Prospective new client it by webspam looking for new resource

[tcmktg]

Background: 
Prospective client recently hit by webspam update. (I have verified hundreds of low-quality links, porn links, backlink exchanges etc.) They want us to step in and remove bad links and start over.

Question:
What is the best way to examine all the links to determine which need to be removed? We can  create the report from open site, but how can we identify the bad links? Here are the site metrics. 5000+ linking domains, so in this example we need to research the 5000 links, and possibly send notifications to thousands of webmasters to remove the links?  Open site states about 25,000 total links, but root links are shown below. Yikes.

Domain Authority 75
External Followed Links 112,000
Total External Links 115,000
Total Links 150,000,  
Followed Linking Root Domains 3,900
Total Linking Root Domains 5,300
Linking C Blocks 2,700

[tcmktg]

Thanks so much, Ben.

You're right, a new domain is not an option. They're a big company.

At this point, I'm not sure what to charge. This is going to take a lot of time and I'm not sure if I even want to get associated with it now. Seems like the old seo should do it and provide documentation it's done.

Anyone find a service to do this?

[BenFox]

So I'm going to guess launching a new domain isn't an option?

You can use Excel to segment the linking root domains (LRDs) by those that are probably directories - those that contain terms like "directory", "URL"  and "submit" in the domain. There are a couple of ways to do it but just as an exmaple if you enter the following into Excel it will tell you whether the URL in A2 contains the term "directory".

=isnumber(search("directory",A2))

It pays to scan over these links and see if there are any decent sites that you've bundled in with the bad but mostly I'd imagine they will just be directories.

You can do the same thing with adult terminology to find those types of sites. The method is by no means perfect but it is quick and easy.

Possibly the easiest way to spot low quality links is looking at the C-block. If you've got 50 links from the same C class IP then it's either a bad link network or a blog service. You'll spot the difference pretty quickly although if you have several clients wanting this service then it might be worth excluding the IP ranges of the major blog services in your spread sheet. I don't think you can currently get IP from OSE but I'm sure you'll figure out a way.

That's all that I can think of off the top of my head but I'll add more if I think of them.