You need to first pin down what kind of scrub. Do "they" want every field scrubbed or is it only certain fields? Is it the work from every department or only one department?
That information helps you to scope the size of the scrub.
What types of errors? Are they something that can be spotted in a pivot table exercise as Truffle mentions, which would pinpoint anomalies, with follow up on errors spotted, or will you need to go back to the files to check source data? If the data made it through submission, it seems likely your issues are differences from what is in the file, which a pivot table won't identify.
Start as I mentioned, narrow down the problems to see if you need to essentially check every field in every file, or a subset.