You don't need to be a jerk. In fact, if you're a real jerk, it's probably not going to be half the product you want it to be.

Here's the trick. You've hired smart people, right? These folks aren't just grunts (not that anybody is); they're creative individuals who happens to be highly skilled in areas that you value. And you're paying them. So given that, if you don't listen to them, or if you make them feel like they aren't valued members of the team, simply put, you're wasting money.

Listening to them, however, is not the same as acting on what they say. Widening the gene pool of ideas is smart; losing focus is not. You know what you want to see, and you know what the end goal is, and your role as a manager is to motivate your team to aim for that. You need to guide them towards the finish line, while simultaneously keeping control over the end product and giving them a sense of ownership. After all, if someone feels like the product is (in part) their baby, and if they have a sense of shared responsibility with the rest of the team, they're going to work harder, be more free to come up with potentially groundbreaking ideas, and be more likely to come up with something that makes people sit up and take notice.

Some people confuse leadership with being a jerk or making arbitrary decisions. That kind of abrasive approach is far more likely to result in a failed project than a successful one.

On a complex creative project like yours, it's far more important to make sure the lines of communication are flowing and the right people are talking to each other. You can't micro-manage everyone's work, but if you have good people on your team and your people are talking to each other (and listening to each other) on a weekly or even daily basis, a lot of problems will sort themselves out.

Make sure everyone's work is reviewed on a frequent basis. It's far less painful to tell someone they've gone down the wrong path and wasted a week than a month. The shorter you can make that feedback the better. (Take a cue from Pixar, where they review work in progress daily and have a culture of constructive criticism.)

The key thing you can do as the team leader is to build a bubble around them so they can do their work uninterrupted. This can mean anything from arranging for food and snacks to reducing unnecessary meetings and outside distractions to making sure everyone has all the computers, software and tools they need to do their jobs. Work hard to eliminate low value work such as voluminous requirements documents that no one reads or long boring reporting meetings where everyone is tuned out with their heads in their laptops.

Sometimes to do these things effectively you need to stand your ground, but there's no reason to be a jerk about it.

No. Oh god no. You have to get things done a certain way, and that might mean cutting out X, Y, and Z that one or another developer likes working on, and rallying people to work long hours, and several tedious i-dotting and t-crossing iterations at the end, but there's no need to be a jerk in the course of it. Only jerks will tell you otherwise.

I dont think you have to be a jerk to have a successful team. I do understand what your alluding to. I find that being a successful team leader you have to be very decisive. This sometimes means giving the opinionated but talented team member a firm no. Sometimes this can bruise egos or offend. But it has to be done for progress to be made.

You don't have to be a jerk, but you may find it difficult to be a friend of anyone on the team.