What to do different ...
Your business is a product. You need to package it up as such, who is going to buy it and why? Are they looking to increase their client base? Looking to get hold of your technology? Looking to get hold of your skillset and knowledge? Looking to get rid of you as competition?
What should you invest in? When you planning to sell out from the beginning you think "in 3-5 years we are selling out" so do you invest in a huge expensive office? maybe not as likely than if you are going to spend the rest of your working life there.
Plan to be replaced. You focus on being easily replaced by new management, you should document all the processes and automate everything you can. Yes you should do this anyway but the level of investment will be different. Spend the time to document it so you can hand it over ... how else is someone going to replace you.
Hiring staff. If your here for a long time, you want good staff who will stick with you thruogh think and thin (at least I do) ... if your going to sell and be rid of them, then you might want people who will drive the business a lot harder.
Features of your product. Some features, functions or indeed the entire direction of the product suite may change depending on if your "trying to get into bed" with a prospective purchaser ... basically you want to complement them without competining with them. If you don't care about your competition and are doing it on your own for your customer base (like 37 signals) then you given them what they want, regardless of who else is doing what in the industry.
Then again if your trying to annoy the hell out of them so they buy you ... you would match their features and raise it to the next level.
Investment you take on. A VC wants to get their money back ... and soon ... this typically means trade sale or publically list. If you are happy to grow for the next 30 years, you may opt for a slower more natural growth and avoid the undue pressures.
Clients you target. I want long term, solidily growing clients that I can make a real difference through automating their business ... If I was selling, I would want more clients that I can sell lots to today so tha it looks better on the balance sheet.
R&D Strategy. This is a near and dear to my heart, you aren't going to build a huge production works, wrap processes and quality control around everything if you not going to be here in 5 years time. You going to get todays feature out the door today and rinse and repeat tomorrow. "Hang the long term consequences, we aren't going to have to deal with that".
Short term contract developers VS long term employees. Long term employees know they have to look after it and maintain it forever (like having a baby) but the contract just has to tick the boxes today and move onto the next job.
There are probably more but that will do to illistrate the differences.