In the car industry big may be beautiful but is usually not enough - only huge will do and this was ultimately Rover's undoing.
While the Longbridge company could have stumbled on until 2010, it was not just a question of selling enough cars which fatally damaged it.
It also didn't make enough - and so did not enjoy the economies of scale enjoyed by its rivals, said Michael Wynn-Williams, a research analyst at Trend Tracker.
Mr Wynn-Williams worked out the optimum production for a car factory to be about 250,000 vehicles a year.
When the Phoenix Four bought the firm in 2000, they targeted a figure of 200,000 but never got close.
In 2002 it made 147,000 - far lower than the 177,000 produced by Honda at its Swindon plant, and dwarfed by the 257,000 built by Ford at its factory in Cologne.
Production at Longbridge continued to fall, while staffing levels remained roughly constant at about 6,500.
By 2004, the last full year of p roduction, output had dropped even further to 107,000, meaning each car built was ever more expensive to make.
Mr Wynn-Williams believed that the company could have limped on for a few more years, but also faced significant problems.
He said: "Only with new products can a company claim long term sustainability rather than long term survivability.
"But the the investment funds can only be generated internally with output approaching full capacity and economy of scale."
Companies producing high volumes of cars also benefit because their large market share denies economic value to their competitors, while research and development costs can be spread across a greater number of units - thereby encouraging greater research.
Large firms can also expand in steps, each of which achieves economies of scale.
Mr Wynn-Williams said: "Toyota for example can open a new factory with full scale benefits for a proportionately small expansion in total output.
"Meanwhile a 50 per cent leap in output at MG Rover, from a 200,000 base, would have meant an extensive reengineering of the only existing site at Longbridge, a genuine problem had that new medium-sized car ever emerged."
He added: "A large firm can save costs on designing new plants due to experience gained from previous projects.
"Finally larger plants output also brings their suppliers up to scale, with consequent cost advantages for both parties."
Rover enjoyed none of these. Alternatives, such as keep-ing extending life cycles of existing cars, were tried, with some success as the remodelled MG variants of the Rover 25, 45 and 75 showed.
The company also sought to keep afloat by practising cost recovery - where a premium brand can price its finished vehicle at a level that compensated for the cost disadvantage of not reaching the industry optimum.
To do this, MG Rover priced the MG ZT at £3,500 more than its competing Ford Mondeo 2.5 Ghia, which was needed to cover the higher production costs.
But with a a new car programme costing about $1 billion to complete, this did not raise enough money.
Mr Wynn-Williams said: "What this tells us is that MG Rover could have survived quite nicely turning out 200,000 cars a year, so long as it did not have to fund new car programmes on its own. The simple answer is that Rover should have abandoned mass production and focused on specialist production as proposed by Alchemy Partners in 2000.
"It would appear the entire company was predicated on a forlorn hope. Even if MG Rover had achieved its planned output, it would only have been enough to sustain the company in its existing state, but not enough to fund a new range of cars to compete with the latest competition."
While MG Rover spent a commendable 5.2 per cent of its turnover on research and development in 2003/4, in real terms this amounted to £90.5 million.
This was dwarfed by the £4.2 billion spent by Ford, the £2.3 billion of Honda and the £1.8 billion of BMW.
The company was also handicapped by the basic problem of not selling enough cars overseas.