Beyond that, it could be that Google wants to show a business page and an informational page on this SERP, and both of those pages are ours. So that doesn’t necessarily mean there’s cannibalization. We’ll talk about how to disambiguate this later.
First let's talk about the classic case. You'll see this chart of a classic, very clear, very obvious case of cannibalization.
Classic eating competition case
You'll see a lot of tracking software rankings on the chart. You can see the time and day along the bottom axis. Seeing a ranking like this, we obviously want to rank as high as possible and close to number one.
Next we see two URLs, color coded, here green and germany mobile database red. When one of them ranks, the other is forgotten, not even in the top 100. Only one is present but not at the same time, they replace each other in the SERP. When we see this kind of behavior, we can be pretty confident that we are seeing some kind of cannibalization.
Less obvious cases
Sometimes cannibalization is less obvious. A great example I found recently is if I searched Google for a place name like Naples, I saw Wikipedia ranked first and second. The number one Wikipedia page was for Naples, Italy, followed by Naples, Florida, ranked second.
Now I don't think Wikipedia is cannibalizing in this case. It just happens that... the nature of Google has decided that this SERP is ambiguous, and that this keyword "naples" needs to serve multiple intents, and Wikipedia happens to be the best page for two of them.