email?
- What problem are you trying to solve with our product?
- Would you be interested in talking further etc (checkbox)
That way you get an understanding of how open they are to talk further. Also there may be an opportunity to upgrade to a call because you know they are open to talking further.
Also somethings to think about
- Maybe they didn't do the demo yet
- Is there anyway you can tell if they opened the email (open rate etc)
Just some thoughts hope you find them useful. I tried out the live demo and it seems like a good concept, especially for anyone that wants to view stats on their sites at a glance.
-
2
Thanks for the input!!
Yeah, I think based on yours and other comments, I will edit my typeform to include a question that gauges this.
Although I'm not keen on tracking email opens. I'd rather just send it without worrying about that.
-
1
I think that’s pretty normal. That being said, you can tweak your email a bit and that may slightly improve the response rate:
- Your email totally looks automated. A lot of people I’ve spoken with almost never respond to automated emails. So I try to make them feel more personalized. For example, I include their first name in the subject and first line, and I tweak the wording of the use case they entered a bit.
- Lots of people prefer a short call over answering emails. Add a 15min calendly link at the end and say sth like “In case you prefer a quick 15 call over emails, you can select a time that works for you here”
- I usually end it with mentioning the value both of us will get out of it: “We’re currently deciding on what features to build over the next months and your input would help us make sure we’re building what you need. “
Keep it up 👍
-
1
I did answer, which means the email was timely and I knew what to answer!
-
1
I did answer. Which means I saw nothing wrong here :)
-
1
Jonathan, Your response e-mail simply states the obvious. Yes you are just repeating what they already know. Also you are assuming they did live demo. Something like thanks for signing up. (hint they are not looking for magic!) Does it look like the app can solve the problem of simplifying analytics? Are you clear on how this app will save you time and make analytics a much less stressful task? If so great. If not would you like to have a conversation regarding how to use the app to get the results you truly desire? Schedule button
-
1
Nice website with the slide! Neat.
-
It is quite normal, it has only been a day + it is weekend, lots of people have overwhelming inbox to take care of. But actually if anything, there is always a conversion rate. If you get 10 signups, you can expect that 6-7 people will never respond to you. That's just how things work, each step is a friction.
-
I think you asking the "did our solution solve your problem" question is not a good opener. If your goal for these people is user research, you can ask them for a 30-min call right after they give you their email. You'll be surprised how many people are so nice to give you time if you show A) clear intent, what are you trying to learn? B) clear ask, how much time you need from them, what the call will be like. I wrote an article about user interviews a few weeks ago, hope it is helpful in some ways.
https://kevoncheung.com/i-discovered-these-customer-interviews-traps-first-hand
Btw who is your target audience?
-
Happy to say Google finally verified my app!
So anyone can safely access the live demo without seeing a sketchy consent screen.
My app (magic-ga.com) is a simplified Google Analytics dashboard.
It aggregates all your GA-tracked websites into a single dashboard so you can see the most important metrics at a glance and quickly.
(10x faster than going to analytics.google.com and tabbing through your properties imo)
You can use it for free and without an account!!
(no data is being stored anywhere, I don't even have a db/backend setup yet haha!)
For a quick preview of what it looks like:
I'd really appreciate feedback/thoughts!
-
2
Congrats @jonathancai on the approval.
It's interesting you say no data isn't being stored since you don't have a db/backend. You could re-phrase it and use it on your website. People love transparency and this is one of things that made me think "ah that's very nice of Jonathan to be transparent and not to collect data" . People really appreciate it.
-
1
Thanks!
And yep! It's written in the live demo page.
Definitely wanted to be transparent :)
-
1
Ah I see. My bad I missed it. I feel when we are transparent with users, they feel they respected and have the power. In most cases when we openly ask for the details they are more than happy to provide it too.
-
-
-
2
Hey Jonathan,
This is great! I really like the clean design :)
This hits home as recently I've been spending time at the day job putting together customized dashboards using Google Data Studio. Have you checked it out before? How does Magic differ? With GDS I'm able to pull data from any source and aggregate it alongside GA which is a big pro for us.
Thanks!
-
1
Yeah, so before building this, I actually used GDS to do this exact thing! I embedded my GDS dashboard on my own personal website lol!
But personally, I found GDS a bit complicated and tricky to set up for 5 data sources/charts.
Right now, Magic just differs in one-click setup. Simple and out-of-box.
But of course, I can understand us hacker types might also prefer more customizable data-viz solutions like GDS :)
-
-
2
this seems like a really cool idea! Hope it goes well
-
1
Thanks! Just not sure if people actually need something like this haha
-
2
Seems plausible and I'm sure you'll find out when you get it in front of people! Did you make it to scratch your own itch? Also I wouldn't be shy about charging $!
-
1
100% to scratch my own itch. Had so many dumb little projects to track through GA and it was annoying haha.
To be honest I don’t believe this is anything valuable enough to charge $ for... especially since the main reason (I think) people would prefer ga is that it’s free.
I’m considering just stopping here, replacing the landing page with the live demo itself, and saying it’s just a “free product”.
I’d say features like email reporting should definitely be paid though.
