Skip to content
Use case

Incident exposure

A provider is in the news. The question is whether you are exposed to it, and through which vendor.

Incident response and securityProblems P5Available now
What changes

The problem, what happens today, and what we do instead.

During an incident you cannot answer "are we exposed to this provider?"

Today Email every vendor and wait days for answers that arrive after the incident has closed.

With Subprocessor monitoring Reverse lookup over the graph. One read against the closure names every vendor of yours that ultimately depends on the provider, and the path through which it does.

The provider is usually not one of your vendors.

Today The question gets asked at depth 1, answered "no", and closed.

With Subprocessor monitoring Exposure is a reachability question, not a contract question. A provider four hops beneath a vendor you signed with is still processing your data.

There is no record of what you knew and when.

Today Screenshots, if someone thought to take them.

With Subprocessor monitoring Every change event is timestamped with the moment we detected it and cites the source URL as published. The history is the evidence.

This is the demo that sells the product, and it is only answerable because the graph is held in reverse. Walking forward from each of your vendors to see whether a provider appears is a walk per vendor. Holding the closure means the answer is one indexed read, which is the difference between an answer during the incident and an answer after it.

The sentence you need to be able to say

Four of our vendors reach this provider. Two at depth 2 through their hosting, one at depth 3 through a support tool, one at depth 4. Here are the paths and the pages they were disclosed on.

What reverse lookup returns

The public version of this page is at /companies — pick any company and follow the reverse-dependency link to see which of the disclosers in our dataset reach it. Inside an account, the same query runs against your own inventory instead of ours.

Why page-diff tools cannot do this

A page differ holds one URL and its last body. It has no graph, so it has no reverse index, so the exposure question has nowhere to be asked. The recursion is what makes the answer possible, and the recursion is the part that is expensive to build and cheap to share once built.

Reverse lookup is P1 in the spec, and the public reverse-dependency pages ship now. We label roadmap items as roadmap items rather than showing them as delivered.

Start with one vendor. Free, and no card.

Search a company, follow it, and see the chain beneath it. Upgrade when one vendor stops being enough.