People keep asking me the same question when they compare us to Keepa, Jungle Scout, and Helium 10: "how often does the data update?"
Wrong question. Here is the real one: whose clock is it running on — theirs, or yours?
Quick answer
- ⦿ Keepa, Jungle Scout, Helium 10 (and similar tools) maintain their own database of Amazon and refresh it on a schedule THEY control — hourly, daily, whatever their infra decides
- ⦿ When you call their API, you are not asking Amazon anything. You are asking their database how stale it feels like being right now
- ⦿ Asgard has no database of "the answer." Every request is a live scrape of Amazon, right now, for the exact datapoint you asked for
- ⦿ That means the refresh frequency is not fixed by us — it is set by YOU, by how often you call the endpoint
- ⦿ Want a keyword rank checked every 5 minutes? Every hour? Once a day? That is a decision you make, not a plan tier you are stuck with
Why this distinction matters more than the word "real-time"
Every scraping tool claims to be accurate. Almost none of them explain the actual architecture underneath, so let's be blunt about it.
Tools like Keepa, Jungle Scout, and Helium 10 are built around a central database. They run their own crawlers on their own schedule, store the results, and serve YOU whatever is sitting in that database when you make a request. Your API call is a lookup, not a live event.
This is a completely reasonable architecture for what they do. Keepa is a historical archive — nobody expects an archive to be "live." But when people ask how often Keepa-style tools "update," the honest answer is: whenever their internal crawler schedule says so, not whenever you asked.
What that looks like in practice
- You send a request. It hits their servers, not Amazon.
- Their server checks its own database. The data in there was captured by their own crawl, at a time you did not choose.
- You get an answer. The response time is fast — because it is just a database read — but "fast" and "fresh" are not the same thing.
- The freshness ceiling is fixed by them. If their crawler touches a given ASIN or keyword once every few hours, that is the best resolution you will ever get, no matter how many times you hit the API in between.
None of that is a knock on those tools — it is just what a shared, provider-scheduled database is. But it means your "real-time decision" — a PPC bid adjustment, a BuyBox alert, a stockout response — is being made on a number that could be minutes or hours old, and you have no control over closing that gap.
How Asgard is architecturally different
Asgard does not have a "the data" database sitting between you and Amazon. There is no central snapshot we refresh on our own schedule and then serve to everyone.
- Every request is a live scrape. When you call the API for a keyword rank, a BuyBox check, or a product page, we go get that page from Amazon right then — not a stored copy from earlier.
- Time-to-response is time-to-Amazon, not time-to-database. The ~2 second response you get back is the time it took to fetch and parse the live page, not the time it took to read a cache.
- The refresh interval is whatever YOU set. Poll a keyword every 5 minutes if you are running active PPC bidding. Poll a BuyBox once a day if you just need a daily brand-protection sweep. That cadence is entirely yours to define — it is not capped by our crawler schedule, because we do not run one.
- Any datapoint, any frequency. Rank, price, BuyBox owner, stock status, badges — whatever you ask for gets scraped fresh, on your terms, every single time.
Frequently asked questions
- So does Asgard update every X minutes/hours? There is no fixed update interval on our side. The "update" happens exactly when you make the request. If you call it every minute, you get a fresh scrape every minute.
- Is that more expensive than a database lookup? It is priced per successful request, and it buys you something a cached database cannot: a guarantee that what you get back reflects Amazon right now, not Amazon at some earlier point their crawler happened to run.
- Why do Keepa / Jungle Scout / Helium 10 work this way? Because their core value is aggregation and history — storing and indexing Amazon data over time. That requires a database they control the refresh cadence of. It is the right tool for historical trend analysis.
- When do I actually need live-per-request data instead? Any time the decision is time-sensitive — PPC bid changes, BuyBox win/loss alerts, stockout detection, competitor price moves, live SERP checks by zip code or device. If the action you take depends on "right now," you need a scrape of right now.
The short version
"How often does the data update" is the wrong frame entirely when you are talking to a real-time scraper. Database-backed tools update on their schedule. Asgard has no schedule of its own — it grabs whatever datapoint you ask for, at whatever frequency you decide, straight from Amazon, every time.
☞ https://asgardata.com/