Thanks for sharing Visualping— can't believe I hadn’t heard of them yet! They seem to be class-leading in “web change detection notifications”, even serving some F500 clients. I checked out some third-party reviews and it seems they sometimes gets complaints around reliability and pricing. That might explain why people have spun up open-source alternatives like changedetection and Huginn, which are more powerful and flexible, but don't match Visualping's user-friendliness and customer support.
I’m definitely doubling down on the agency use case since I have a connection there. But based on what I’ve gathered, here are a few other potential uses:
- For tech marketing managers managing multiple products daily, ensuring everything’s purchase-ready.
- For developers, monitoring forums for important updates, filtering out the noise.
- For warehouse managers, keeping an eye on product pages for restocking purposes, maybe checking in about once an hour depending on demand.
- For marketers, detecting website outages—this should run as often as possible without causing issues.
- For academics, staying updated on relevant research grants to quickly apply.
- For me personally, as a marathon runner who switches out training shoes every 300 miles or so, I want to catch discounts on last season’s models which perform just as well at half the price.
- For everyday consumers, checking international shipping updates, which can sometimes require multiple form submissions.
It also appears perception of web clipping tools is "advanced mode" and not easy to use. I’m excited about making them simple enough for anyone to use. I'm imagining an interface where you can easily save what matters to you, record your activities, and keep those sessions safe and private on your own device for future replay. Also hoping that making it open source builds trust, but I'm still checking how well that works. The goal is a tool that "just works" out of the box while still keeping your data local, private, and secure.
Also noticing that notification filter setup can be complex, and require a lot of upkeep. I’m thinking about how to make that easier and more reliable. Currently plan to use a fine-tuned local LLM for parsing & interpreting HTML and JSON. This approach would help the AI be smarter about timing & sending notifications. For example it would ignore minor changes like a one-minute increase in your commute, but alert you about significant changes like a sudden 15-minute jump.
Let me know if you have any more thoughts on this! Also if you’re open to it, I’d be interested in learning more about how you use Visualping.