Scrapingdog Weekly Changelog: July 6-12, 2026

Keeping APIs reliable is a constant job. Websites change their layouts, platforms tweak how they respond, and every one of those changes can quietly break the data flowing into your applications. Our job is to catch those breaks before you do and this week, we fixed a handful of them across Google AI Mode, Amazon, and Google Hotels.

If you've been pulling data from our Google AI Mode API, you may have noticed that some references quietly disappeared from the response even though they were technically present in Google's page.

Here's what was happening: Google embeds certain references inside the AI Mode HTML in a "hidden" format that isn't immediately visible in the visible text. Our parser was only picking up the references that were clearly displayed, and skipping the ones tucked away in the hidden markup.

We've now updated our parser to dig into that hidden layer as well. So starting this week, every reference Google includes visible or hidden will show up in your JSON response. If you're building anything that relies on source citations or reference tracking from AI Mode results, you'll now get the full picture instead of a partial one.

This one's a big quality-of-life improvement if you're building a price tracker, deal finder, or any tool that depends on knowing whether a product can actually be bought right now.

Previously, our Amazon Product API returned details like price, title, ratings, and images but it left out one important field: availability. That meant developers had no direct way to know if a product was in stock, out of stock, or running low, without writing extra logic to guess based on other fields.

That gap is now closed. The API response includes a clear availability status for each product, so you can instantly tell whether something is:

• In stock

• Out of stock

• Limited in quantity

This is a simple addition, but it saves you from writing workarounds and gives your application more accurate, decision-ready data straight out of the box.

Over the past couple of weeks, some of you may have noticed occasional failed requests when calling our Amazon Search API or Amazon Product API not constant failures, but random, intermittent ones that made success rates a little less predictable than usual.

We traced this back to an internal issue affecting how requests were being processed under certain conditions. It's now fixed. Both APIs are back to delivering consistent, reliable success rates, so you can count on fewer retries and more predictable performance in production.

If you had monitoring or fallback logic in place to catch these failures, you should see that logic triggering far less often, ideally, not at all.

Last but not least, our Google Hotels API was going through a rough patch. An internal bug was causing a noticeably higher number of failed requests than normal, which likely affected anyone pulling hotel listings, pricing, or availability data at scale.

We identified the root cause, rolled out a fix, and the API is now back to operating normally. You should see significantly fewer errors and much more stable performance going forward.

None of these updates are flashy, and that's kind of the point. Reliable scraping isn't about big, dramatic overhauls, it's about catching the small things before they turn into bigger problems for you:

• A missing reference here

• A missing stock status there

• A quiet dip in success rate that adds friction to your workflow

• An internal bug that shows up as random failures

Each of these, left unfixed, chips away at trust in the data your application depends on. Our goal every week is simple: make sure the data you get from Scrapingdog is as complete, accurate, and dependable as possible, so you can focus on building, not debugging someone else's website changes.

If you're using any of these APIs, here's the short version:

• Google AI Mode users — you'll now get hidden references included automatically, no extra work needed on your end.

• Amazon Product API users — you can now filter or sort based on real-time stock availability.

• Amazon Search & Product API users — expect fewer failed requests and more consistent performance.

• Google Hotels API users — enjoy a much more stable, error-free experience.

All of these updates are already live, so there's nothing you need to change in your integration to start benefiting from them. Just cleaner, more complete, and more reliable data flowing into your applications.

That's it for this week's update. As always, if you run into anything unexpected or have feedback on these changes, our support team is just a message away.

We'll be back next week with more updates. Until then, happy scraping!