Duplicate Content Filter/Penalty

What is the string length that flags the duplicate content penalty?

[Tigrou]

When pages on different sites have the exact same content they run into a penalty. Fine.

But then imagine you take affilate content and change every tenth word. Would that limbo under the flag for the penalty?

Or put another way, what is the length of the duplicate text string that flags the penalty? 100 characters? 200? 500? More? And does the filter work more strongly in the affilate rich sectors (e.g. hotels?).

Don't worry, I am not about to do anything evil like stealing someones site -- but a friend/client who does well on affilate work finally got hit by dup filter.

Peace,
Colin

[Never_again]

Tigrou:

I think you are talking about Googles dup. content filter. I think you would be more apt to get an answer if you posted this over on the Google forum.

[rogerd]

All SEs, not just Google, try to avoid duplicate content. So far, it seems, they are best at avoiding mirror-type dupes. I can't imagine a few hundred characters triggering a dupe filter unless that was the entire content of the page. Simple quotes can run longer than that.

The biggest risk, IMO, comes from nearly identical pages dominating the SERPs - even if the SE algo doesn't catch it, a disgruntled competitor will complain.

[Tigrou]

Thanks for the replies Roger and Never_again.

It is an interesting story; the only thing the guy did wrong was to jump from a less popular program to a more popular program. Well, more precisely went with the content of the new one.

He was making an extremely healthy living from his only site and is now is down to 20% of it. Ouch.