Having second doubts
The title basically says how I feel right now.
I have validated my idea a ton, people like it, they would pay for it. Then I showed a close friend my product and he asked a single question that made me doubt everything about my current project:
"Why should a developer implement your solution and pay for it, if they could just as easy implement the solution you provide themselves?"
I was stumbled. And I couldn't think of anything else rather than the "support" they'd get.
After a pretty sleepless night, I still have no answer that would answer this question.
I am seriously thinking to scrap the project.
It depends on your target market and the value of your solution. If you're selling something simple that can be done in a few hours to DIY developers who love to build rather than buy, then it's not going to work. If you're selling something more complex to businesses/eng teams/managers, then there's at least one argument for it - if it's cheaper than the dev time it'd take to reimplement (not including TCO), it makes sense to buy rather than build.
cheaper to run and maintain is a valid argument. I believe the solution as is, is not that complicated, but I do provide additional insights into usage, tracking and documentation alongside my solution, drastically increasing the development time of developers if they need that additional statistical data.
thank you, I did not think about that angle.
If you tell us what it is, I'm sure some of us (myself included) would be happy to critique it for you.
Oh yes, of course. It's https://vatfy.com
Bear in mind though, it is not yet finished as some things are undocumented and thus not really usable without the documentation. The backend however is fully fleshed out.
Hi,
Firstly, some background. I have some recent experience with VAT rates. I write Gig'o'Books (www.gigobooks.com), an accounting app which uses VAT data and calculates VAT.
I don't think you're gonna like what I'm gonna say. Brace yourself ...
There's a public API (provided by the EU member states or something) for this, isn't there? So I don't see a reason to use a third-party service. As a developer, if I was writing my own software, I'd just integrate with the public API myself.
The data for this is publicly available. It's also available in a convenient form at https://github.com/staaky/vatrates (albeit slightly out of date). So I don't see a compelling reason to get it via a third-party API.
(For myself/gigobooks, I even just hardcode the data into source code myself. See https://github.com/gigobooks/gigobooks/blob/4ee41ee1bfaf3c492e2a59b3ae5e45f48ec2d587/src/core/tax.ts#L583)
Uh, this is a standard and well-known calculation/algorithm. I don't see a compelling reason to get it via a third-party API.
----
Overall, it doesn't look good.
Sorry, but as a developer, I don't see a strong reason to use your API.
The VAT rates and calculation I can code myself. The VAT number validation ... maybe I'd use your API if it was 100x faster than the publicly available API or something. But I'd have to be doing a crazy amount of volume for that to be significant.
Is there some secret sauce or magic in your service that I don't know about?
If not, I'm sorry, but 99% of the time, my personal opinion is that ... I don't see a need to use your API.
----
Sorry, I'm not trying to be mean here. I'm just trying to give constructive, neutral, unbiased, and brutally honest feedback.
just had a look at your sourcecode. line 592 has an error:
DE: ['19', 'reduced:7'],
should be:
DE: ['16', 'reduced:5'],
until 31.12.2020. On 01.01.2021 it goes back to being
DE: ['19', 'reduced:7'],
:)
Hi,
Oh, yeah, Germany temporarily reduced their VAT rate this year due to COVID. I'm aware of this but I didn't bother to update the source code because Gig'o'Books doesn't support Germany VAT yet ... and when I get around to it, it'll be 2021 anyway.
Thanks for the heads up, though.
point taken, no hurt feelings.
quick question. how fast can you implement said thing, worry-free, with the appropriate level of documentation to stay compliant? how do you make sure that each and every price you show is calculated the way it should be?
Yes, EU VAT rates are publicly available. The documentation that when you charged your customer had a valid VAT rate, and what that rate was and you can show regulators proof that you did it the correct way, is not.
Hi,
Huh? I don't quite understand what you're asking.
The VAT calculation is a well-known and standard calculation. It's used in non-EU countries as well (where it's known as GST). If I've implemented an algorithm correctly, it's not going to need 'documentation' to stay compliant.
Hmmm ... I feel like I understand the individual words that you've written, but when I try to understand the entire sentence, it doesn't make sense. I feel like I'm misunderstanding something somewhere.
Again, I feel like I don't quite understand something. What documentation are you talking of?
For example, I know that EU invoices need to contain business name, address, VAT ID etc. for both parties, date of invoice etc. Is this the documentation you are referring to? Or something else?
ah sorry about that. english is not my mother tongue, so please forgive me here. I'll try to break it down.
