Survey

SANS 2022 SOC Survey

Written by Chris Crowley and Barbara Filkins
May 2022

©2022 SANS™ Institute

Executive Summary
The content of this year’s SANS SOC Survey explores the ongoing development and
progress of the security operations center (SOC). Herein we explore details of who
answered the survey, the key takeaways we observe in their responses, and the challenges
everyone seems to face. The survey explores what people consider SOC capabilities, as
well as the staff, technology, deployment strategies, and the funding it takes to secure and
operate this gamut.
What Makes a SOC?
So how are SOCs evolving? To date, our definition of a SOC
remains conceptual, built around the capabilities required by We consider that a SOC is defined by its capabilities and
how these capabilities are prioritized by the organization
business-specific goals of an organization. A SOC framework is
owning the SOC. Capabilities are process-based, the related
not necessarily aligned with a reference architecture but comes service driven by the business needs or mission statement
from the technologies in use and the individuals who make up of the organization.
the SOC team to accomplish capabilities. We consider a SOC architecture as how organizations decide
to arrange their staff and technology to gain visibility into
In planning this survey, we took a capabilities-based approach to
protected systems, perform the required work, and take into
determine the current SOC landscape, with the goal of surfacing account the complicated logistical and jurisdictional issues
results that can help you assess your performance compared to address when monitoring information systems.
with your peers. In that regard, here are the top five questions
you might want to consider and our insights from this year’s survey:

1. Are trends going in the right direction?

Results from 2021 to 2022 show a decrease in both incidents and breaches from
incidents. This is a positive trend, but the question is, can it continue?

2. Does staffing match growth?

Hiring, retention, and turnover are key challenges. Consider comparing how
your organization lines up against the survey results.

3. Do capabilities match business need?

Survey Globalization
The leading items for survey respondents are detection/
Something new this year, which the authors are extremely
monitoring, vulnerability assessments, incident response,
proud of, is that the survey questions and answers were
and alert triage and escalation with capabilities balanced translated into Spanish and Portuguese. The intention is
between internal staffing and outsourced resources. that respondents who speak English as a second language
How do the capabilities your organization defined rank or non-English speakers can provide responses and share
thoughts succinctly and naturally. The intention is to
against these results?
scale this translation into more languages going forward.
4. Is the technology working? Optimistically, we’ll cover Europe, the Middle East, and Asia
with regionally specific languages to develop the SOC Survey
What technologies received a grade of an A and why?
into a truly global perspective on cybersecurity operations.
Compare how your organization lines up against the
survey results
in the section “Staffing: Meeting the Key Challenges.”

5. Are metrics measuring investment and what’s effective?

6. Are the discrepancies noted by this report being taken into account
as we move forward?

SANS 2022 SOC Survey 2

We all have our shortcomings, and this survey is no different. We’d love
to hear from our readers on how to make it better if we didn’t answer
your question in the following content. If we didn’t answer your questions
(or you are skeptical of our findings) and you want to perform your own
analysis of the data set in this survey, download it from
Within your organization, is use of the internal SOC viewed as
https://soc-survey.com/2022.
mandatory or is it acceptable for members of your organization
to acquire services from external parties/providers?

SOC Landscape Yes, use of the internal

SOC is mandatory. 243
No, we may acquire services
from an external provider. 170
When most of us think of a SOC, we probably envision No, we have no internal SOC. 88
some type of command center, housing InfoSec
Unknown 18
professionals who will detect, protect, and defend an 0 50 100 150 200 250
enterprise from cyberattacks. Indeed, most respondents,
Figure 1. SOC Services Optional or Obligatory (Q3.2: n=519)
at 47% (n=519) indicated that their SOC services are
obligatory for their organization (see Figure 1), with
Select the option that best reflects the size and
the majority of 53% (n=240) citing their SOC size and structure of your SOC environment.
structure as single and centralized, followed distantly by
Single, central SOC 240
multiple, hierarchical SOCs at 19% (n=85). See Figure 2.
Multiple, hierarchical SOCs 86
But looking ahead 12 months, while survey results
Multiple, standalone/siloed SOCs 33
show the single, centralized SOC (n=138) as the leading
deployment model, the real growth is occurring in Multiple, unorganized SOCs 33
Multiple, parallel, redundant
cloud-based SOC services.1 This opens the door to what SOCs looking at same data 25

we envision as the true definition of SOC, one based Other 37

on capabilities rather than a formal structure. See 0 50 100 150 200 250

Figure 3, noting that the second line (“Informal SOC, no Figure 2. Structure of SOC (Q3.5: n=454)
defined architecture”) doesn’t seem to be
an aspirational future state, because it Infrastructure Arrangement
represents low maturity. Current Next 12 months

129
Centralized into a single SOC
138

Informal SOC, no defined 75

architecture 26

69
Cloud-based SOC services
107

Centralized and 66
distributed regionally 65

47
Full SOCs distributed regionally
40

23
Partial SOCs in regional locations
19

29
Other
24
0 20 40 60 80 100 120 140

Figure 3. Infrastructure Arrangement Sorted by Current Deployment (Q3.6: n=454)

1
If expressed as a growth percentage (future/current counts divided by current count), the growth in the cloud-based SOC services sector is 55%,
whereas the growth in the central SOC is 7%. All other arrangements indicate a decline in the next 12 months.

SANS 2022 SOC Survey 3

Meeting the Key Challenges
Given that SOC capabilities mirror key security functions, such as incident response, we
looked at the survey responses to see if the challenges to fully utilizing SOC capabilities
would follow those barriers faced by most security professionals.

Not surprisingly, high staffing requirements is the

What is the greatest challenge (barrier) with regard to full utilization of
biggest stumbling block (46 of 235 respondents), your SOC capabilities by the entire organization? Select the best option.
followed by a lack of skilled staff (34). A lack
High staffing requirements 46
of automation and orchestration (32/235)
followed. The fourth largest challenge—a lack Lack of skilled staff 34

of management support (23)—brings the top Lack of automation and orchestration 32

four barriers to account for 57% (135/235) of all Lack of management support 23

responses. See Figure 4. Lack of enterprisewide visibility 20

We wanted to hear the respondents’ challenges Silo mentality between security, IR, and operations 19

in their own words, so we gave a free format Lack of processes or playbooks 13

opportunity to write it down. The word cloud in Too many alerts that we can’t look into
13
(lack of correlation between alerts)
Figure 5 captures a frequency-weighted depiction Too many tools that are not integrated 12
of the words used in the responses. Human
Lack of context related to what we are seeing 11
resources issues, money, and management
Regulatory or legal requirements 5
support are repeated almost universally.
Other 7
0 10 20 30 40 50

Figure 4. Challenges to Using SOC Capabilities (Q3.61: n=235)

Figure 5. Barriers to SOC Use (Q3.59: n=130)

SANS 2022 SOC Survey 4

Demographics
Survey respondents provided insight from their personal experiences with results heavily
influenced by individuals whom SANS asked to participate in the survey through outreach and
marketing efforts. To conduct a survey at this level of detail, we depend on respondents to
volunteer 30 to 60 minutes of their time to answer thought-provoking and detailed questions.
It is the opinion of the authors that we still don’t have a globally representative sample of
cybersecurity operations centers, but we’re striving to get there. Figure 6 provides a snapshot
of the demographics for the respondents to the 2022 survey.

Top 4 Industries Represented Organizational Size

Small
Technology (Up to 1,000)

Banking and Small/Medium

finance (1,001–5,000)

Medium
Cybersecurity (5,001–15,000)

Medium/Large
Government (15,001–50,000)

Each gear represents 10 respondents. Large

(More than 50,000)

Operations and Headquarters Each building represents 25 respondents.

