I’m not surprised that people on a developer focused forum would find little to no value in something that enables automation without programming.
Y’all can probably automate interactions with hundreds of web based services in your sleep, using multiple languages, with unit tests, and a Turing complete ML-based proxy server for high availability.
For the rest of us, Zapier is a little like magic. Granted there are other servothay do the same thing, but none that I’ve found do it as easily and as well as Zapier.
The thing is that no-code solutions are nothing new or unique; they're a dime a dozen, and have existed almost as long as microcomputers have. What's not obvious is why Zapier's version is (apparently) a revelation. I'm not saying there is no reason, but that's probably where the GP is coming from.
To add to your point - so much amount of software already exists in the world. Tools like zapier enables people to put these different software services in an massive assembly line and kind of use them to get their workflow.
why wouldn't someone just hire a developer to do this? My understanding is that the market for contractors is oversaturated right now. You could probably pay someone hundreds of dollars to write software that interacts with web services etc if you look hard enough
Like the grandparent comment, it's hard for me to imagine that there are enough people in need of this service to justify a $5B valuation
When I was a non-programmer trying to build a business that needed software, the last thing I wanted was to rely on something I couldn’t maintain myself. For example, I could wire up DNS by hand, but the last thing I’d want is to have to learn Terraform just so I could update my email MX records. Using Herkou because I didn’t want to manage a server was another example. Using something like Zapier to add a row to a spreadsheet or send an email is also amazing.
Re-reading, you're definitely right. Where I got confused was the "Terraform/Heroku is bad because it adds another layer" but then not understanding why Zapier is better?
Creating a Zap that is hosted by Zapier isn’t just easier than hiring a programmer, it’s also operationally easier. Zapier’s team handled all the devops headaches. Yes, it could be more expensive than a budget developer, but I’d have to maintain an ongoing relationship to handle problems in a way I would not with Zapier. The headache isn’t worth it until you really need something so custom that Zapier couldn’t easily do it anyway. Hire a developer when you need one, and not before.
Regarding Terraform, installing it, learning HCL, understanding the state management, maintaining a place to run it, and learning how not to screw it up were all headaches I did not want.
The process of hiring someone for 'Download attachments from this email address to my Dropbox, then create a note in Quickbooks to pay the invoice' is a lot more complicated / expensive than just using Zapier.
I think you're overestimating what people use Zapier for.
why wouldn't someone just hire a developer to do this?
Because developers are relatively expensive.
Because hiring anyone requires identifying someone who can do the job and trusting them to do it well. And even before that, you have to figure out what the job actually is.
Because hiring anyone requires some sort of contract, and that has legal implications that might require approvals etc.
As a developer with the required knowledge and skills, these services offer little value to me, but I'm not their target customer. That knowledge and those skills are the result of decades of study and practice, which is experience that most people don't have, and for all of those people the cost/benefit comparisons are going to look completely different.
Also, no developer you’d hire with confidence is gonna get out of bed for an “I can barely explain myself, but all I need is...” job that pays a small amount.
Create in zapier, send the login to dev; please recreate. Works fine and can, in very many cases, be done for well under zapier's pro account prices. Especially for tasks you need to run 100k+ per month, which is not uncommon if you are trying to build a product yourself.
Because people want to click and get it done without dealing with hiring someone else.
Tasks that need to be automated could be also very small and not worth the effort of hiring contractors.
BC you have to maintain the code then. Also, a non-engineering team member can quickly get an integration they need done without waiting for a developer.
How to hire capable developers is an open question for even the most sophisticated technology companies. Anything that allows you to avoid that problem is extremely valuable.
I'm in a similar boat thinking where are all those people in need of these things but then I look around and realise that 99.99% of the population will ever go as far as using Zapier and not because there's something wrong with it but because it's a good choice for most. Just last week had my colleague with zero tech knowledge connecting Zapier to a couple of systems in minutes savings the company money and more importantly,not delaying pretty important integration.
> why wouldn't someone just hire a developer to do this?
...
> You could probably pay someone hundreds of dollars to write software that interacts with web services etc.
The reason is exactly that. They don't want to pay hundreds for a dev to dev for them, and that's just for 'one' integration, with Zapier they can always add whatever they want. They'd rather have the 'cheaper' option (which to be real doesn't always end up being cheaper in the long run, but sometimes does).
Why would I pay someone hundreds of dollars to build something that I don’t know how to maintain, when Zapier can do the same thing for free or $20/mo?
Give me a break - there’s no developer that anyone could reasonably trust to write even the simplest program and maintain and modify it throughout they year for less than $20/mo.
Maybe I spend several hours finding someone online who will do the job for $200, then I spend an hour communicating and clarifying my request, then another hour testing the result. 1 months later I decide I need the automated email to have a different subject line, or include an attachment that wasn’t there before, so I find the developers’ email address and pay him another $100 to make the change - or maybe I don’t because it’s too much trouble.
In the alternate world (in which I live), I simply fire up my browser and create a task in about 15-20 minutes. When I need or want a change, I change it myself.
Less time, less hassle, less risk, and in all likelihood less cost.
Your opportunity cost for the time spent maintaining, or managing people to maintain it is much higher than $20, if your business has more important problems you could spend time fixing. It’s very obvious that this is true if you’re a yc-type startup, which could be worth $0 or $1B depending on how you spend a few years.
Y’all can probably automate interactions with hundreds of web based services in your sleep, using multiple languages, with unit tests, and a Turing complete ML-based proxy server for high availability.
For the rest of us, Zapier is a little like magic. Granted there are other servothay do the same thing, but none that I’ve found do it as easily and as well as Zapier.