If your government asks you to provide proof on how and when you calculated VAT (or moreso with no VAT), you need to provide the identification number VIES gives you. And you need to do that for every transaction made.
One quick example, that others may find useful as well:
You sell a monthly subscription service. Said customer provides you with a VAT ID number. Now, you can validate this one, but you also need to make sure to save the VIES identification and you need to do it for every monthly invoice. The VAT ID may be invalidated, the company may have switched name and or address, and so forth. By being able to prove that you identified it for every invoice may be required by your tax man. It's not a biggie to do, my service just takes care of that for you, so you don't have to, plus it does some minor other things that helps you show correct pricing for every visiting prospect by calculating net, gross and moss prices based on their visiting country.
Sure, you could do that yourself, but isn't your time better spent working on refining your product, instead of worrying how, when and where you store all of that data and how you show whom what prices?
Hi,
Ah, I'm unaware of this.
For example, I mainly use https://www.revenue.ie/en/vat as my VAT reference since it's in English, and there's no mention of this requirement (that I can find).
(I wonder if it's a Germany-specific thing?)
Can you point me to the government documentation on this? Any of the following ...
... that mentions this requirement would be appreciated.
I'll read up on it first so I don't make comments about something I'm not knowledgeable about.
ah yes, here you go:
https://en.wikipedia.org/wiki/VAT_Information_Exchange_System
Article 138, Paragraph 1 of Council Directive 2006/112/EC
Hi,
Ah, I didn't know about this. Thanks for pointing me to it. It's interesting.
So, going back to your previous questions ...
Are you asking me how long it'd take me to integrate the VIES API into Gig'o'Books and store the unique 'Consultation Number' for each VAT number verification request?
It depends on the API is and how easy-to-use it is. If it's an average API ... maybe 1-2 days?
----
So ... I presume the VIES API is publicly available? If it is ... then I don't see a compelling reason to use your API. Unless your API is much easier to use than the VIES API, or it's much faster, or some other advantage.
Or, is the VIES API private instead? And you reverse engineered/web-scraped it and made an API for it? If so ... okay, that's a good reason to use your API.
Yeah no, I’m a dev and if I needed something like that, I would definitely wanna use an easy solution so I didn’t have to spend any time figuring it out myself.
You just have to find your segment of devs who really need this, and also have some money. Who value their time more than it would cost to buy your solution.
I wouldn't be so discouraged. You should check out the Hacker News post about Dropbox. By that reasoning, there shouldn't be any open source libraries for simple things.
If it is easy, just build it in a weekend and share it here or on a niche community.
Yes, I will probably make it available, free of charge, to the IH community. Just ping me with your account email and I'll upgrade you to the paid plan for free.
It definitely has potential. There are companies for all types of apis - last one that comes to mind is a cost of living api.
This will still take at least a week to setup everything properly, and writing production grade code takes time. Why build something from scratch, if it's cheap, and it already exists, and there is some guarantee on reliability?
Your api might fit the use case for companies who'd rather pay than setup the entire thing.
Good luck!
Can you elaborate?
Yes, most of those I have spoken too were developers and they seemed very interested. What I did not account for, was "generalistic" developers, kind of entrepreneurs, like my friend, that implements everything themselves. Maybe this persona is not my target market?
Exactly. Not everyone is your market. Your challenge now is to prove the notion people will pay for it.So how are you going to do that?
that's a hard thing to answer. i have never been this far yet. any pointers?
You can implement everything by yourself. The question is if it's worth it. If I can pay a bit for something and never think about it again I'll consider. If the documentation and support are good I'm sold.
The bigger issue are open source libraries, if there is none that's doing what you're doing you probably don't have to worry.
Thanks. Already working on the documentation. :)
It depends a lot on the type of developer. I have been burned by 3rd party dependencies and tend to minimize them and build more things by hand. I'll use larger libraries with very clear benefits (e.g. an MVC framework like Phoenix/Rails/Laravel) and I'll use very small utility libraries (like Lowdash.js), but I'm not going to use something like Auth0, or a third party referral service, etc.
I can build that kind of functionality faster and easier myself than integrating somebody else's code and be sure it will work exactly how I want it to. Similarly, I wrote my own API wrapper libraries for ConvertKit, Typeform and other services I work with. I often realize when debugging issues that an external library I'm using isn't quite as flexible as I needed it to be and also that I can write 20 lines of code that replaces it in my app.
Many other developers, including me of 6 years ago do want to dish everything possible off to 3rd party services. Part of the difference is cultural, part of it is experience and part of it is how low a level you're working at and how much you're worried about library churn. But there are many people who want 3rd party services. In the case of your service specifically, it's even more compelling since it's tied to regulatory issues and the will likely require updates even if they wrote it internally for themselves.