Ops: 74
Top 4 Roles Represented
HQ: 21 Ops: 67
HQ: 12 Security administrator/
Ops: 150 Security analyst
HQ: 92

SOC analyst

Security manager or
Ops: 246 Ops: 42
HQ: 3
director
HQ: 207

Ops: 42 IT manager or
Ops: 235
Ops: 32 HQ: 6 director
HQ: 10
HQ: 168
Each person represents 10 respondents.

Figure 6. SOC Survey

Demographic Highlights Respondent Demographics
• The majority of respondents are from smaller organizations. Of the 519 participants who
answered, half (50%) are from organizations with fewer than 1,000 people. This tendency
toward smaller organizations may be due to fact that the high-tech and cybersecurity
companies function in a technology support role and, traditionally, have smaller staff.

• Respondents are primarily in high tech, financial, cybersecurity, or government. If you’re

in one of these sectors and don’t have a SOC, your peers likely do.

SANS 2022 SOC Survey 5

 • Of the 519 respondents, 270 (52%) have analyst, administrator, or architect roles, while
214 (41%) serve as managers, directors, or c-level executives. The remaining 35 (7%)
listed their roles as other. The other roles included specific forms of analyst, engineer,
specialist, and manager not specifically cited in the list of choices but still cyber-
focused. There were several educator, consultant, repair, sales, and other write-in titles
which had no corresponding choice in the list.

• The respondents’ companies are primarily based in North America, Latin America, and
Europe. Previous years’ SOC surveys speculated the prevalence of North American
and European organizations as a feature of who participated, rather than a feature
of the actual prevalence of SOCs globally. This year’s report and survey targeted Latin
America, and this region took the second among headquarters for the first time. Our
mission for this survey going forward is to continue similar outreach globally.

Factors for SOC Success

Survey readers tell us that they frequently seek budget, technology, Has your organization suffered an incident
and staff based on the results and insights described in our or intrusion in the past 12 months?
SOC surveys. The authors take this seriously. We’re providing the
guidance in this section around some key questions you should ask 47
yourself and your organization to help you make difficult choices.
71 151 Yes
And we have included a few key findings from this survey to help
No
you guide your decisions and requests.
Unsure/Unknown

Success Factor One: 250

Prefer not to answer

Are Trends Going in the Right Direction?

This year’s survey, more respondents as a percentage (48%)
answered “No,” indicating their organization had not suffered
Figure 7. Intrusions/Incidents in
an intrusion in the past 12 months. This statistic is better than last year’s SOC survey, the Past 12 Months (Q3.3: n=519)
where 39%2 answered “No.” In the way the question is phrased, more people responding
“No” indicates fewer incidents occurring. Of course, there were numerous responses that
expressed ignorance (71, 14%) or unwillingness (47, 9%)
to acknowledge if a breach happened. See Figure 7. Did this incident or intrusion result in a breach, implying the
generally accepted definition that your organization had to publicly
Of the respondents who had an intrusion, we followed notify specific parties potentially implicated by the breach?
up with the next logical question: Did this result in a
Yes 48
breach? Of the 128 people who answered Question 3.4,
No 68
68 responses (53%) indicated that the intrusion didn’t
result in a breach (see Figure 8). In comparison to the Unsure/Unknown 16

2021 SOC Survey, 95 responses were collected to the Prefer not to answer 4

“result in a breach” question, and 47% of those said 0 10 20 30 40 50 60 70 80

Figure 8. Breaches in the Past 12 Months (Q3.4: n=136)

2
“A SANS 2021 Survey: Security Operations, Center,” October 2021,
www.sans.org/white-papers/sans-2021-survey-security-operations-center-soc/?socsurvey=1, p. 2. [Registration required.]

SANS 2022 SOC Survey 6

“No.” In other words, in 2022 more people answered that the intrusion did not result in
a breach, which is an improvement. Bravo for finding and stopping it! We started asking
this question in 2021, and so far, the trend seems to be going in the right direction.

We didn’t have a follow-on question related to what enabled the detection and
removal of the intrusion prior to the breach. There are a multitude of other reports and
surveys documenting the details of breach and loss from the cybersecurity community.
But we will continue to assess if this trend of success we are seeing in reduced
incidents and subsequent breaches can be attributed to the presence of a SOC.

Track metrics to ensure your trends are moving in the right direction.

Success Factor Two: Does Staffing Match Growth?

This year staffing levels remained fairly consistent with the past. The most popular SOC
size is 2–10 people, with 66 of 234 respondents selecting that answer.3 See Figure 9.

Despite a constant team size

What is the total internal staffing level (i.e., all related positions) for your SOC,
year after year in the survey,
expressed in terms of full-time equivalents (FTEs)?
staff turnover and retention are
70 66
leading concerns. Staff turnover
60
remains high: 70% for individuals
with five or fewer years of 50

experience, with the majority 40 38

remaining in their current 30 26 27

position for fewer than three. 20 17 19

Retaining talent is also critical, 10

5
3
requiring a critical and continual 0
<1 (part-time) 1 2–10 11–25 26–100 101–1000 >1000 Unknown
pitch by the SOC to organizational
Figure 9. SOC Size by Number of FTEs
management. This pitch needs
(Q3.46: n=234)
to be supported by metrics that demonstrate value and clear planning to articulate
expected improvements. Organization management considers the business
operational environment and weighs the cyber threats against the business impacts.

Compare how your organization lines up against the survey results in the section
“Staffing: Meeting the Key Challenges.”

Success Factor Three: Do the Capabilities Fit the Business Need?

We will go into detail on capabilities in a later section, but if your team isn’t performing
against one or more of these capabilities listed in Figure 16 (see page 11), your team
probably won’t be considered a SOC.

The leading items for survey respondents are detection/monitoring, vulnerability

assessments, incident response, and alert triage and escalation, with capabilities balanced
between internal staffing and outsourced resources.

Rank the capabilities your organization defined against these results.

3
Please note that this is not organizational size or sector adjusted.

SANS 2022 SOC Survey 7

Success Factor Four: Is Technology Working?
Technology elicits strong opinions from most cybersecurity professionals. This year, we
assigned our technologies a GPA based on the grades individual respondents assigned to
each technology. The more successful technologies are focused on the mature and stable as
opposed to the new and exciting, but even the leading technologies only got a B– overall.

Explore the section “Technology: What Is Getting a Passing Grade?” to see how to
assess your technologies as well as the part on tying the capabilities together.

Success Factor Five: Are Your Metrics Really Measuring the

Effectiveness of Your Investment?
With an increased emphasis on staffing as a leading challenge/barrier to SOC effectiveness,
organizations must consider how executive management not only listens but also acts on what
their SOC leadership is telling them. Although 39% reported that executive management and
SOC leads work closely together in allocating funds for cybersecurity, 55% believe the decision
is wholly that of executive management, despite any recommendations from the SOC team.
And although 41% report that management pays close attention to the recommendations and
needs of SOC leads with regard to hiring and retaining skilled, experienced staff for defending
the enterprise, 55% again think that, while executive management may again listen, they do
not act on the urgency to retain, not just hire, skilled staff.

Calculate a metric that measures the effectiveness of the SOC.

Staffing: Meeting the Key Challenges

Let’s first talk about people. There is no change in overall SOC team size from previous years.
Again, 2–10 staff members is the most commonly cited team size, regardless of the size of the
organization. See Figure 10.

