Prime Day Price Tracker: ASINs, Big Deal Days, 2027

Prime Day 2026 ran 23-26 June (it was 8-11 July in 2025), and Prime Big Deal Days ran 6-7 October 2026. The 2027 dates aren't announced. To track Prime Day prices, record your ASINs weekly for 4 weeks before the event, check at each deal-drop time during it, and compare the event low with your own 30-day median rather than with Amazon's list price.

This guide covers the Amazon-specific parts: the event history and how dates are announced, tracking the same product in several Amazon countries, exactly what a signed-out scraper can and can't see during a members-only sale, and a schedule timed to Amazon's deal drops. For other events, see the retail sales calendar.

Prime Day and Prime Big Deal Days dates

Two things follow for tracking. First, don't schedule off last year's date: Prime Day moved by about two weeks between 2025 and 2026. Start the baseline when the event month is announced, not when the dates are. Second, during the 2026 events Amazon released new "Today's Big Deals" at 12 a.m., 8 a.m. and 1 p.m. PT each day. Those are the moments to check.

What a scraper sees during a members-only sale

Prime Day deals are for Prime members. A scraper that doesn't sign in sees the page a signed-out visitor sees, which matters:

So a scraper measures the public price around the event, which is still the right baseline. If you are a Prime member, a member-only deal price is the public price minus a discount you can confirm by hand on the day.

Track the same ASIN in several countries

Amazon runs Prime Day in many storefronts at once, and the same ASIN often exists in several of them. The E-commerce Product Scraper Pro actor reads 24 Amazon storefronts. Its README rates amazon.com and amazon.de as supported and the other 22 (including amazon.co.uk, amazon.fr, amazon.ae, amazon.sa) as best-effort. In its latest cloud tests, amazon.de delivered 84% and amazon.co.uk 95% of deliverable products. Prices come back in each storefront's own currency.

import os
from decimal import Decimal
from apify_client import ApifyClient

client = ApifyClient(os.environ["APIFY_TOKEN"])
run = client.actor("constructive_calm/ecommerce-product-scraper-pro").call(
    run_input={
        "detailsUrls": [
            {"url": "https://www.amazon.com/dp/B09B8V1LZ3"},
            {"url": "https://www.amazon.co.uk/dp/B088NRLMPV"},
            {"url": "https://www.amazon.de/dp/B0DHJRLQPB"},
        ],
        "monitorMode": True,
        "onlyChanges": False,
        "maxProductResults": 3,
    },
    max_total_charge_usd=Decimal("0.20"),
    logger=None,
)
for r in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(r["site"], r["price"], r["currency"], "was", r.get("originalPrice"),
          "| seller:", r.get("seller"), "| tier:", r["tierUsed"], "|", r["changeType"])

Real output, 2026-10-20 (pip install apify-client first). This was the second run of the list that day:

amazon-us 79.99 USD was None | seller: Amazon.com | tier: datacenter | unchanged
amazon-de 9.34 EUR was 13.99 | seller: SimpleLife LLC | tier: residential | unchanged
amazon-gb 7.49 GBP was 9.49 | seller: AnkerDirect UK | tier: residential | unchanged

tierUsed shows how each page was fetched. amazon.com went through a datacenter connection; the UK and German pages needed residential traffic, which is billed per MB (see costs below). changeType was new on the first run of this exact URL list and unchanged here; on an event day you'd expect price-down and price-up.

Was the event price a real deal?

The question to answer after the event: was the lowest event price lower than what the product normally cost? This script compares the event low with the 30-day pre-event median and the lowest pre-event price. The CSV is a made-up example (B0EXAMPLE1 is not a real ASIN) to show the logic; in practice it's your own exported snapshots.

