I am not sure when Spark management 'decided' that 10% was a reasonable ROI to aim for when running datacentres. Now I have figured out that ROI as defined by Stefan and IRR imply the same growth rate (post 2434), it could be that Stefan thought he could make an argument for 10% ROI 'sound convincing', and that the downstream consequences of that decision saw an IRR of 10% naturally fall into place.
I see on slide 23 of the result presentation that Spark is targeting 'data centre revenue growth' of 15% over FY2025. That is all very well. But there is a big difference between targeting 15% growth for FY2025 and 15% growth for 25 years! I haven't checked your figure of CAGR of data revenue between FY2022 and FY2024 being 31% Bull. But in the early stages of pioneering a new market it is quite believable. What is not believable is that such a growth rate will continue as the market expands. That 'super high growth rate' was only possible because of a low starting base level. Even a growth rate of 15% for 25 years yields scarcely believable valuations based on where we sit today, unless you 'discount those future valuations back'.
The problem with valuing datacentres inside of Spark is that you are trying to measure a 'growth asset' inside of a company that is normally valued on the basis of the income streams it produces. That doesn't make a happy melting pot for valuation purposes.
You make a good point about valuing the up-selling of 'package deals' for other services that might interest data centre customers as well. I am not sure how I would reasonably measure that though! As a conservative assumption, it might be best to think of such extras as a hook for getting new customers into a Spark data centre in the first place? So you might model it as increasing the growth rate of new customers signed up, rather than trying to put a dollar value on the extras.
SNOOPY
Bookmarks