Combining both services solves both disadvantages! I use Dapper to create a high-level API that extracts the relevant portions of the webpage and this is fed into an Open Table as xml. The Open Table allows for detailed analysis and formatting of the data. The coding is minimal.
The only thing that worries me, is the reliability of Dapper. It's reporting server overload quite often. Is Dapper really the only visual html scraping tool around?