Can you find success by copying a name or category prefix from a big Web 2.0 site or company? I wanted to find out.

Here's what I did. (Warning: This may get a little boring/technical/nonsensical. Go to "Findings" below the table if you're not interested in my process.)

1. I found a list of the top (approximately) 1000 Web 2.0 sites and companies compiled by Seth Godin. The list is ranked by Alexa - as good a source as any for my gauge of success.
2. I scanned the list for prefixes, words, numbers, and letters that were at the start of the website name. (This didn't have to be an actual word - it could be a single letter that is meant to be sounded alone like in ebusiness, or numbers, like "321contact.)" This was not a scientific process. I used a spreadsheet and sorted by alpha to locate groupings.
3. For each common prefix I counted up the number of "hits" there were in the top 1000 and the top 100 (the latter number being a somewhat arbitrary measure of success.)
4. I measured the ratio of companies with each prefix in the top 10% to the companies in the top 1000.
5. I then subtracted out the "initial mover" that brought about the trend in usage (if one existed in the top 100) - assuming that if there is at least one in the top 100 that they are the attracting factor for the term. (I know it isn't really true in all cases, but I gotta start somewhere.)
6. I measured the ratio of copycat prefix users to see how effective the names have been at drawing traffic.
7. I completely ignored everything about the companies, websites, users, and any external factors that might be influencing one website to draw more traffic than the others with the same name prefix.

Here's the resulting table:



Findings:

1. About 19.5% of the top 1000 Web 2.0 sites fell into a recognizable prefix/first-word usage group.
2. 20% of the top 100 websites were a part of the prefix groupings
3. The corresponding success rate (for being in the top 10%) for all companies in the prefix groupings was a approximately 10.26%, meaning that those companies not in the prefix groupings had a success rate of just below 10%.
4. But when the First Movers are subtracted and we analyze only the copycats the success rate is reduced to 4.1%, implying that non copycat names have an approximately 11.4% chance to succeed. This is an increase of over 180% achieved just by not following the prefix groupings of other top 1000 sites.
5. The best success rates for pattern matching names are for beginning with the word "news", any grouping of numbers, or a variant of the word "You" (as in you or your). And even these success rates aren't exactly awe inspiring.
6. Personalization is well represented in the top 100 (just factoring in the prefixes - there's probably more that hide the personalization elsewhere in the name) There are multiple examples of each (I, My, You) prefix in the top 100. And yes, not every "I" refers to personalization - but I'm going to stick by my story.
7. Success rates for companies that have first mover status for names and have attracted copycats in the top 1000 is 40.00%. (This is mitigated by the fact that we assume the top-ranked name is always the first mover - something that is not always true.)

Conclusions:

• Generally speaking, copycat naming does not work.
• The impressive success rate for first movers with copycats likely isn't a causal relationship (e.g., naming with a new prefix won't get you a 40% chance of being in the top 100) but it certainly makes the case that starting trends is more likely to get you attention than following them.
• More research in this area would be absolutely fascinating for me - I'll be looking to write a deeper study for publication in the near term.

What do you think? Are the outcomes as you thought they would be? Is my logic horribly flawed?

Will you read my amazingly dry research report when I have the time to publish?

Tate Linden Principal Consultant Stokefire Consulting Group 703-778-9925