The first reaction is, "of course I need a good idea."
Yes you do, but not when you are first thinking about that new business. 
Then where do I get that great idea?
You get it from your customers.
How do I do that?
You do it during what Steve Blank, a professor at Stanford University, calls the customer discovery phase. This occurs when you interview a large number of potential customers about their burning problems in your business area of interest. You are looking for what he calls their migraine problems.  If you can find a potential solution to one of those problems, you create what is called a value proposition, the means of bridging between the problem and the problem solution by means of your proposed product or service.  
You then embody that solution by means of creating a Minimum Viable Prototype or MVP which can be a very rough solution model. By demonstrating this MVP to potential customers you can verify that it will satisfy their need.  If the customer is not fully satisfied, you pivot and change the MVP to fully meet their needs.  It is this iterative effort of modifying the MVP that allows you to home in on the solution that really fills the bill for the customer.  When you have verified that the MVP will solve the customer’s problem, you have completed what Steve Blank calls the customer validation phase.  After verifying that the customer would be willing to pay an amount for the problem solution that covers all of your costs and profit, you then have a great idea to start a business. 
This process allows you to create a business that has assured customers and expected profitability, thereby minimizing risk and providing an increased probability of success. 
More next time on the details of converting that customer-approved big idea to a product/service that you can take to market.

You Don’t Need a Good Idea to Start a Business