Size of Organization Versus SOC Team Size

<1 (part-time) 1 2–10 11–25 26–100 101–1,000 >1,000 Unknown

14 14
14

12

10
10

8
7
66 6 6 6
6
5 5 5 5 5 5 5 5
4 4 4 4 4 4
4
3 3 3 3 3 3 3 3 33
2 2 2 2 2 2 2 22 2
2
1 1 1 1 1 1 1 1 1 1 1 1 1
0 0 0 0 0 0 0 0 00 0 0 0 00 0 00 00 0 00 0
0
Fewer than 100 101–500 501–1,000 1,001–2,000 2,001–5,000 5,001–10,000 10,001–15,000 15,001–50,000 50,001–100,000 More than 100,000

Figure 10. Size of Organization Versus SOC Team Size

SANS 2022 SOC Survey 8

The top two barriers to full SOC utilization
What is the average employment duration for an employee in
are staffing-related (see Figure 5). What are your SOC environment (how quickly does staff turn over)?
respondents saying about the challenges they
1 year or less 13
face? Frequent turnover appears to be the first
1–3 years 84
obstacle. In Figure 11, we see that the average
duration of employment is predominantly 3–5 years 68

less than five years, with the most cited (n=84) 5–10 years 29

remaining one to three years before leaving. 10+ years 15

So, what are most effective methods of Unknown 27

retaining staff? The leading approach is to 0 10 20 30 40 50 60 70 80

provide staff clear career progression (n=76), as Figure 11. Average Employment Duration (Q3.48: n=236)

shown in Figure 12. The authors see this as providing two benefits:

• Employees stay with the organization

What is the most effective method you have found to retain employees?
longer, reducing the cost of hiring and
training new staff. Career progression 76

• People who want to grow and develop Money 54

within an organization tend to be more Meaningful work 37

productive and effective employees. Training 31
Shifting roles and
Employees who have a mind to grow and responsibilities regularly 26

develop are likely caring workers. If you Other 15

don’t facilitate that growth, they’ll find it in a 0 10 20 30 40 50 60 70 80

different company that has a plan in place for Figure 12 Retention Methods (Q3.49: n=239)
career growth.

Retention seems to be an important part of managing the SOC The short story of our guidance here is calculate the
team, implying likely negotiation between SOC and organizational costs involved in hiring new staff. Show the value
management to address key factors to keep staff: career plan proposition by comparing the hiring cost to the cost
assurance with additional training, monetary incentives, work-life of training and developing existing staff.

balance, and relaying the value of the work performed.

So, what did the respondents say about

How is human capital addressed in your environment? Select the best option.
their management’s investment in staff?
Management pays close attention to the
Many respondents (n=96) indicated that needs of SOC leads/managers with regard
to hiring and retaining skilled, experienced 96
organizational management coordinates with staff to defend their environment.
SOC management and team leads to hire Management listens to the requests of SOC
leads/managers regarding hiring skilled,
69
and retain the right people to defend the experienced staff but does not understand
the urgency to retain these skilled people.
environment. But, while this indicates that Management does not pay any mind to
the unique staffing needs of a SOC and
the situation many be improving from past does little to encourage hiring skilled, 46
experienced staff or retain them.
surveys, the balance (n=129) still feels that Management thinks hiring multiple,
less-skilled employees to stare at alerts
executive management may listen, but they do is an acceptable strategy for mitigating 14
cybersecurity threats in their environment.
not act on the urgency to retain, not just hire,
Other 9
skilled staff. See Figure 13.
0 20 40 60 80 100

Figure 13. Human Capital Management Approach (Q3.53)

SANS 2022 SOC Survey 9

Working Remotely Do you allow SOC staff analysts to work remotely?
The pandemic forced a new reality on the workplace. Working remotely
is now the norm. And that includes the SOC team. You probably 22
already allow your SOC staff to work remotely. If you don’t, you’re
53
competing for that staff with companies that do. See Figure 14. Yes

Organizations now need to balance the ease of hiring and retaining No

staff who expect to work remotely versus the increased difficulty in Unknown
296
training, developing, and onboarding staff working from home.

So, what factors do organizations take into consideration for a SOC

staff analyst to work remotely from home? The leading consideration
(n=165) is if the platform supports it. (See Figure 15.) We take this to Figure 14. Remote Work (Q3.13: n=371)
mean that the data sensitivity is weighed against the risk of accessing
the data from off-premises, and a risk-weighted decision is made
about the work from home. This seems a reasonable consideration The authors concur that few organizations can justify
that can be made in coordination with risk management around the an on-premises-only SOC based on data sensitivity.
sensitivity of the systems protected. While some organizations must maintain this
security posture for specific reasons, in general this
The next few considerations, however, are more of a SOC management position is becoming more difficult to justify.
judgment call based on the individual’s performance and perceived
skill set (n=139). But the growth in remote work/work
from home due to the pandemic appears to have What factors are considered in determining whether a SOC
changed how organizations determine whether an staff analyst can work remotely? Check all that apply.
individual can work remotely. In 2022, work ethics Platforms securely support
remote workforce 165
rose to the third most important evaluation factor
Skill set 139
(n=115): 43% in 2022 versus 33% in 2021. (Note: While
Work ethics 115
an individual’s work ethic seems a reasonable basis
for consideration for remote work, it might be hard to Role 96

quantify fairly. Ostensibly, the rationale here is that Seniority 78

those with a lesser work ethic are motivated to be more Individually negotiated 68
effective by the oversight provided on-premises.) Other 32
0 40 80 120 160

Figure 15. Factors in Work From Home (Q3.14: n=266)

SANS 2022 SOC Survey 10

Capabilities: Does Your Team Count as a SOC?
Do you ever wonder if your team counts as a SOC? You’re not alone. We asked
respondents to identify the capabilities they have within their SOC. Figure 16 shows a
count-based, ranked list of what the most
Capability – Sorted by Total
respondents said they do. The leading
In-house Outsourced Both
items are detection/monitoring (376),
vulnerability assessments (373), incident Security monitoring and detection
209 58 109
response (373), and alert triage and Alerting (triage and escalation)
219 68 86
escalation (373). Incident response
227 47 99
We define a SOC through capabilities and Vulnerability assessments
178 75 120
architecture. So, if you’re not performing Compliance support
243 57 71
the capabilities listed in Figure 16 either Data protection
264 39 68
internally and/or by outsourcing, that Security tool configuration,
integration, and deployment
particular group wouldn’t be considered 217 55 98
Security administration
a SOC. 267 66 36
Security architecture and engineering
(of systems in your environment) 253 43 73
Much of what SOCs do can be outsourced,
Digital forensics
so we wanted to understand what SOCs 175 102 91
Threat research
choose to outsource. There is basically no 163 88 116
Remediation
change in the outsource portfolio from 223 53 90
Security road map and planning
previous years’ SOC surveys. See Figure 17, 268 39 59
SOC architecture and engineering
where the data from Figure 16 is re-sorted (specific to the systems running your SOC) 218 69 79
by how many respondents outsource that Pen-testing
133 149 83
capability. Threat hunting
189 76 99
Threat Intelligence (production)
155 108 101
What’s interesting to the author (Crowley) SOC maturity self-assessment
195 76 90
is how much consensus exists in this
Threat intelligence (attribution)
response set. If “Other” is excluded 149 103 108

(112), then the range is 348–376 from 383 Threat Intelligence (feed consumption)
149 89 121
respondents. There were only 20 other text Red-teaming
142 138 76
responses, and 14 of these were none or
Purple-teaming
N/A. A couple were comments about the 162 116 70
question. There were three relevant other Other
56 22 48
responses: business review, research, and 0 100 200 300 400
NIST framework management. The research
we’d put with “threat research” and the Figure 16. Capabilities Ranked on Total Reporting (Q3.10: n=383)
NIST framework we would expect to group
with “compliance support.”

