Home

About TimeYourBuy

Hi — I built TimeYourBuy on nights and weekends because I got sick of Amazon labeling something a "deal" when it was actually the highest price it had been all year.

The site tracks 88,000+ Amazon products and 28 million price points across years of price history. When you look up a product, you see the full chart, where today's price sits in that range, and a plain-English signal: Strong Buy, Buy, Fair, Wait, or Strong Wait. I backtested the scoring on historical data — the Strong Buy signal was right 73.6% of the time.

I don't sell products. I don't push you to buy. I just show the data. If Amazon's price today is a rip-off vs. its own history, I want you to know before you check out.

How the signals work

My scoring engine compares the current price against a rolling 12-month price history. The score (0–100) reflects how far today's price sits below its historical average, adjusted for category, volatility, and — for commodity categories like Memory and Storage — live market conditions. I validated the signals on a survivorship-corrected backtest: Strong Buy calls were correct 73.6% of the time.

Who's behind this

Just me — one person. If you spot a wrong signal, a stale price, or a broken guide, email me at timeyourbuy@gmail.com. Real inbox, real reply.

Affiliate disclosure

TimeYourBuy is an Amazon Associate. I earn a small commission when you buy through my links — at no extra cost to you. It's how the site stays free and how I pay for the data feed.

Privacy Policy · Terms of Service

Find us on

Home

Privacy Policy

Last updated: May 9, 2026

TimeYourBuy doesn't sell your data, doesn't track you across the web, and doesn't set tracking cookies. We store your email only if you ask us to send you a price alert — and we delete it the moment you unsubscribe.

Here's exactly what that means.

What we collect

Email address — only if you sign up for a price alert. We use it once: to send the alert you asked for. We never sell it, share it, or send you anything else.

Anonymous usage analytics — we use Umami, a privacy-respecting analytics tool hosted on Umami Cloud. It records page views, referrer, browser type, and operating system in aggregate. No cookies. No fingerprinting. No cross-site tracking. No personal identifiers. Data is kept for 12 months, then deleted.

Search queries — the searches you run on the site are logged so we can improve product matching. They are not linked to your email or any personal identity.

What we don't collect

• Payment information — we don't process payments
• Advertising or tracking cookies of any kind
• A behavioral profile of you
• Anything we'd need in order to sell your data, even if we wanted to

Price alerts

When you register an alert, we store two things: your email and your target price. When the price hits your target, we send one email. That's it.

Every alert email includes an unsubscribe link that immediately deactivates your alert. We also send a List-Unsubscribe header so Gmail and other email clients can offer a one-click unsubscribe button. If you don't unsubscribe and your alert sits unused for 12 months, we delete it.

Amazon affiliate links

TimeYourBuy participates in the Amazon Associates Program. When you click through to a product, Amazon may set its own cookies under their privacy policy — we have no control over that. We earn a small commission on qualifying purchases at no extra cost to you. This is what keeps the site free.

Cookies

We don't set any first-party tracking cookies. Umami, our analytics tool, is cookie-free by design.

Data retention

• Alert data: deleted when you unsubscribe or remove the alert, or after 12 months of inactivity — whichever comes first. Alerts stay active until you remove them (at most one email every 3 days per product).
• Analytics data: kept in anonymous, aggregate form for 12 months, then deleted.

Your rights (GDPR / CCPA)

Depending on where you live, you may have the right to:

• See the data we hold tied to your email
• Delete all of it
• Opt out of any future emails (the unsubscribe link in any alert does this immediately)

Email timeyourbuy@gmail.com to exercise any of these rights. We respond within 30 days.

Contact

Questions, concerns, or you spotted something we should fix? Email timeyourbuy@gmail.com.

Home

Terms of Service

Last updated: April 8, 2026

By using TimeYourBuy, you agree to these terms. If you don't agree, please stop using the site.

What TimeYourBuy is

TimeYourBuy is a price intelligence tool. We track publicly-available Amazon pricing data and provide buy/wait signals based on historical price analysis. We are not a retailer, marketplace, or financial advisor.

No purchase guarantees

Prices and availability change constantly. Our signals represent our best analysis of historical data — they are not guarantees of future pricing. Always verify the current price on Amazon before purchasing.

Accuracy of information

We do our best to keep price data accurate and up to date. However, prices are sourced from third-party data providers and may not reflect real-time availability or seller-specific pricing.