scrapedAt,asin,price
2026-09-10T07:00:00Z,B0EXAMPLE1,49.99
2026-09-17T07:00:00Z,B0EXAMPLE1,44.99
2026-09-24T07:00:00Z,B0EXAMPLE1,49.99
2026-10-01T07:00:00Z,B0EXAMPLE1,52.99
2026-10-06T00:00:00Z,B0EXAMPLE1,52.99
2026-10-06T08:00:00Z,B0EXAMPLE1,39.99
2026-10-06T13:00:00Z,B0EXAMPLE1,39.99
2026-10-07T08:00:00Z,B0EXAMPLE1,42.99
2026-10-08T07:00:00Z,B0EXAMPLE1,47.99

import csv
from datetime import datetime
from statistics import median

EVENT_START = datetime.fromisoformat("2026-10-06T07:00:00+00:00")  # 12:01 a.m. PDT
EVENT_END = datetime.fromisoformat("2026-10-08T07:00:00+00:00")    # 11:59 p.m. PDT, Oct 7

rows = [(datetime.fromisoformat(r["scrapedAt"].replace("Z", "+00:00")), float(r["price"]))
        for r in csv.DictReader(open("prime_hourly.csv"))]
before = [p for t, p in rows if (EVENT_START - t).days in range(0, 31)]
during = [(t, p) for t, p in rows if EVENT_START <= t < EVENT_END]

base = median(before)
low_t, low = min(during, key=lambda x: x[1])
print(f"30-day pre-event median {base:.2f}, event low {low:.2f} at {low_t:%m-%d %H:%M} UTC")
print(f"real discount vs your baseline: {(base - low) / base * 100:.1f}%")
print(f"lowest pre-event price {min(before):.2f} -> event beat it: {low < min(before)}")

30-day pre-event median 49.99, event low 39.99 at 10-06 08:00 UTC
real discount vs your baseline: 20.0%
lowest pre-event price 44.99 -> event beat it: True

Notice the example's price rise to 52.99 just before the event. A "25% off 52.99" banner would overstate the cut; against the median it's 20%. The fake discounts guide goes deeper on this pattern, and Amazon price history without Keepa covers building the long-run history.

A Prime Day schedule

The event cron assumes Pacific Daylight Time (UTC-7), which applies to both a June/July Prime Day and an early-October Big Deal Days. Hourly (0 * * * *) catches more, at more cost.

Keep the ASIN list fixed from the baseline to the end: the actor's monitor state is keyed by the exact set of input URLs, so adding an ASIN mid-event starts a fresh comparison. With onlyChanges: true, unchanged products aren't written but are still counted and billed at the lower "product checked" rate.

What it costs

Pay per event, from the actor's README. FREE tier: $0.0048 per product row, $0.0012 per unchanged product with onlyChanges, $0.005 per run start, $0.018 per residential MB. GOLD: $0.0008, $0.0002, $0.005 and $0.014.

• 100 amazon.com ASINs, deal-drop schedule (3 runs a day for 2 days = 6 runs), onlyChanges, nothing changes, FREE: 6 × (100 × $0.0012 + $0.005) = 6 × $0.125 = $0.75. Changed products cost $0.0048 each instead.
• The same, hourly for 48 hours: 48 × $0.125 = $6.00 on FREE, or 48 × (100 × $0.0002 + $0.005) = 48 × $0.025 = $1.20 on GOLD.
• Non-US storefronts add residential traffic: the README measured amazon.co.uk at 0.3-0.5 MB per product, billed at 89% of the measured MB. 100 UK products on GOLD come to about $0.46-$0.71 per full snapshot.

amazon.com product pages use datacenter connections, so they have no traffic fee. Apify's free plan gives $5 of usage every month, no credit card required; the actor has no free trial of its own. The free e-commerce price tracker query builder builds the input without code and shows a fixed example of the output shape.

Limits worth knowing

• Amazon reviews: only the 8-13 top reviews on the product page; full review pages need a login.
• Run-to-run variation: Amazon occasionally serves empty pages or captchas. The actor reports those per input in a Status dataset instead of an empty success, and blocked pages aren't charged.
• "Cannot be dispatched": a non-US page fetched from the wrong location can refuse to show a price. The actor retries on another route; if it's still blocked, the product comes back without a price and with a note.

Related guides

• Retail sales calendar 2026-2027
• Amazon price scraper by ASIN
• Amazon price history without Keepa
• Price drop alerts by email or Slack

Disclosure: I built the Apify actor and this site. The baseline script works on prices from any source.

Sources: Amazon: Prime Day 2026 dates, countries and deal-drop times, Amazon: Prime Day 2025 date, 9to5Toys: Amazon announced June Prime Day on 29 April 2026, NBC News: October Prime Day 2026, PCWorld: Prime Big Deal Days 2025, the actor's README (Amazon coverage, yields, pricing) and a live actor run on 2026-10-20.