SANS 2022 SOC Survey 11

While you might have good reasons to
Capability – Sorted by Percentage Outsourced
retain pen testing and its variants, threat
In-house Outsourced Both
intelligence, and forensics internally,
Pen-testing
many of your peers continue to outsource 133 149 83
Red-teaming
these capabilities, probably for two basic 142 138 76

reasons. Retaining this expertise on staff Threat intelligence (attribution)

149 103 108
can be cost-prohibitive, because talented Threat intelligence (feed consumption)
149 89 121
specialized staff are rare. An added Threat intelligence (production)
155 108 101
complexity is the requirement for training Purple-teaming
162 116 70
specialized staff, because keeping up to Threat research
163 88 116
date with constantly changing technology
Digital forensics
and techniques, such as in pen testing 175 102 91
Vulnerability assessments
(and related) and threat intelligence, 178 75 120
Threat hunting
can be difficult. Finally, budgeting for a 189 76 99
SOC maturity self-assessment
full-time individual specialist in a small 195 76 90

team usually doesn’t make financial Security monitoring and detection

209 58 109
Security tool configuration,
sense, as general-purpose analysts are integration, and deployment 217 55 98
usually internal staff. So, focused expertise SOC architecture and engineering
(specific to the systems running your SOC) 218 69 79
is procured in an outsourced fashion. Alerting (triage and escalation)
219 68 86
In larger teams, calculating the value Remediation
223 53 90
proposition of outsourcing usually directs
Incident response
the outsource action. 227 47 99
Compliance support
243 57 71
So, what compels retention of staff Security architecture and engineering
(of systems in your environment) 253 43 73
internally? The rationale for capabilities
Data protection
which tend to be performed internally 264 39 68
Security administration
are those which require tailoring to the 267 66 36
Security road map and planning
organization. The institutional knowledge 268 39 59
Other
to perform the tailoring is necessary for 56 22 48

this work as the work tends to stay internal 0 100 200 300 400

(also, the perceived need of of data control Figure 17. Capability Ranked
Greatest to Lowest for
and privacy for these capabilities and the data handled therein). Defensible or
Outsourcing (Q3.10: n=383)
not, this is the stance that many organizations use to keep the activity internal
rather than outsourced.

Technology: What Is Getting a Passing Grade?

From a technology perspective, we discussed technology deployed and how far along it is
in the deployment. Of course, we also asked what technologies people like and don’t like.
The two questions we asked to discover this information included long lists of choices
and took at least 10 minutes to answer, so we allowed people to skip them, but 53% (Q3.26,
n=308) agreed to answer them.

SANS 2022 SOC Survey 12

A perennial problem for IT environments is the “partial deployment” of systems. In the
authors’ experience, many SOCs’ maturity, efficiency, and improvement efforts are pending
completion of some technology deployment. Figure 18 indicates where the respondents’
organizations are in their deployment efforts.

Production Production Purchased Not

Category: Question (All Systems) (Partial Systems) Implementing Implemented Planned
Host: Vulnerability remediation 58 42 24 5 14
Host: Malware protection system (MPS) 60 43 22 6 12
Host: Behavioral analysis and detection 51 29 27 9 24
Host: Data loss prevention 42 35 27 13 26
Host: Ransomware prevention 56 27 26 11 19
Host: User behavior and entity monitoring 39 34 23 10 30
Host: Endpoint or extended detection and 59 28 26 5 19
response (EDR/XDR)
Host: Application whitelisting 46 34 23 11 24
Host: Continuous monitoring and assessment 61 37 18 11 16
Log: Endpoint OS monitoring and logging 57 38 18 12 17
Log: Endpoint application log monitoring 54 33 27 10 15
Log: Log management 56 40 23 10 13
Log: DNS log monitoring 47 43 25 6 20
Net :Network segmentation 64 41 15 10 9
Net: Email security (SWG and SEG) 72 33 15 8 12
Net: DNS security/DNS firewall 70 33 21 4 13
Net: Asset discovery and inventory 47 39 26 12 17
Net: VPN (access protection and control) 79 30 17 9 7
Net: Full packet capture 32 35 26 9 32
Net: Packet analysis (other than full PCAP) 41 33 23 10 26
Net: DoS and DDoS protection 51 48 15 13 15
Net: Network traffic monitoring 52 42 21 11 15
Net: Web application firewall (WAF) 53 52 16 11 10
Net: Next-generation firewall (NGF) 73 30 15 11 13
Net: Egress filtering 53 42 22 5 18
Net: Deception technologies such as honey potting 33 29 22 11 39
Net: Web proxy 58 36 19 6 16
Net: Network access control (NAC) 46 31 23 14 25
Net: NetFlow analysis 32 33 23 16 30
Net: Malware detonation device 33 31 23 11 36
(inline malware destruction)
Net: Network intrusion detection system 67 41 19 7 7
(IDS)/intrusion prevention system (IPS)
Net: SSL/TLS traffic inspection 40 39 23 6 24
Net: Ingress filtering 56 39 18 10 15
Analysis: Risk analysis and assessment 53 38 27 9 11
Analysis: S  IEM (security information and 63 34 25 6 12
event manager)
Analysis: C  ustomized or tailored SIEM 53 32 22 13 19
use-case monitoring
Analysis: AI or machine learning 36 26 20 6 42
Analysis: Frequency analysis for network connections 38 33 25 7 27
Analysis: E  xternal threat intelligence 44 32 23 12 22
(for online precursors)
Analysis: Threat hunting 38 36 28 9 23
Analysis: Threat intelligence platform (TIP) 40 28 27 8 33
Analysis: T  hreat intelligence 42 29 29 14 22
(open source, vendor provided)
Analysis: E  -discovery (support legal requests for 36 29 22 5 40
specific information collection)
Analysis: S  OAR (security orchestration, 39 30 19 9 37
automation, and response)
Other 13 10 11 5 12

Figure 18. Technology in Use and Deployment Status (Q3.26: n=150)

SANS 2022 SOC Survey 13

This view is useful, but it might also be beneficial to consider this in terms of
accomplishing “all systems” deployment. To this end, Figure 19 shows a sorted view of the
data shown in Figure 18. It loses the categorical grouping as a result. But this figure might
be used as a guide on the likelihood of getting your intended technology fully deployed.

