From Holgate's table, it appears that hardware was increasingly the driver of Autonomy's growth in the runup to its buyout by HP.
Are you sure? It looks like the percentage of growth "lost" if you remove hardware is lower in 2009-10 than in 2008-9? So that's a decreasing effect.
Also there's a really important question here. How do you define which sales are growth? Because if you can claim all the hardware sales are in the "growh" bucket of Autonomy's sales, then you can make it look like Autonomy's sales were 10% hardware - but it could be that their total sales were 2% hardware and he, or HP, have selected all the hardware sales as "growth"? Unless there's a standard definition of growth use to break down sales - which seems problematic. Surely growth is just the difference between one year's turnover and the next?
But this looks pretty bloody shaky, if it's all the evidence HPE have got. We've got the one self-reported case of "fraud" from the US, with the witness who got immunity if he would allow his testimony to be agreed in advance - and that's not at all dodgy oh no!
But other than that we're talking about a moderate percentage of sales being entirely legally booked, but HPE being sad because they're low margin hardware, not high margin software. But even on their definitions of the accounts (which I would want to see confirmed independently) is at worst is reducing 47% growth to 36% growth. And that's on sales figures - so the effect on growth of profits is going to be less - given the assumption that this hardware is low-to-no margin and most of the profits come from the software.
Split between software and hardware revenue and margins should have been the first question asked in due dilligence. And if Autonomy lied about that - then HPE would have a case. Was there any evidence of that? I don't remember it, and in fact remember the opposite. I'm going to be fascinated by what the judge says at the end of all of this.