^Yeah this is the question I would like an answer to.

Does it depend on how well previous instalments have sold in the series if its a sequel? Say a new game (non-sequel) is made. How many sales is considered to be successful? Say a total of  500k or one million copies sold world wide? Sales of games are definitely not in relation to review scores some of the better games have struggled to sell 2 million whereas shovelware games have sold a lot more. Say a Halo game, MGS game and Mario game have to sell 7 million. Both GT and Final Fantasy games must sell over 10 million copies.

Well, I think it depends on a lot of factors really:

-past series performance
-public hype
-cost of development

Some games might be happy to sell 250k, while the big series are generally looking for multi-million sales.

The amount a game needs to sell to be successful depends on how much it costs to make and its history ...

A cheap new IP costing say $5 million, might only need to sell 200k copies worldwide. While a game like Halo needs to sell 6 - 8 million (lifetime) to be considered sucsessful due to its higher cost AND fanbase expectations ( ie Microsoft might of needed to sell 800k copies of Halo 3 to break even ... but if the game only sold 1.5 million copies it would be considered a failure even though it generated profit ... while Folklore or Warhawk might just break-even for SONY or make a slight profit and be considered a success ...

Personally I believe every developer seeks to sell a million, even though in many cases its not going to happen, but I've never seen anyone call a million seller a failure.

For the developer the only measurement is profit.

In general terms I think you can say that a PS360 game over 1 million and a Wii game over 500.000 can be considered a profitable and therefore successful game. Of course there are the bigger budget exceptions.

Success in terms of expectation is something completely different.

Let's take MGS4. It's not Konami's fault that the PS3 is not (yet) the success it was supposed to be. Yet a lot of people expect the game to sell millions of consoles. In that regard MGS4 can be considered a failure by the fans, while still be very profitable for Konami.

The big names like Zelda, Mario and FF of course have a history that has to be matched in order to be called a success. If Mario Galaxy doesn't outsell Sunshine it will be considered a failure, same with FF13 vs 12.

The questions are : is it a new or an old IP ?
What console is it on ?
How big the hype has the game before release ?

For a 360/PS3 game it will allways be bigger the needed copys to be sold , for the game to make any profit , while the Wii/DS/PSP/PS2 versions usually are much lower budget games ...

i only see it as being a success if it moves enough to cover marketing/dev cost

 I agree that for the developer the measurement is profit.  But, I think your numbers are too high.  Those would give a fairly large profit.  Rev. of $60M/25M, minus store cut of $5M each, other overhead of say $5M, minus production costs of $20M/$10M= $30M/$5M.  And my numbers are being very generous as well.

 Especially a new IP that is being build up, Wiki & Zack, NMH, to get the name out there with high quality means lower sales okay if sequel is any good.

Enough copies to make a profit.

That's pretty much all there is to it. Puzzle Quest, a DS/PSP game sold around 500,000 copies on their respective systems, and has sold around 150,000 on XBLA. Guess what? It saved Infinite Interactive, and made them a pretty penny. Guess how many copies Killzone 2 is going to have to sell before the developers see any return on their investment?

 Ultimately, it comes down to the Cost of Goods Sold (R&D, Development, Distribution of Content), versus Gross Revenue from Shipped products, or Sales via Steam or a DLC Service on a console. There is no "solid watermark" like there used to be. 10 years ago, 100,000 units LTD would of been enough to be considered a succuess. Some games, like a Final Fantasy needed to make more due to bigger dev. teams, and costs of translation to the US/PAL markets.

As for previous installments vs. new ones, it really just depends on how big the developer is wanting to go. From what I've seen, Halo 3 didn't cost a whole lot more to make than Halo 2 (again, this is just speculative, but the developer teams and length to completion were kind of similar), so the need of H3 out-selling H2 wasn't as huge. But for a game using a bigger dev team, definately needs more (like Metal Gear Solid 4).

Right now, your probably looking at, to make a profitable return:

XBLA/PSN/WiiWare game:

100,000 Units Sold

Small Budget Retail Title:

200,000 (Wii), 300,000 (X360), 350,000 (PS3)

Middle Budget Retail Title:

350,000 (Wii), 475,000 (X360), 550,000 (PS3)

High Budget Retail Title:

500,000 (Wii), 800,000 (X360), 950,000 (PS3)

Add 10% Total LTD sales for each multi-platform system. Subtract 10% if good DLC is available. I tend to think that a game like Call of Duty 4 made a profit on the 360 version alone, multiplied many times, and the same for the PS3 version. Likewise, Mario & Sonic at the Olympics probably made Sega just as much.

And those are just really rough estimates. The biggest thing is finding what a game cost to make (such as Gears of War's $10m USD), and then understanding the developer makes about 60% of the retail cost of the game.

Wow thanks mrstickball ... thats actually less/game than what I was expecting :?