Percentage Production Production Purchased Not

Category: Question (All Systems) (All Systems) (Partial Systems) Implementing Implemented Planned
Net: VPN (access protection and control) 55.6% 79 30 17 9 7
Net: Next-generation firewall (NGF) 51.4% 73 30 15 11 13
Net: Email security (SWG and SEG) 51.4% 72 33 15 8 12
Net: DNS security/DNS firewall 49.6% 70 33 21 4 13
Net: N
 etwork intrusion detection system 47.5% 67 41 19 7 7
(IDS)/intrusion prevention system (IPS)
Net :Network segmentation 46.0% 64 41 15 10 9
Analysis: SIEM (security information & event manager) 45.0% 63 34 25 6 12
Host: Continuous monitoring and assessment 42.7% 61 37 18 11 16
Host: Malware protection system (MPS) 42.0% 60 43 22 6 12
Host: E ndpoint or extended detection and response 43.1% 59 28 26 5 19
(EDR/XDR)
Host: Vulnerability remediation 40.6% 58 42 24 5 14
Net: Web proxy 43.0% 58 36 19 6 16
Log: Endpoint OS monitoring and logging 40.1% 57 38 18 12 17
Log: Log management 39.4% 56 40 23 10 13
Host: Ransomware prevention 40.3% 56 27 26 11 19
Net: Ingress filtering 40.6% 56 39 18 10 15
Log: Endpoint application log monitoring 38.8% 54 33 27 10 15
Net: Web application firewall (WAF) 37.3% 53 52 16 11 10
Net: Egress filtering 37.9% 53 42 22 5 18
Analysis: Customized or tailored SIEM 38.1% 53 32 22 13 19
use-case monitoring
Analysis: Risk analysis and assessment 38.4% 53 38 27 9 11
Net: Network traffic monitoring 36.9% 52 42 21 11 15
Net: DoS and DDoS protection 35.9% 51 48 15 13 15
Host: Behavioral analysis and detection 36.4% 51 29 27 9 24
Log: DNS log monitoring 33.3% 47 43 25 6 20
Net: Asset discovery and inventory 33.3% 47 39 26 12 17
Net: Network access control (NAC) 33.1% 46 31 23 14 25
Host: Application whitelisting 33.3% 46 34 23 11 24
Analysis: External threat intelligence 33.1% 44 32 23 12 22
(for online precursors)
Host: Data loss prevention 29.4% 42 35 27 13 26
Analysis: Threat intelligence 30.9% 42 29 29 14 22
(open source, vendor provided)
Net: Packet analysis (other than full PCAP) 30.8% 41 33 23 10 26
Analysis: Threat intelligence platform (TIP) 29.4% 40 28 27 8 33
Net: SSL/TLS traffic inspection 30.3% 40 39 23 6 24
Host: User behavior and entity monitoring 28.7% 39 34 23 10 30
Analysis: SOAR (security orchestration, 29.1% 39 30 19 9 37
automation, and response)
Analysis: Threat hunting 28.4% 38 36 28 9 23
Analysis: Frequency analysis for network connections 29.2% 38 33 25 7 27
Analysis: E-discovery (support legal requests for 27.3% 36 29 22 5 40
specific information collection)
Analysis: AI or machine learning 27.7% 36 26 20 6 42
Net: Deception technologies such as honey potting 24.6% 33 29 22 11 39
Net: M
 alware detonation device 24.6% 33 31 23 11 36
(inline malware destruction)
Net: Full packet capture 23.9% 32 35 26 9 32
Net: NetFlow analysis 23.9% 32 33 23 16 30
Other 25.5% 13 10 11 5 12

Figure 19. Technology in Use and Deployment Status, Sorted by Percentage “All Systems” (Q3.26: n=150)

SANS 2022 SOC Survey 14

While each organization has distinct challenges, there is some benefit to projecting based
on this chart. The author’s (Crowley) speculation on why the projects at the top of the list
of all systems have gotten there is that network security controls have been in place for
many years. A network-implemented solution represents the older paradigm of perimeter
protection. These solutions can be deployed with the IT teams as networks are updated.
While users are affected by VPN solutions in a way that requires changes in behavior, the
rest of the technologies topping the list have no requirement of user behavior change.
Related, a VPN forces users through a new channel to access data, but once configured, it
is minimally impactful on the user.

Technology, regardless of deployment status, is a source of strong opinions from most

cybersecurity professionals. To assess the reasons why some technology scored well,
such as VPN (Grade=+79), which is the highest-ranked technology in Figure 20, will
require speculation.

The highest-ranking tools score at the top because they do something well and the tools
are likely in the enlightenment or productivity phase of the hype cycle.4 In this continued
speculation, the lowest scorers are relatively new or fail to do the task for which they were
purchased. That may be no fault of the technology (full-PCAP not allowed to capture due
to legal restrictions) or failure of the technology to adapt (full-PCAP being outmaneuvered
by advancing encryption protocols).

Finally, the middle of the pack looks like the technology that works well only when
you’ve applied the appropriate customization and tailoring for your environment. This
takes time, dedicated staff, and cooperation with the protected systems’ owners and IT
administrators. Cooperation and time are often an unnecessarily scarce commodity in the
cybersecurity space.

Figure 20 (shown on the next page) represents the GPA for each technology based on the
grades respondents assigned to that technology. The GPA is calculated on a 4-point scale,
where A is 4 and F is 0, divided by the number of responses per technology.

To be fair, the GPA is based on the respondent’s opinion, and we do not have a fully
developed assessment rubric for this. Take this respondent’s opinion-based GPA scoring
as anecdotal opinions on a product category, not specific products.

4
“Gartner Hype Cycle,” www.gartner.com/en/research/methodologies/gartner-hype-cycle

SANS 2022 SOC Survey 15

Category: Question A B C D F GPA
Net: VPN (access protection and control) 53 52 16 11 2 3.11
Analysis: SIEM (security information and event manager) 49 44 25 7 5 3.08
Log: Endpoint OS monitoring and logging 47 42 29 7 9 3.03
Net: Next-generation firewall (NGF) 45 44 30 8 6 2.99
Host: Endpoint or extended detection and response (EDR/XDR) 45 37 27 10 7 2.98
Net: Web application firewall (WAF) 43 47 25 11 6 2.97
Host: Ransomware prevention 46 36 25 13 8 2.96
Log: Endpoint application log monitoring 38 41 37 3 9 2.96
Host: Vulnerability remediation 44 42 38 6 3 2.95
Net :Network segmentation 44 36 28 11 7 2.95
Analysis: Customized or tailored SIEM use-case monitoring 39 45 28 10 7 2.93
Host: Malware protection system (MPS) 40 50 30 11 3 2.91
Net: Ingress filtering 40 41 34 9 3 2.90
Net: DNS security/DNS firewall 38 47 29 11 8 2.90
Net: Egress filtering 39 34 39 7 8 2.88
Net: DoS and DDoS protection 45 33 38 11 7 2.88
Net: Email security (SWG and SEG) 44 35 29 15 9 2.88
Log: DNS log monitoring 38 45 28 13 8 2.87
Net: Web proxy 38 33 35 9 14 2.87
Host: Continuous monitoring and assessment 39 38 37 10 6 2.85
Analysis: Risk analysis and assessment 31 45 34 7 8 2.85
Net: N
 etwork intrusion detection system (IDS)/intrusion 41 38 32 14 5 2.85
prevention system (IPS)
Other 10 7 11 2 7 2.83
Log: Log management 36 49 28 15 5 2.83
Net: Network access control (NAC) 36 36 30 13 13 2.83
Host: Behavioral analysis and detection 39 39 36 13 5 2.82
Net: Network traffic monitoring 33 40 41 9 6 2.79
Analysis: Threat intelligence platform (TIP) 32 36 30 13 13 2.78
Host: Application whitelisting 30 41 30 13 13 2.77
Analysis: Frequency analysis for network connections 31 28 34 11 21 2.76
Analysis: External threat intelligence (for online precursors) 32 36 29 16 10 2.74
Net: Deception technologies such as honey potting 31 26 33 13 22 2.73
Analysis: Threat intelligence (open source, vendor provided) 28 37 43 10 9 2.70
Analysis: E-discovery (support legal requests for specific 24 39 30 14 14 2.68
information collection)
Net: SSL/TLS traffic inspection 29 32 37 14 14 2.68
Analysis: Threat hunting 32 30 36 17 10 2.67
Net: Packet analysis (other than full PCAP) 24 40 30 16 15 2.65
Analysis: AI or machine learning 27 26 37 13 20 2.65
Host: User behavior and entity monitoring 27 41 34 18 7 2.64
Net: Malware detonation device (inline malware destruction) 25 27 37 13 23 2.63
Host: Data loss prevention 30 32 35 20 15 2.62
Net: Full packet capture 27 30 36 17 17 2.61
Net: Asset discovery and inventory 31 27 46 16 10 2.61
Net: NetFlow analysis 27 24 43 13 12 2.61
Analysis: SOAR (security orchestration, automation, 23 26 38 12 17 2.61
and response)

