Workflow Automation
Hi,
I have some idea but want to ask you about it. I would like to create a platform that will automate some actions. For example if some data will be added to the specific table. This will trigger some actions that will cause creating some PDF (let's say invoice) in specific catalog and the specific API endpoint will be executed.
This kind of software is called Workflow Automation or Robotic Process Automation (RPA is more like making some actions on the view not in the back-end). Obviously it's nothing new. There is for example Zapier, MS Flow.
Why I would like to make my own software?
I can see that a lot of companies in my country and probably around the globe that make some custom code (background services) for making this kind of integration and often it is not optimal. I would like to create a software that can detect if trigger fired (row added to the database, file modified/created, email received etc.) and then it can execute actions based on that.
What would be different from Zapier, MS Flow?
You can install multiple agents/workers on your local or virtual machines not necessarily on the someone's else servers. You will also be able to configure everything via Web App (also managing agents/workers) etc.
Do you think it makes sense to do this? What features would be crucial for you? Any thoughts appreciated!
Thanks,
Przemek
We, developers tend to build things either out of curiosity or to challenge ourselves. If you want to build this to make money (which you should) and not to scratch the itch then you need to first identify the market you want to penetrate ("who is your potential customer").
Eg. Email Automation is huge in ecommerce, cold email marketing (email triggers), onboarding users, etc. Most of the solutions out there are not for developers to install on their server, it's hosted solution doing specific thing.
If you are building for developers then each institution will have their own unique requirements and they would rather build it themself.
That being said, there are quite a few products (hosted solution) out there and you have named couple, how do you plan to compete with them or better than them?
Thank you for your answer imomin! I really appreciate. In my solution (if hosted) you can have a multiple agents and you can set their properties separately and manage them through the web ui. Also you can move all solution to your platform (cloud or your own servers/local machines).
Here is my use case:
I have been working for pretty large organization that was selling pretty much all electronics: like tv, consoles etc. They have a small development team (and sometimes they hire external software house) and for the integration of software they needed to create some custom windows services that for example were getting data from the table/queue and with the api dll they were creating order in their ERP system. Everything has been done like you mentioned... "If you are building for developers then each institution will have their own unique requirements and they would rather build it themself". But the disadvantage was that after a years they had a ton of undocumented windows services / scheduled tasks / sql jobs and nobody knew what is going on there. With my solution they can achieve similar result but in the web app there is a list of all workers and you can manage them remotely. What is more they can hire me to create custom plugin (for example to integrate some custom software) or create plugin with the public SDK.
What do you think?
Hello @dzimiq! I'm working on something similar. Take a look to https://www.tideflow.io/ perhaps you can find some ideas or contribute! It's fully open source.
Hi Guys,
I had a few meetings last time that gave me additional view on this app. Also your comments are really valuable. Now I need to put this all requirements on paper and create some roadmap, milestones etc.
Then I will start with the product development!
Thanks,
Przemek
Like how to think. But I'm worrying if you've done enough research.
Who would be your ideal customers and how do you measure the switching cost from one application to yours?
Like you said there's Zapier out there, it might also worth the time to look at Parabola.IO and IFTTT. I'm using these tools to simplify tedious routine for my little project. It works pretty well so far.
Happy to exchange ideas!
Seeing that many in-house enterprise applications have a lot in common (CRUD, form-based data processing, API integrations, you name it ...) I've been thinking about creating such a product myself time and again.
However, in addition to the points already mentioned in other comments, one important aspect to this is that while developing bespoke software to do this again and again is expensive the resources (i.e. developers and development and deployment environments such as Java application servers) to do so usually are readily available in larger companies anyway while new software has to go through a lengthy sales and approval process.
Hence, if such a requirement comes up - and Excel doesn't cut it (because Excel frequently is the first go-to self-service solution for automating processes) - stakeholders tend to opt for repeatedly developing solutions in-house rather than purchasing an off-the-shelf solution that already covers 80% of the requirements.
Decisions in larger companies by and large aren't motivated so much by the financial bottom line (though that certainly helps with making your case) but sadly rather by politics and defending entrenched fiefdoms. If you can win someone high up in the hierarchy as a proponent of your solution selling it might work but otherwise you'd be in for an uphill battle.
With cloud solutions and low-code products becoming more popular this might change in the future, though.
Still, incumbents such as Microsoft or Oracle already offer solutions in that space. In a somewhat typical enterprise fashion they're often not exactly user-friendly but - large company dynamics again - no one ever got fired for choosing Microsoft or Oracle ...
If you'd like to discuss this further please get in touch (either via public comment or the email address mentioned in my profile). In my opinion, such a solution from a product perspective absolutely does make sense. It's really a question of how to position and market it to potential buyers.
Thank You Bjoern! I will put some stuff on the paper before starting development. Then maybe we can discuss that either public or private?
Have a good day!
Sure. You have nice day, too.
Who is your potential customer for this ?
Because , being a developer we always use event based programming.
An event get triggered / dispatched in one module . An event listener will catch the event and do some action in another module.
So what makes me to use your solution ?
Thanks for your response. I will give you my example that I wrote for the comment above. This should describe my initial thought how my customers can use my system:
I have been working for pretty large organization that was selling pretty much all electronics: like tv, consoles etc. They have a small development team (and sometimes they hire external software house) and for the integration of software they needed to create some custom windows services that for example were getting data from the table/queue and with the api dll they were creating order in their ERP system. Everything has been done like you mentioned... "If you are building for developers then each institution will have their own unique requirements and they would rather build it themself". But the disadvantage was that after a years they had a ton of undocumented windows services / scheduled tasks / sql jobs and nobody knew what is going on there. With my solution they can achieve similar result but in the web app there is a list of all workers and you can manage them remotely. What is more they can hire me to create custom plugin (for example to integrate some custom software) or create plugin with the public SDK.
This comment was deleted 6 years ago