-
2
hey @jonathancai, just read this article and was thinking of you: https://training.kalzumeus.com/newsletters/archive/saas_pricing It talks about how technical people often undervalue their product because it seems easy to make, and even gives an example of someone trying to compete against google by making their product free.
I think it might be helpful to change your notion of what "valuable" is here. It may not seem valuable to you, but to someone busy at a company or startup, their time is extremely valuable. So being able to see their entire web traffic in an easy format could provide a heck of a lot of value.
-
1
Ahhh good ol' "Charge more!" patio11
Definitely can be difficult for more technical people to understand business value.
Thanks @Glench for thinking of me and sharing!
-
-
-
-
-
-
2
I like the direction you're going. Personally I really dislike the complexity of Google Analytics and also of the current Google Ads. Would love to connect :)
-
1
Thanks! I’ve never used google ads but I can imagine the same bloated complexity might be there too haha :)
-
-
1
Hey, I using YouTube data API in my app and each time I record a video, google is asking me to submit again huh! Can you guide me on how to create the video please. TIA
-
1
If they are rejecting your submission, they will leave the reasoning why, e.g. you didn't show the client id, or you didn't display the url/domain in the video.
Did they respond with such reasoning for you?
-
-
1
Hi Jonathan. I remember when I got that approval from Google regarding my app and man, that feeling was glorious. Congrats on getting things approved and hoping that whatever limit they gave you gets filled up quickly and you're having to go back and ask for more.
-
2
Haha!
I've gotten approved twice, and the first time took longer than a month! Happy this one took just a few days, its definitely a good feeling :)
-
-
3
This comment was deleted 5 years ago
-
2
hey I think since the author talks about Tailwind it's likely they're using Tailwind UI components which are shown here: https://tailwindui.com/components/marketing/sections/heroes
-
1
Yeah, I certainly took a lot of inspiration from Plausible Analytics, their interface is amazing!
Like @Glench mentioned, I just began using tailwind ui (free components) and tailwind css, so atleast the marketing stuff will look a lot like many many other saas companies that are also using tailwind ui free stuff. Obviously not great being generic, but I thought it fit the aesthetic I was going for
-
1
This comment was deleted 5 years ago
-
-
Just came on here to give @adamwathan some praise for one hell of a framework that I was naive enough not to use until now.
TailwindCSS + TailwindUI (free components) have literally 10x'd my frontend work productivity.
Built the landing page in lightning speed.
Plus, if you want to use TailwindUI with React, I'd recommend this chrome extension to convert the HTML to JSX.
I could not recommend Tailwind enough. Stop using vanilla HTML + CSS.
-
2
Hey Jonathan, I came across your post when trying to find folks working with TailwindCSS. Glad to see others enjoying the framework as much as me. I'm currently building www.tailwindflow.io and wanted to see what you think of the tool.
Also I tried using Magic but got stuck on a loading icon in the Live Demo after logging in using Google
Magic is a simpler Google Analytics dashboard.
See metrics for all websites in one place, faster, and simpler.
I had been thinking about scratching this itch of mine for a while now.
Forced myself to "launch" within a week, it's a landing page, signup form, and live demo of the dashboard.
Check out it out here: magic-ga.com.
(Holding out on the domain purchase before further validation!)
I've only tested it with two of my Google accounts, so I'm not confident in the live demo's capabilities.
Here's what it is supposed to look like:
It's pretty much two minimal features:
- Aggregate dashboard (all websites)
- Individual dashboards
You can see if it works for you 😂
Pretty sure no-one will read this, but if you're reading this, I'd love to hear your thoughts!
-
2
Nice visual design (and you seem to be addressing a very frequent complaint about GA). Keep at it!
-
1
Hey @jonathancai this is fantastic! I used to run a design agency and had about 15+ Google analytics dashboard to review. GA collects a lot of useful data but their presentation really isn't up to the mark and Magic is the magic for it ;)
I did something similar for a client of mine using Google Data Studio. Do you enable which data users can choose to be shown? And can an user see multiple clients dashboard in the same page (like say if I have clients and want to see their data in just one page) ?
-
1
Toggling the shown data is for sure a planned feature.
The second ask is interesting though... Because right now, I’ve only considered all dashboards/websites under a single account. You’d want to aggregate dashboards from multiple accounts?-
1
I honestly believe this has great potential.
Here is an idea of how you can make it even more useful for users? Asking users the top five to seven metrics which are important to their businesses and then suggesting the charts which they need to look at it.
As I mentioned I have Google analytics but there is just way too much info and I dont know which charts would give me the best metrics that are key to my business. So when you suggest the top 5 to 7 charts, user can see them more clearly on Magic and you can send them daily reports via emails.
Yes, I meant aggregated dashboards from multiple accounts but I doubt many users would need it. I just create the GA accounts for my clients but dont see their data anyways.
-
1
Absolutely I need to talk to more GA users.
Still trying to validate that something like this would help people :)
Tysm for your input!-
1
It's safe to say most people these days use GA by default when they track analytics on their website.
Got to figure out to whom would this be most useful? Would it be for founders who are just starting their online business or for businesses which have been operating for quite some time but see the importance of conversion/how people use their website in order to optimise it further.
-
-
-
-
Comment