Figure 20. GPA Rating (Q3.27: n=132)

SANS 2022 SOC Survey 16

If you want to see these two SUM Prod
Category: Question (Full + Partial) GPA Total
charts combined (like we did), Net: VPN (access protection and control) 109 3.11 132
see Figure 21. Net: N
 etwork intrusion detection system (IDS)/intrusion 108 2.85 125
prevention system (IPS)
Net: Web application firewall (WAF) 105 2.97 126
Net :Network segmentation 105 2.95 119
Net: Email security (SWG and SEG) 105 2.88 123
Net: Next-generation firewall (NGF) 103 2.99 127
Host: Malware protection system (MPS) 103 2.91 131
Net: DNS security/DNS firewall 103 2.90 125
Host: Vulnerability remediation 100 2.95 130
Net: DoS and DDoS protection 99 2.88 127
Host: Continuous monitoring and assessment 98 2.85 124
Analysis: SIEM (security information and event manager) 97 3.08 125
Log: Log management 96 2.83 128
Log: Endpoint OS monitoring and logging 95 3.03 125
Net: Ingress filtering 95 2.90 124
Net: Egress filtering 95 2.88 119
Net: Web proxy 94 2.87 115
Net: Network traffic monitoring 94 2.79 123
Analysis: Risk analysis and assessment 91 2.85 117
Log: DNS log monitoring 90 2.87 124
Host: Endpoint or extended detection and response (EDR/XDR) 87 2.98 119
Log: Endpoint application log monitoring 87 2.96 119
Net: Asset discovery and inventory 86 2.61 120
Analysis: Customized or tailored SIEM use-case monitoring 85 2.93 122
Host: Ransomware prevention 83 2.96 120
Host: Behavioral analysis and detection 80 2.82 127
Host: Application whitelisting 80 2.77 114
Net: SSL/TLS traffic inspection 79 2.68 112
Net: Network Access Control (NAC) 77 2.83 115
Host: Data loss prevention 77 2.62 117
Analysis: External threat intelligence (for online precursors) 76 2.74 113
Analysis: Threat hunting 74 2.67 115
Net: Packet analysis (other than full PCAP) 74 2.65 110
Host: User behavior and entity monitoring 73 2.64 120
Analysis: Frequency analysis for network connections 71 2.76 104
Analysis: Threat intelligence (open source, vendor provided) 71 2.70 118
Analysis: SOAR (Security Orchestration, Automation, Response) 69 2.61 99
Analysis: Threat intelligence platform (TIP) 68 2.78 111
Net: Full packet capture 67 2.61 110
Analysis: E-discovery (support legal requests for specific 65 2.68 107
information collection)
Net: NetFlow analysis 65 2.61 107
Net: Malware detonation device (inline malware destruction) 64 2.63 102
Net: Deception technologies such as honey potting 62 2.73 103
Analysis: AI or machine learning 62 2.65 103
Other 23 2.83 30

Figure 21. Deployment State and GPA Rating (Q3.26 and Q3.27: n=142,132)

SANS 2022 SOC Survey 17

Tying Pieces Together
The equation for successful SOC operation is additive:
people (staff) plus capabilities (process) plus
technology. Our survey addressed several instances
that tie these elements together.
What is included in your security monitoring activities? Select all that apply.
Monitoring
Detection of threats 315
Drilling down into the monitoring
Access and usage monitoring 305
capability, we asked exactly what this
Protection of data 258
entailed. (This question is new this year.)
Determine threat landscape/
Detection of threats is at the top. But in Identify emerging threats 222

many cases, it seems that the monitoring Support for IR 194

team doesn’t support incident handling, Other 8

at 47% (174/368). This is an area of 0 50 100 150 200 250 300 350

improvement the authors think warrants Figure 22. Security Monitoring Details (Q3.11: n=368)
focus: better integration between monitoring and
incident handling. See Figure 22. Does your SOC operate 24/7?

Because most computer networks never shut down,

12
the SOC should probably monitor 24 hours a day, every
day. We asked if this is the case, and it is in most cases Yes, in-house only
62
(details are in Figure 23). Only 62 respondents (17%) 144 Yes, outsourced only

indicated that the SOC doesn’t operate 24 hours per  es, mixed internal/
Y
outsourced
day. The “Yes” contingent answered that a purely in- 89 No
house (144, 38%) 24-hour operation is the most popular
approach to this, trailed substantially by mixed (89, 66 Unknown

24%). See Figure 23.

While running 24/7 operations, the primary task is

Figure 23. 24/7 Operations or Not? (Q3.12: n=373)
identification of issues (see Figure 24).
We have a section later about technology,
Primary Technology for Event Data Correlation
but capability is integrally connected to
the technology used to perform it. So, Through our SIEM 151
we asked how the data correlation of Through our security orchestration,
36
automation, and response (SOAR) platform
event data for identification of issues Through our EDR/XDR platform 35
is performed. To the authors, the top Through services provided
by our MDR vendor 24
answer “through our SIEM” (151, 47%) is
Through a workflow tool 21
no surprise. Our projection is that this
Through our aggregated log
15
obvious dominance changes soon, as management system
Through home-developed
15
more SOCs shift this effort into SOAR APIs and dashboards

(36, 11%), XDR (35, 11%), and MDR (24, 7%) Don’t know. It all happens in the cloud. 14

platforms. See Figure 24. Through a threat intelligence platform 8

Other 5
0 25 50 75 100 125 150

Figure 24. Event Correlation Technology (Q3.20: n=324)

SANS 2022 SOC Survey 18

The author (Crowley) isn’t suggesting this is necessary or necessarily
advantageous based on the technologies involved. But rather, his
opinion is that the staff’s assessments of performance of the tool drives
replacement of the tool rather than reconsidering the implementation
and programmatic elements surrounding
the technology. This opinion is not derived Does your SOC support nontraditional computing devices
directly from survey responses. It is a fusion such as smart sensors, building devices, building monitoring,
manufacturing, industrial control systems, OT (operations
of assessment and observation of SOCs, the
technologies), and system assets considered as part of the IoT?
marketing pressures within the technology
market, and generalizations around human 1
 artly. Our SOC supports some of our
P
connected, at-risk smart systems.
behavior when faced with criticism over SOC
26  es. Our SOC supports all our
Y
performance. The tools don’t usually speak at-risk smart systems.
up to defend themselves and point out the 33 79  o. We are planning to support nontraditional
N
computing devices within the next 12 months.
deficiencies of implementation or tool operator.
 o. We have no plans to support smart
N
systems.
69
Visibility Across Systems  e haven’t assessed and inventoried smart
W
79 systems yet, but we plan to.
As these SOCs are correlating events, there are
45 Unknown
varying categories of systems into which they Other
provide visibility. This occurs to a greater degree
on different types of systems, so we include Figure 25. Smart Device Support (Q3.16: n=332)
smart devices in Figure 25 and mobile devices in Figure 26.