Look for slightly less technical customers. People who are running their apps off WordPress are probably one subset worth looking for. JS-only devs are another group with a lot of promise (due both to average experience and cultural reasons).
Solid advice. Thank you! I am thinking of modularizing Vatfy and offer direct integrations to Wordpress and other e-commerce projects for ease of implementation.
I'm curious what this is, too. The people who said they would pay for it - did they say why?
unfortunately, I blanked out to ask that question ;( I was just happy to hear, they would pay for it... So I guess, I need to get back to interviewing people about it.
I can give you another developers opinion, but I'll need to know more about it than is here.
https://vatfy.com
Would love some feedback! Thank you.
Looking at the product (and your answer to me below) I actually do see the value in this and the timesaver. More important I see where you can grow from here for internationalization, US Sales tax rates and compliance and the promise to always keep things updated. Given your pricing I think it still has value at scale and you have a path to build on it to further it. I appreciate the doubts, but I'm not sure I would totally shelve it at this stage. Get some users and see what happens.
:) thank you @Khidr9. I appreciate it!
Hey, site looks good!
If I or the companies I work for and with would need something like this I think we could consider going for it. The cost seems quite low when compared to putting that week or so into it to do it yourself (approximately what you're saying on site).
Just because devs could do something like that themselves doesn't mean they would. Quite often it makes more sense to buy it as a service and focus on what is unique about whatever we are building.
I love that you're calling these second doubts. So, I don't know a ton about your product, but this question is what ultimately led us away from using nylas for our google integrations. It's a fair question. Can your solution be useful to non-developers? Can you make it a 10x time saver so implementing your solution saves tons of time for a developer to implement on their own? If not, then it's fair to have those doubts. But, you said you validated your idea, people liked and said they'd use. Were those people devs? Is your friend? Who is closer to your market?
Can your solution be useful to non-developers? Not yet. I have build some manual usage in the dashboard for non-devs and am currently implementing another non-api use case to account for non-devs.
I guess, statistics, documentation/history (for compliance) etc. would be a development time drain, that most companies would shun here. So that could be an angle.
Most people I talked to were either entrepreneurs or developers. I did not have a single entrepreneurial developer at hand (except my friend) and him being identified as my target persona, this busted my idea at this time. He is very thorough and smart, and I value his opinion. Hence, me having second doubts.
Is this for hashr?
Why target devs?
Sometimes the person making the buy decision might be different than the dev so, even if the dev can implement it on their own, having a process and an sla can be better for the overall business.
Engineering teams always say stuff like this, in fairness.
'I could code this over the weekend!'
Yeah, maybe, but even if you could, you won't. Show me the developer with free cycles to take on some extra bonus project out of the blue. Devs always say stuff like this, and they believe it, even if you shouldn't, but they really aren't your market. The market is the management who is trying to keep their team working on something productive and keep accounting happy. Make it easy for the guy running the team, don't worry as much about programmers flexing their muscles.
If people start pushing back on you, the response is all the weird corner cases that result from government bureaucracy that have to be handled and most importantly maintained by somebody. Who's gonna own that task forever? "Oh some country somewhere passed some tax relief bill, so the VAT is different for the first half of this year." I actually had to deal with something very recently where there is one tax rate for one class of products and an entirely different one for regular products. Who tracks that? Who makes sure it's right all the time? It's a lot of work to do-- they should pay you a little bit of money and not get constantly interrupted by the shifting landscape.
(I don't live in the EU and I don't know jack about VAT, but I know tax assessment and collection is a giant pain in the ass for me personally, and Avalara is a big big company for a good reason.)
Great post David - I think this is what entrepreneurs feel like all the time. One day you've spoken to a customer who might buy and you're building something thats going to explode, the next day you feel like you've wasted months of your life!
If you've done good customer validation, that's really all you need. Get real commitments though - everybody you speak to will tell you its great! Ask them to pay and push them to commit. Not easy and often awkward initially- but thats sales!
I'm sure someone has dropped in the Dropbox story on Hackernews, where someone said pretty much exactly what your friend has pointed out .. but there's a tonne of busineses that are successful, simply because they make things easier. Or it looks nicer. Or developers don't want to spend their time doing that exact thing.
Also, it looks like you're doing for VAT, what IPinfo has done for IP's - and they look like they're doing very well!
This comment was deleted 6 years ago
Solid thought. Thank you. I will try to flesh that out more via copy of the website.