Best Practice Checklists
An exclusive publication of

Essentials every leader should understand about IT
Issue 5 – Authentication Security, Network Traffic & Activity Analysis
Authentication Security & Activity Analysis
Most intrusions do not pick the lock. They sign in.
Almost every serious intrusion now begins the same quiet way: with a valid username and password, used by the wrong person. The defenses that matter here are not exotic. They decide how an identity is proven, how administrative privilege is separated, and whether anyone can later see what happened. When those choices are weak, an attacker does not break in. They log in, and often no one notices until long after.
When it goes wrong
No set of controls makes an account impossible to misuse, and chasing that is not the goal. The aim is narrower: to make a stolen credential hard to use, and its misuse easy to see. The situations below are ordinary, not exotic, and they occur more often than most organizations realize.
The reused password
An employee uses one familiar password for a personal shopping account and, for convenience, the company VPN. Months later the shopping site is breached and that password appears on a list traded among attackers. Automated tools then replay it against business logins.
One consumer breach becomes a working key to your network.The second factor that was not there
Multifactor authentication is switched on for the main sign-in page. An older mail protocol, left enabled for a legacy application, still accepts a password on its own. An attacker holding a phished password connects through that path and is never prompted for a code.
The second factor is in place, and quietly bypassed.The breach no one could see
An intruder is present for months. The affected device keeps only sparse, volatile logs that clear on reboot, and nothing gathers them in one place. When the activity is finally noticed, there is no record of what was reached or taken.
With no trail, the company must assume the worst and notify everyone.The shared administrator login
Several IT staff, and one departed contractor, all knew the same administrator password. A change is made that should not have been, or data leaves. The audit log shows only “administrator.”
The record cannot say who, and the credential walked out the door.The common thread
None of these turns on a firewall or an exotic exploit. Each was decided in advance by three things: how an identity was proven, whether one credential could be traced to one person, and whether the activity left a trail anyone could read.
Every one of these outcomes could very likely have been avoided. Not with a larger security budget, but with a few deliberate choices about how people prove who they are, how administrative privilege is separated, and whether activity is recorded. That is the whole of this issue.
What to Ask Your IT Manager or Provider
and how to know if the answers are good ones
You do not need to administer an identity platform or a monitoring system to oversee one. You need to know what to ask. A short primer on the finer points is provided at the end, for those who want the detail. First, the part that matters: run the checklist below with your IT professional. If any item is not marked green, it may warrant a closer look.
Passwords Built for Strength, Not Forced Complexity
No Reused or Shared Passwords
Phishing-Resistant Multifactor Authentication
Separate, Unique Administrative Accounts
A Master List of Credentials Held by Management
A Proper Password Repository
Third-Party Vendor Access on Request
Automatic Screen Locking
Seeing What Happens on the Network
detection, logging, and the record you will wish you had
Strong authentication decides who gets in. This second group decides whether anyone can see what happens next, in time to act and, later, to prove. The same rule applies: if an item is not green, it is worth a closer look.
Traffic Analysis or a SIEM With Alerting
Centralized, Durable Audit Logging
Web Use Monitoring, Where Appropriate
Physical Access Control and Visitor Logging
Going deeper: the finer points worth pinning downTerminology and the newer thinking, for whoever wants the detail. Optional.+
Why the multifactor method matters
Not every second factor resists a determined attacker. Codes sent by text message can be exposed through SIM-swap fraud, and both texted and app-generated codes can be captured by a convincing fake login page that relays them in real time. Repeated approval prompts can also wear a user down until one is accepted by mistake. Methods built on the FIDO2 standard, including passkeys and hardware security keys, are bound to the genuine site and are designed to resist these attacks, which is why they are described as phishing-resistant. Where a platform does not yet support them, an authenticator app is generally preferable to text-message codes.
The shift in password thinking
Guidance has moved away from short, complex passwords changed on a fixed schedule. Current NIST guidance (SP 800-63B) emphasizes length over enforced complexity, screening new passwords against lists of known-breached ones, and allowing long passphrases. It also advises against forcing routine password changes on a calendar, since that practice tended to produce weaker, predictable variations; a change is warranted when there is evidence a password may have been compromised. Organizations subject to specific compliance requirements should still follow the rules that apply to them.
Web use monitoring: the legal considerations
Where web use monitoring is deployed, a few precautions reduce liability. Personnel whose roles require visiting destinations that might otherwise be considered inappropriate can be exempted from filtering, and it is often recommended that such work be done through virtual browsing sessions to reduce the risk of malware. For roles likely to surface objectionable material, organizations sometimes obtain written legal acknowledgements, adjust position descriptions, add privacy screens, or position displays to limit what others can see. In every case, the organization should seek the guidance of its legal counsel to determine what is suitable.
Contact Sandbox Technologies for a free, no-obligation consultation.
Headquartered in Los Angeles. Serving clients nationwide and internationally since 1997.
Nationwide (Toll-Free)
877-930-8324
Los Angeles
424-207-5130
Support Helpdesk
424-207-5140