Most respondents (158, 48%) indicated that they either partially or fully
support smart devices.

When asking what is being used for this, the MDM is What are you using to monitor your mobile devices, extranet, and
clearly ahead, but not by enough to call it a certain choice cloud partner (AWS, Azure, etc.) resources? Select all that apply.

if there are new deployments. Importantly, this question MDM 178

(Q3.18) asked respondents to include all that apply for
EDR 143
monitoring technology for multiple items, and some
XDR 120
respondents answered multiple items. MDM, EDR, and XDR
CASB 83
are specifically intended for this type of nontraditional
EMM 64
compute device monitoring. The authors see continued
development of this technology, in addition to the use of Other 35
0 20 40 60 80 100 120 140 160 180
cloud-provider-native monitoring tools.
Figure 26. Monitoring Technology (Q3.18: n=311)

SANS 2022 SOC Survey 19

Also, on the topic of nontraditional arrangements, we
Are you monitoring your OT (operations technologies) systems
inquired into the techniques for monitoring OT networks. separately or with IT SOC resources? Select the best option.
Of the respondents who monitor OT as part of their SOC,
Together with IT SOC resources 103
most (103, 32%) do so converged with IT systems. This is
closely followed by separately (80, 25%) as the next most Separately 80
Separate systems for
common. Another technique is to use separate technology monitoring but same SOC staff 47

but the same staff for monitoring, and if these two groups Other 8
are considered together (because physical separation of the 0 25 50 75 100

visibility and protection instrumentation is consistent for Figure 27. OT Monitoring Strategy
both groups), then that grouping (127, 39%) exceeds the “together” responses (see Figure (Q3.17: n=326 but n=88 excluded
from chart because there’s no OT)
27). This is important because there appears to be a strong urge for compartmentalization
of these resources onto their own network, and the conceptual boundary appears to
be extended to the defensive monitoring systems as well. This compartmentalization
approach has strong advocates on both sides of the subject in the OT cybersecurity
community. From this survey, it’s about evenly split in terms of how the SOCs monitor OT.

Relationships
What is your SOC’s relationship to your IT operations?
In discussing OT/IT convergence, it’s apropos Note: This question refers to IT operations in whatever form,
to also highlight the SOC to IT operational such as general IT or a network operations center (NOC).

monitoring. It is the opinion of the author Our IT/NOC team is an integral part
of our detection and response,
(Crowley) that there are opportunities for tool although our SOC and IT/NOC activities 119
are not technically integrated.
reuse, converged visibility, collaboration, and
Our SOC and IT/NOC teams work together
only when there are emergencies. 101
coordinated hunting activities if these teams
Our IT/NOC team and SOC team are
are empowered to share data and ideas. kept well-informed through integrative
77
dashboards with shared information,
APIs, and workflow, where needed.
In Figure 28, it looks like a lot of the SOCs
Our SOC and IT/NOC teams have
54
agree. The strongly segmented responses very little direct communication.

“very little direct communication” (54) and We don’t have an IT/NOC team. 37

“there is no relationship” (27) are 19% (81) There is no relationship. 27

of responses, whereas the strongly positive Other 17

“integral part … not technically integrated” 0 25 50 75 100 125

(119) and “…integrative dashboards…” (77) are Figure 28. SOC to IT Relationship
45% of the responses (196). This is a call to action to leverage your scarce resources to (Q3.8: n=432)

further the collaboration or integration of two core operational capabilities: SOC and IT.

The only likely counter-indication the authors see in this situation would be that the
SOC might lose some oversight capability of potentially malicious or negligent system
administrators; the IT admin could also see what the SOC sees about his or her (insider/
malicious) activity and adjust to avoid detections. The value of the visibility as a deterrent
likely outweighs the risk of a crafty insider threat intentionally evading monitoring.

Our advice? Monitor IT admin access by using a behavioral monitoring strategy to identify
patterns that could be attributable to malicious insider activity. This would cover the other
common fear of sharing the security visibility data with IT peers, if an attacker seizes
credentials and is using your tools to see what you see about the attacks. This scenario of
loss of control and use would likely represent a behavioral change from the normal baseline
of that account, giving you a potential alert late in the phase of an attacker intrusion.

SANS 2022 SOC Survey 20

 What is your estimated annual budget for new hardware, software
Investment: licensing and support, human capital, and any additional costs?
The Determining Factor
Unknown 71

Looking at SOC budgets reveals some interesting Less than $100,000 USD 39

observations. First, 30% of respondents (71/240) $100,001–$250,000 USD 28

are not aware of the overall SOC budget. (See $250,001–$500,000 USD 21
Figure 29.) This may be indicative of respondent $500,001–$750,000 USD 14
role according to Q2.5 where 270 (52%) were
$750,001–$1,000,000 USD 13
analyst, administrator, or architect roles.
$1 million–$2 million USD 13
However, the next most popular funding amounts $2 million–$4 million USD 12
in Figure 29 indicate budgets of less than
$4 million–$8 million USD 12
$500,000: $100,000 (39), $100,001–$250,000 (28),
$8 million–$16 million USD 2
and $250,001–$500,000 (21). While regional staff
$16 million–$48 million USD 4
salary variation could certainly play a part here,
this doesn’t represent a realistic view of the Greater than $48M USD 11
0 10 20 30 40 50 60 70 80
investment to run a SOC, especially if there’s a 24/7
Figure 29. Overall Budget (Q3.51: n=240)
performance expectation.

In short, a realistic model of required investment

How would you characterize your process for establishing
should be developed, specifically around the SOC IT/security budget for next-year funding?
team size of 2 to 10 members.
Formal 133
Most respondents (56%, n=133/236) follow a formal
Informal (akin to ad hoc) 62
budget process. (See Figure 30.) But there may be
Unknown 38
issues in actually determining funding.
Other 3
We asked how the funding was determined, and
0 25 50 75 100 125
Figure 31 shows the breakdown. Most respondents
Figure 30. Budgeting Method (Q3.56: n=236)
(n=93) indicated that the SOC management and
organization management work closely on this. How is funding allocated in your organization? Select the best option.
However, the majority still believe that management
Management and SOC leads/managers
does not heed recommendations from SOC leaders work together closely to decide how 93
to allocate funds for cybersecurity.
in allocating funds (n=83+25+25).
Management takes recommendations from
SOC leads/managers but ultimately decides
how to allocate funds, sometimes against 83
SOC management’s recommendations.
Management takes recommendations
from SOC leads/managers but
frequently goes against SOC 25
management’s recommendations.
Management pays little heed to the
recommendations of SOC managers
and allocates the cybersecurity 25
budget as they see fit.
Other 14
0 20 40 60 80 100

Figure 31. Funding Allocation (Q3.52: n=240)

SANS 2022 SOC Survey 21

While there seems to be more cooperation and less frustration
Budget versus Metrics
expressed with organizational management not heeding SOC
Formal Informal (akin to ad hoc) Unknown
management advice on funding, a problem may still remain,
25
25
suggesting that a formal budgeting method should be derived
from metrics showing the need for the work. So, we checked to
20
see if the respondents who said they provided metrics tended
to operate in an environment with a formal budgeting process. 16
15
Figure 32 shows that there is a clear difference, where formal
budgeting is more present where metrics are delivered. We
10
suggest taking this as an attribute of maturity for your SOC. 8

