User Tools

Site Tools


threat_modelling

Threat modelling: risk identification and prioritisation

This section sets out a basic approach for contemplating cybersecurity.

The aim is to stop you from running out and buying whatever product some shiny suited salesperson might be promoting, but rather to think about your security needs holistically, so that you spend your time, money and effort where you can get the best results.

You’ll sometimes see this described as “threat modelling” or understanding your “risk scenario”.

Key points

What are trying to protect?

The reason you do this is that, without knowing the threats against which you’re trying to protect, you don’t know what mitigations you need to have in place. And, since you probably can’t do everything at once, you’ll need to understand the greatest threats you face, and so which are deserving of the greatest attention, and what measures are “nice to haves”, which could be done at some point in the future.

Let’s consider some different examples.

Why you?

A normal domestic conveyancing transaction may not be of particular interest to an overseas government’s intelligence service, but may be very much of interest to someone hoping to steal the buyer’s funds.

If you are working on litigation against a government, might that government have an interest in trying to understand your case before it is formally presented? One would hope not, of course.

Could you simply be the weak link in the chain, which someone wants to break to gain access to the information which you hold? Some useful insider information which might affect a share price, for example, or designs for a new or improved product?

Do you hold information about someone which might be valuable from the perspective of blackmail? Or of interest to the media?

Information useful to a competitor of a client?

Or could it be that the information you hold about a client might be useful in some other attack – someone wanting to get into the client, to exploit what information they have, and getting names and addresses and contact details from you is an easy way to do that?

Or are you just generally concerned about holding up the confidentiality of your clients’ matters, and ensuring that clients can communicate with you in a reasonably secure manner, to take advantage of their right to seek independent, expert legal advice? Threats to this might be more casual — someone reading over your shoulder on a train, for example, or listening in on a phone call you are having in a public place.

Create an information asset register

In a spreadsheet, list every device and service you have which stores data — computers, phones, external hard drives, USB keys, servers, online services, and potentially even printers and scanners.

Against each device and service:

  • identify what data might be stored on it
  • state who is permitted to access it
  • describe the risks that it might face (for example, that you might lose your computer, leading to someone else getting access to the data on it, and you being unable to work on client files because the only copy is on the computer)
  • describe the security measures currently in place to protect against those risks
  • if relevant, state who is responsible for keeping it updated and ensuring it is wiped or destroyed properly at the end of its life

Keep this up to date, amending it as you add and remove devices from your firm.

Who is your threat?

Is your attacker motivated, and focussed on you? Do they have lots of resources at their disposal? If so, chances are you are going to need substantial security measures, and probably professional assistance.

Perhaps you are just of passing interest and, if you have “good enough” security, the attacker will simply find another target — a less secure law firm, more vulnerable to their attack, for instance.

Perhaps it is not even an “attacker”, but rather a fellow commuter, or someone else present in the place you are working. What about private companies tracking what you do online? Are you happy if the operator of your favourite coffee shop’s Wi-Fi network is keeping an eye on what cases or statutes you are researching, or sites you visit, or even your communications with clients?

How are you vulnerable?

Once you’ve identified why you might be of interest, and who you might be defending against, the next step is to identify how you might be vulnerable.

You might find the European Union Agency for Cybersecurity's threat report to be useful in identifying key threats.

Prioritise your response

Chances are, you'll have quite a few risks on your list, so you'll want to prioritise your approach to tackling them.

A simple means of prioritisation is by multiplying the likelihood of a bad thing happening with the impact if that bad thing does happen.

Something which is likely to be exploited, and which would cause a high level of harm, is a greater priority than something either unlikely to happen, or which is unlikely to cause much harm if it did happen.

Bear in mind that some security controls are appropriate for highly confidential information, but some are less appropriate for less confidential information — the likelihood of harm, or the severity of the harm, does not justify the intrusion or inconvenience. There comes a point at which providing security makes it more difficult for the client to work with you, contrary to their best interests.

Write it down

You might find it useful to write this down, so that you have a register of threats and risks, with reasons why you have included, or excluded certain things.

Make this a regular thing

You probably want to make this appraisal a regular exercise — perhaps yearly, or even more frequently, depending on the likelihood of a risk arising, and the severity of the impact if it did.

If you think you fall into a higher risk category, it is probably something you’ll want to do even more often than that.

If you have a compliance calendar, to help you meet your numerous regulatory obligations, make perhaps a quarterly or half-yearly action to review your risk categorisation, and your security measures.

threat_modelling.txt · Last modified: 2019/08/15 15:07 by neil