The problem is two fold:
1) Legacy MIS' are a nightmare to migrate from. Legacy MIS' have horrible tie ins and proprietary data storage. Clever isn't open source. So despite it having an API the software is still proprietary. Clever doesn't solve this problem. Getting your data out of the legacy MIS will still be a nightmare and the resources required to move data from each MIS available in the market at the moment will be extremely costly.
2) Training to use a new MIS often outweighs the cost of the software itself. Because the training costs are so high any monetary saving from moving to lower cost MIS' is canceled out.
Clever isn't what the market needs. Y-Combinator doesn't have the resources to compete w/ Capita/Pearson/RM/Serco just to name the leaders in the UK market. Imho it is a naive effort unless they are going to sink a lot of resources in.
I do wish Clever the best of look but Y-Comb should know that this is a really mature market place and making software that's main pitch is "REAL TIME LEARNING" shows a complete misunderstanding of the market.