For a technology invented over 10 years ago it does seem a solution looking for a problem. Ambitious white papers and press releases don't matter if your tech doesn't do anything.

Blockchain too often seems a single, excessively complicated, technical approach to solving a particular type of problem, and really shouldn’t be on the radar of people who don’t build systems themselves. 

Non-developers should restrict themselves to getting excited about actual real world problems getting solved rather than how they might be solved! As it happens, real world problems are usually solvable using simpler and cheaper methods. A trusted central authority works just fine for most if not all situations. There are ways of adding tamper resistance without going full blockchain.

Another key issue- it is one thing to establish the identity of parties involved in, say, transactions. But that is of no use if the content is invalid, fraudulent, or incomplete. Blockchain does nothing to validate content so the old adage "Rubbish in rubbish out is as valid for blockchain as any other technology.

Read MITSloan Management Review "Before jumping on the bandwagon, companies need to carefully consider how ledger technologies fit into their overall strategy."