Affiliate relationships

TimeYourBuy participates in the Amazon Associates Program. We earn a commission on qualifying purchases made through our links. This does not affect the integrity of our buy signals — we score products objectively based on price history.

Acceptable use

You may use TimeYourBuy for personal, non-commercial use. You may not scrape, copy, or redistribute our price data or signals in bulk without written permission.

Limitation of liability

TimeYourBuy is provided "as is". We are not liable for any purchasing decisions made based on our signals. Always do your own research before making a purchase.

Changes

We may update these terms from time to time. Continued use of the site constitutes acceptance of any changes.

Contact

Questions? Email us at timeyourbuy@gmail.com.

Home

Home

How My Signals Work

Here's exactly how I calculate buy and wait signals, and how well they actually performed when I tested them.

How I calculate the score

Every product gets a score from 0 to 100. The score combines four things:

The score becomes a signal: STRONG BUY (79–100) · BUY (55–78) · FAIR (41–54) · WAIT (23–40) · STRONG WAIT (0–22). When I have less price history for a product, I show a lower confidence number so you know the signal is less certain.

How well do the signals actually work?

To check my signals, I set aside 20% of my data and ran the algorithm on it as if I were seeing it for the first time. A buy signal is "right" when the price doesn't drop sharply afterwards — meaning you didn't miss a better deal. A wait signal is "right" when the price does drop — meaning waiting saved you money. I measure each two ways: looking at the price exactly 30 days later, and looking at the lowest price seen during those 30 days.

Buy signals

A buy signal is "right" if you didn't miss a better deal — i.e., the price didn't drop more than 5%.

"Held throughout" is the stricter test — it catches cases where a cheaper price appeared briefly even if the price recovered by day 30. For comparison: always buying immediately would pass the strict test 50% of the time.

Wait signals

A wait signal is "right" if a cheaper deal appeared — i.e., the price dropped at least 5% (it doesn't measure whether prices rose).

"Dropped at some point" is more lenient because a price can briefly dip and then recover. For comparison: always waiting would catch a drop at some point 50% of the time, and random guessing ≈ 40%. FAIR signals (29,026 cases) had an average price change of +0.9% — neither a clear buy nor a clear wait.

Does my confidence number actually mean anything?

When I say a signal is "75% confident," is it actually right 75% of the time? Below, I plotted what I predicted against what actually happened. Dots close to the dashed line mean my predictions matched reality. Dots far from the line mean I was over- or under-confident.

On average, my STRONG BUY confidence was off by 2.6 percentage points across 12,477 cases. BUY was off by 1.2 across 29,629 cases. Both are well within the range I'd consider reliable (under 5 points). Based on my test set, looking 30 days ahead.

What about products that got discontinued?

My catalog is products that are still on Amazon today. Products that disappeared — often after a fire-sale discount — aren't in my data, which could make my results look better than they really are. To check, I compared two groups in my test set: products that are still being sold, and products that got pulled off Amazon afterwards.

The gap is tiny — under 0.2 percentage points for every signal. STRONG WAIT showed a slightly bigger gap (+0.3 points), but only 18 discontinued products fell into that bucket, so the difference isn't meaningful.

Where my signals fall short

• Discontinued products — my data is mostly products still on Amazon today. Products that got removed are underrepresented, so my success rates may be slightly higher than reality.
• Pricing assumptions — my test assumes prices tend to bounce around an average. For memory and storage products, prices follow industry-wide trends, which the test doesn't fully capture (the live system handles this separately).
• What I tested — the backtest covered ~10,000 products, mostly consumer electronics — a subset of the 88,000+ the site tracks today. Results apply to that cohort, not all of Amazon.
• Stable-price products — my results focus on products whose prices change at least a few times a month. Products with rarely-changing prices weren't included in this test.
• STRONG WAIT is rare in practice — real products seldom stack every bearish factor at once, so on any given day only a handful of live products carry a STRONG WAIT. The 84.4% figure is backtest-validated, not a live-catalog guarantee.
• Data source — all my price history comes from Keepa, which records prices only when they change. For products with stable prices, I have fewer data points to work with — I account for this by lowering the confidence shown to you.

Questions or feedback?

Email timeyourbuy@gmail.com. If something here is unclear or you spot a problem, please let me know.