Editorial Standards

These are the rules we apply to every deal before it appears on back9.deals. We publish them so you can hold us to them.

The core test

Every deal must pass a single threshold before we consider anything else: would we tell a friend? Not “would we link to it for a commission,” not “is it technically a sale.” Would we text this to someone we play with and feel good about it the next time we saw them?

If the answer is no, the deal doesn't run. Full stop.

Genuine savings requirement

We verify the claimed discount against price history before publishing. A deal is only listed if the sale price represents a meaningful reduction from the item's typical street price — not the MSRP that nobody pays.

The minimum discount threshold varies by category. For equipment, we generally require at least 20% off. For apparel and accessories, 25–30%. For green fees and travel, we evaluate against published rack rates and look for at least 15% off.

Deals that don't clear the threshold may still appear if the price represents an all-time low or the item is genuinely hard to find discounted.

Confidence scores

Each deal carries a confidence score between 0 and 1 representing how certain we are that the listed price is accurate and the discount is real. The score reflects:

• Price history depth — how much historical data we have for this SKU.
• Retailer reliability — whether this retailer has a track record of honoring listed prices without bait-and-switch at checkout.
• Deal freshness — whether the price was verified within the last 24 hours.
• Source quality — manually curated deals score higher than automatically ingested ones until verified by a human editor.

Deals with scores below 0.7, or deals that haven't been verified in more than seven days, are flagged with a visible price warning. We'd rather tell you to check before you click than let you get surprised at checkout.

Retailer standards

We only publish deals from retailers that meet our baseline requirements:

• Established e-commerce presence with a clear returns policy.
• No pattern of bait-and-switch pricing or fake MSRP inflation.
• Responsive customer service (we check this by looking at public reviews and our own experience).
• Secure checkout.

We maintain a running list of retailers we will not work with. Retailers on that list are excluded regardless of how good the deal looks.

Affiliate relationships and editorial independence

We earn commissions on some of the deals we publish. See our full Affiliate Disclosure. The relevant editorial rule is:

• We do not accept payment to feature specific deals.
• Deals are not ranked by commission rate.
• Deals from retailers with no affiliate program appear alongside those that do.
• A deal we earn nothing on that passes our threshold will always beat a deal we earn from that doesn't.

The only honest concession: when two deals are otherwise equal, we may prefer the retailer with an affiliate relationship. We're a small operation and need to keep the lights on. We document this here rather than pretending it doesn't happen.

Corrections

When we get something wrong — wrong price, wrong discount percentage, deal that expired before we caught it — we correct it promptly and mark the deal accordingly. We don't delete errors silently.

If you spot an error, use the Submit a Deal form with the note “correction” in the subject, or email hello@back9.deals.

Automated vs. manual curation

Some deals on back9.deals are ingested automatically from price monitoring tools and partner feeds. These automated deals are clearly labeled with a source badge and start life as drafts — they are held for human review before being published.

No deal goes live without a human editor signing off on it, regardless of source. Automated tools help us find candidates faster; they do not replace judgment.

Community votes

The upvote system reflects reader interest, not editorial endorsement. A highly-voted deal has resonated with readers; it hasn't been vetted any differently than any other deal on the site. Use votes as a social signal, not a quality guarantee.

We monitor vote activity for manipulation. Votes that appear coordinated or automated are removed. The vote log is hashed and rate-limited to one vote per reader per deal per 24-hour period.