5 ms·
> But that's not really how people consume media. Most people don't spend time on thousands of websites. I found it difficult to read past that bullet point. I
by hn_throwaway_99 9y ago
> But that's not really how people consume media. Most people don't spend time on thousands of websites.
I found it difficult to read past that bullet point. It's such an obviously stupid mischaracterization of what "the long tail" is that it makes the rest of the arguments suspect.
- arkitaip 9y agoIt gets dumber, yes.
- gt_ 9y agoI am not familiar with the "long tail" book or the particular internet theory, but I used to work for a company which owned an actual windsurfing website. They only sold advertising to what they called "endemic" advertisers, which means they are of the industry that supports the niche; in this case windsurfing. They dealt directly with the advertisers. The reasoning was that "non-endemic" ads would sour their allegiance with the niche audience they served. I would be interested in some research on this theory.
- rjbwork 9y agoIt may be a reasonable theory. But there are lots of ancillary industries that help support the niche - boat manufacturers, plastics/cloth companies, clothing and misc gear companies that are native to windsurfing (such as brands like Revive, Vans, Sector 9, etc are to skateboarding), and dozens of other related brands. You may also be hurting your income if you are only doing direct buys/run-of-site buys, since you don't really know what the market value of placements on your site are, you only know what the few brands you consider "endemic" value your site for. I could likely come up with a few other reasons given the time, but I think this makes the point decently. EDIT: Also, remember the incentives of the many mainstream high-traffic sites who are pushing this narrative about advertising.
- autokad 9y agolong tail? depends on what you are talking about. previously, companies couldn't focus on the long tail because there were many limitations. For example. say you want to sell sodas, there are only so many types you can fit on a truck / shelf, etc. Then came the internet, so the idea was you could chase the long tail and sell xyz soda that only .01% of the market would drink. the problem is, people tend to do things that fit Dirichlet distributions, and a consequence of that is double jeapardy: https://en.wikipedia.org/wiki/Double_jeopardy_(marketing) https://en.wikipedia.org/wiki/Double_jeopardy_(marketing) the part in the link thats important is that brands with lower market share aka 'niche' products, people also tend to buy less frequently. I see this all the time dealing with people. someone will be like 'i love x, its the bset, i dont understand why people dont love x so much, people who love y aren't dedicated as i am'. (y being the mass liked thing). but when you crunch the numbers, almost always people who like y consume y more than the zealot x. even more damning, the majority of the time the x zealot actually consumes y more than x.