This is something I don't see in the .com and tech world. Why not have multiple brands that separate levels of service, instead of having the same service with different subscription levels for features.<p>Using flickr as a example. They offer free and pro. Instead, why not have two web fronts (both using the same backend still). One is called Flickr and would be free, but the other would be Blingr which is paid and looks completely different, a new brand.<p>A good real world example is Toyota + Lexus (imagine on the backend it's the same people working).<p>Why do this?<p>It's easy to implement, and you can really tap in to the folks who have ego and love exclusivity and want to spend. Plus it's like having two separate companies/fronts now.
Counterpoint:<p>It took millions of dollars to launch the Lexus brand. Probably hundreds of millions.<p>When you divide your brand awareness efforts into two distinct parts, you'll effectively cut your marketing budget in half. One of the biggest hurdles new companies face is getting people to know and trust a new brand. Getting them to give you money is even harder.<p>Admittedly, like Toyota and their existing dealer/repair network, you get some of the "brand split" at no significant extra cost (servers, bandwidth), but the most expensive thing is reaching customers and making them aware of you.
I don't think it is a good choice for the generic web startup with 2~4 levels of activity. If you're pitching the different levels to totally different markets, I could see it. (Example: Heroku versus Big Freaking Enterprise-ku.)<p>For startups that don't have that severe of a segmentation between the needs of the different tiers, it sounds like a way of time and money. Your maintenance needs just increased proportional to the number of brands. (Easy to implement? <i>gah</i> I shiver at thinking about it.)<p>Want to do an A/B test? Enjoy writing it four times, and potentially ending up with 4 different optimal choices that you have to maintain for forever.<p>You also split your link equity -- 4 domains with 1,000 links split between them won't rank nearly as well as 1 domain with 1,000 links pointed at it. Because search is a Winners Take Most game, it won't end up with each of the domains having 25% of what they could have, it will be closer to .01% of what they could have.
Well for starters it would make conversions from free to paid that much more confusing and unlikely... I have to switch to a different website to get these other features???? What about all my friends on this website? Uh... wait, what? They're the same people? I'm confused!
The primary purpose behind different Brands is to serve <i>different markets</i>, which usually involves providing multiple levels of service.<p>To skip from A to C, while trying to capture the same market, is largely a recipe for confusion, and extra unnecessary costs (particularly in marketing / positioning).<p>To complete your example, people who buy a Lexus were about to buy a Mercedes or BMW, not a Toyota.<p>To add another example, people who fly Jetstar (low cost Australian airline fully owned by premium airline Qantas) weren't about to fly Qantas. They were going to holiday within driving distance of home - the new brand opened the bigger business up to a new market.
The goal of freemium is the upsell - "hey you liked my x, try my y!" - which requires a tight association between your products. You upsell in the same market. You do brand differentiation in different markets.<p>Basically, if you can imagine someone deciding to buy one product instead of another, they're in the same market. Nobody's sitting in the lot thinking "hmm, Lexus or Toyota?" but they might be thinking XLi vs GLi. That means you can convince them to get the more expensive one.<p>Some software does do brand differentiation, though. Off the top of my head, Hiveminder and Request Tracker - one is personal, the other enterprise, both made by Best Practical. Hiveminder is actually freemium itself, and RT has support & hosting tiers.