5
5
3 3
Measuring for Success 2 2
1
0
Yes No Unknown
Metrics are a critical component of the SOC’s interaction with Figure 32. Budget Method and Metrics (Q3.35 and Q.56: n=240)
the organization, in the authors’ opinion. Yet, most of the metrics
used fail to effectively characterize the value the SOC provides to
the business. Admittedly, we’re taking the optimistic view that the Does your SOC provide metrics that can be used in your
reports and dashboards to gauge the ongoing status
SOC does provide value and could calculate it.
of and effectiveness of your SOC’s capabilities?
Among respondents, 70% (193/274) indicated that they provide
metrics to accomplish this communication. See Figure 33.
30
We asked about satisfaction with these metrics, and of those who
answered, 78% (136/187) are either satisfied (92) or very satisfied Yes
51
(54) with the metrics. See Figure 34. No

We have been conducting this survey for several years, so we 193

Unknown
looked back to see what the responses from previous years said
about metrics and metrics satisfaction. We are using percentages
to compare because the numbers varied in each year. In 2019,
57% said they provided metrics, but the survey didn’t include Figure 33. Metrics Provided (Q3.35: n=274)
questions on metrics satisfaction. In 2021, 77% provided metrics,
of which 67% were satisfied. This year’s survey shows a 7% How satisfied are you with current SOC metrics used
drop from 2021 in providing metrics but an overall increase in in reports and dashboards to help gauge the ongoing
status and effectiveness of your SOC’s capabilities?
satisfaction of about 12% on how these metrics help gauge the
effectiveness of the SOC. 1

Very satisfied
40
54 Satisfied

 ot satisfied/Need
N
serious improvement

92 Unknown

Figure 34. Metrics Satisfaction (Q3.36: n=187)

SANS 2022 SOC Survey 22

Looking at Metrics That Tie Value to Effectiveness Have you calculated a “cost-per-record”
from an actual incident?
But how to tie funding to metrics that ascertain how SOC value
is determined? Questions 41 to 44 aimed to determine this by
first investigating cost-per-record values related to cybersecurity
31
incidents and then exploring potential methods of calculating loss
69 Yes
prevention provided by the SOC. Ultimately, it is the promise of loss
prevention that compels organizations to fund SOCs. Let’s trace out No
what the respondents said.
Unknown
139
It seems like most respondents are not calculating cost-per-record
values. Figure 35 shows that 58% answered that they don’t calculate
this. The author’s recommendation to these respondents and the
readers of this report: Start this effort. It will be imprecise at first, Figure 35. Cost-per-Record Calculation
so be patient and nurturing. But this tactical operational value Percentage (Q3.41: n=239)
should be correlated to funding to secure
Value per Record Type
the right amount of funding to protect the
organization’s information. Other Credit card Customer account information Internal user account

16
16
Where they have been calculated, the numbers
14 14 14
are all over the map, with no clear consensus, 14

as is displayed in Figure 36. 12

12 12 12 12
11
Consolidating this to eliminate the distinction 10
10
9 9 9 9 9
between the record types, there’s still no 8
8
7 7 7
clarity on which is the most common value, 6
6
as shown in Figure 37. This suggests that the 5

calculations aren’t consistent and that the 4

3
conditions of intrusions aren’t consistent 2
1 1
among the respondents. A per-record cost 0
$25+ $10–$25 $5–$10 $1–$5 < $1 Unknown
should be consistent between organizations
Figure 36. Cost-per-Record by Type (Q3.42: n=42)
for the same record type. The takeaway here is
that this measurement isn’t done often enough
All Record Types
and has yet to achieve consistency across
50
various organizations. But there’s still value to 47
44
the effort. As one example, your insurers are 40 39 38
using an estimated value for their calculations
of your insurance premiums related to this. 30
25 24
You should have an idea of handling costs and
20
impact costs.
10

0
$25+ $10–$25 $5–$10 $1–$5 < $1 Unknown

Figure 37. User Account Cost-per Record (Q3.42: n=42)

SANS 2022 SOC Survey 23

Of course, we went on to ask the next obvious question (in the authors’
Do you have an estimated or calculated “incident
manner of thinking): whether respondents are calculating the value of the with a SOC vs. incident without a SOC” value?
SOC by using some form of comparison between the SOC intervening and
the SOC not intervening. Like cost-per-record shown in Figure 37, most
(58%, n=140/244) said that they don’t calculate this, as shown in Figure 38. 30

You can see where this is going. If the respondent is calculating it, we 74 Yes

wanted to know. So, the next question asked is what help the SOC is No
(if it is, in fact, helping). Only a small number said the SOC’s existence
140 Unknown
made the handling effort more costly than without the SOC (n=7), and
the impact more costly (n=4). But many more saw 10% handling (n=19)
and impact (n=24) reduction. The most popular response for handling
reduction was 50% reduction (n=22), with the incident impact reduction
Figure 38. Estimated or Calculated Value
of 50% for (n=18) respondents (see Figure 39). This provides a
Provided by the SOC (Q3.43: n=244)
compelling story to show value to management when
you go into that formal budget to try to assure the Estimated Relative Handling Cost with SOC
organization management works closely with you to Incident cost Handling cost

allocate the SOC budget. Multifold (2x or more) 7

reduction of cost 1

90% reduction of cost 12

14
Summary 50% reduction of cost 18
22
24
We’ve covered a lot of territory with this survey. It 10% reduction of cost
19
represents a major expansion of our linguistic offering, Actually, cost is higher with 4
the SOC than without it 7
and we’re optimistic that we’ll soon offer the survey
8
natively in other languages. N/A or unknown reduction of cost
8
0 5 10 15 20 25
We have taken a capabilities-based approach to the
Figure 39. Estimated SOC Value
concept of a SOC. The way to use this survey is to assess
(Q3.44: n=73)
your capabilities compared to your peers. In doing so, you’ll see that most SOCs have the
same capabilities but accomplish them through varying levels of internal performance
and outsourcing. Most SOCs deliver metrics, and many are starting to deliver calculations
on the value of the defense provided.

For SOCs with lesser maturity, adding missing capabilities is the next step. Doing so
through outsourcing often provides speed and high-value proposition without the
accompanying tailoring and customization. The next step would be assuring performance
of metrics. For more mature SOCs, delivering calculations related to data protected and
loss prevention provided is the step to take.

SANS 2022 SOC Survey 24

Follow These Calls to Action
Match your organization against our demographics and related results.
4 Compare the size of your organization’s workforce versus survey-reported workforce size.
4 Check SOC staffing levels in comparison to your business growth over time.
4 Evaluate the capabilities of your SOC against those reported in this survey as commonly
present to develop what you lack.
4 Compare what survey respondents frequently outsource with what your organization
outsources to evaluate whether there’s an opportunity to outsource capabilities or bring
them back in-house.
Be sure that the trends which are important to your organization are going in the right direction!
4 Define and track the critical metrics to your organization.
4 Make sure you have at least one metric that depicts the value your SOC provides.
 xplore the “Technology: What Is Getting a Passing Grade?” section to see how your
E
organization stacks up against the community.
4 Compare your implementation of technology with how other organizations have
implemented it.
4 Ask the following hard questions:
• Are you still just investing in technologies that most other organizations have
successfully transitioned to production?
• When you buy technology for your SOC, do you have a plan to get it deployed to full
production?
Are you considering the effectiveness of the SOC in budgeting for its resources?
4 Make sure your organizational management listens to SOC leadership where important!
4 Keep your metrices in mind!

Sponsor

SANS would like to thank this survey’s sponsor:

SANS 2022 SOC Survey 25