Let me put smart contracts' strengths and weaknesses in three lines:
Smart contract is a technology that allows us to create a bounding contract in a blockchain.
It is bounding: when done properly, it can not be broken.
It is in a blockchain: we can not make it bounding if part of it happens outside of the blockchain.
So - what can we do with smart contracts?
First of all (apologies to all people that will call me scamer
this is gambling.If money is in a blockchain, it is bounding and you can play... Looks like a sandbox,
doesn't it? Nevertheless, it is a multi million business. All I did so far in this
direction was refurnishing an old idea of "King of Ethereum" (I added countries,
and option to create/destroy them) - and I already have haters
Here is the link: http://duke-of-ether.com/duke_of_ether.htm
Second, services that serve contracts. Being transparent, contracts can not do certain
things, like producing true random values, and being simple (complexity costs) they
can not parse Internet sites looking for info. So they need services like stoke quotes
or random number generators. Oraclize holds this niche and I think more will come.
Third, this are exchanges of all kinds, from Ethereum tokens (lucky us, we do not have
to leave the blockchain to trade them) to BTC/ETH/USD/Whatever conversions that have
huge demand but no clean in-blockchain implementation so far. And Out-of-blockchain
implementations are (as usual) clumsy, expensive, ask you for ID, photo with credit card
and neighbour's cat and to kiss. I think the next Vitalik Buterin will have to create
the between-blockchains protocol.
Fourth (finally!) it is contracts in the original sence of this word. Fund rising,
fund management, multisig wallets, living trusts and all other ways to keep money and
terms of spending together. There are some severe limitations compared to real world's
contracts that are solvable and therefore, someone will provide services soon.
Example: I can create a contract saying "if I do not postpone the action, then
I am dead; and if I am dead, give my money to this Ethereum address". But I can not
say "give my money to Maria-Louise", not if I do not know an address of her wallet.
Also I can not say "build a nice garden where the ugly Central Park used to be" -
this is an off-blockchain action.
However, imagine that a large and influential law firm start offering services
of "helping smart contracts to fulfil off-chain actions". It is going as safe as
their reputation, which might be good enough, especially if a contract hires
a second company to do auditing.
Fifth (almost same as fourth actually): we can expect that number of off-chain
companies will create standard contracts bounding them to provide standard services,
from pizza delivery to investing into Hyperloop. The point is that if their
services are standartized, there is no need to sign contracts on paper: provided
the government makes this sort of services legally bounding. Or, as I mentioned above,
it it is a mater of reputation.
To summarize, smart contracts have to find their way off-chain. To pay for services
that are not limited to taking money from one place and sending it to another.
There are two ways of doing it.
First, more and more things will be available through smart contracts. Say, a Univercity
can create a smart contract allowing you (via your smart contract) to pay for your
kid's education.
Or imtermediaries, companies receiving orders from your smart contracts and carrying
them in the "real world" will do the "off-chain" part of job.