Methodology and Feedback about Systematic Cybersecurity Experts
Auditing in Belgium
Christophe Ponsard
1
, Jeremy Grandclaudon
1
and Nicolas Point
2
1
CETIC Research Centre, Charleroi, Belgium
2
Multitel Research Centre, Mons, Belgium
Keywords:
Cybersecurity, SME, Assessment, Lightweight Framework, Survey.
Abstract:
Increasing the maturity of SMEs with respect to cybersecurity threats is crucial as they are less prepared and
less resilient. They are also increasingly exposed and targeted by malicious actors. Providing support means
ensuring an effective ecosystem is available to help companies all along the process. Resources have to be
available, from raising awareness to performing audit, increasing protection and building response capabilities.
In this paper, we report about the progress achieved after one year of deployment of a Belgian cybersecurity
initiative focusing on SMEs. An important goal is to make sure minimal requirements will be checked and
enforced by cybersecurity experts while letting them use their own methodology. We explain how the expertise
is validated using an evaluation grid based on the NIST Cybersecurity framework and CIS 20 criteria directly
reflecting protection priorities for SMEs. We also highlight some interesting characteristics and lessons learned
in our data set of 25 experts evaluated so far.
1 INTRODUCTION
Small and Medium Enterprises (SMEs) are a strong
driver of the world-wide socio-economic develop-
ment. At European level, they contribute to more
that half of the economic value and hire about two
third of the workforce (Muller et al., 2015). Given the
fast move to a digital society, Information Technology
(IT) has become business critical for SMEs and needs
to be protected against cybersecurity attacks. How-
ever, SMEs tend to be highly focuses on their busi-
ness and less on the quality of their process and IT in-
frastructure. They may also lack expertise or time to
be fully protected against cybersecurity threats. They
may also wrongly assume their size will not attract
attackers. In the past few years, the rate of attacks
targeting them has increased dramatically with esti-
mations around 60% to 70% (Keeper Security, 2018).
As SMEs are also less resilient than bigger compa-
nies: more than half of the hacked SMEs do not re-
cover and cease their activity a few month after an
attack (NCSA, 2018).
The need to support SMEs in the management
of cybersecurity threats is widely acknowledged. At
European level, many organisations such as ENISA,
SME Alliance, the European Commission, European
Cybersecurity Organisation (ECSO) are devoting ef-
fort in this area. At national level, most countries
have set up some form of program to raise aware-
ness and to provide guidance as reported in our previ-
ous work (Ponsard and Grandclaudon, 2018; Ponsard
et al., 2019). Examples of such initiatives are the Cy-
berEssentials in UK (UK Gov., 2016) or the Finnish
Cyber Security Certificate (FINCSC, 2018).
In Belgium, the effort is currently structuring at
the two levels, depicted in Figure 1:
The Regional Level is responsible for the non-
certifying audits which aim at ensuring SMEs
have identified key cybersecurity risks and have
taken adequate measures to manage them. They
are helped in this task by a pool of cybersecu-
rity experts. In Wallonia, the driving initiative is
called ”Keep IT Secure” (KIS for short) and is led
by Digital Wallonia (Digital Wallonia, 2018). Ex-
pertise is validated by an advisory board which
checks that experts master cybersecurity funda-
mentals and are able to conduct assessment and
improvement activities with SMEs. A funding
scheme is also available through specific cyber-
security vouchers that will only support the inter-
vention of validated experts.
The National Level is concerned about providing
certification based on a light certification scheme
inspired by ISO27K (ISO, 2013). It under de-
754
Ponsard, C., Grandclaudon, J. and Point, N.
Methodology and Feedback about Systematic Cybersecurity Experts Auditing in Belgium.
DOI: 10.5220/0009190307540759
In Proceedings of the 6th International Conference on Information Systems Security and Privacy (ICISSP 2020), pages 754-759
ISBN: 978-989-758-399-5; ISSN: 2184-4356
Copyright
c
2022 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
Figure 1: KIS Ecosystem in Belgium and scope of this survey.
velopment by Center for Cybersecurity Belgium
which is also supporting awareness actions such
as cybersecurity guide for SMEs (CCB, 2016).
The purpose of this paper is to detail how the KIS
initiative is practically organised to validate that its
experts are qualified to help SMEs. The aim is to be
sure they have the required competencies while letting
them enough freedom about the methods and tools
they want to use in this process. We do not directly
report here about the cybersecurity maturity level of
SMEs although the required expertise is calibrated to
address the threats to which they are exposed.
Our work is structured as follows. First, Sec-
tion 2 describes how we designed the expert valida-
tion phase based on reference frameworks such as the
NIST Cyber Security Framework (CSF) and basic se-
curity controls such as specified by the Center for In-
ternet Security (CIS). Then Section 3 reports on our
application of the resulting expert validation toolkit
for the first nine months of its application. During that
period, more than 25 experts have been interviewed
and all the data has been recorded. Even if the process
is still ongoing, we already give a first sketch analysis
of key characteristics of our local ecosystem which
we believe is quite representative. Our next contribu-
tion is to detail some lessons we learned so far along
in Section 4. Finally, Section 5 draws some conclu-
sions and presents our planned work.
2 EXPERT ASSESSMENT
PROCESS AND TOOLKIT
Keep IT Secured emerged after a long maturing pe-
riod started in 2017 and involving public authori-
ties, research centres, a local cybersecurity cluster and
end-user SMEs through specific awareness-raising
events. During our elaboration process, we were in-
spired by other European initiatives detailed in (Pon-
sard and Grandclaudon, 2018).
Unlike other domains where posterior control is
possible, the sensitive dimension of cybersecurity re-
quires ensuring, prior to any service, that service
providers are qualified experts w.r.t. their ability to:
Identify and manage risks related to the various
types of information held by the company espe-
cially in the SME context
Implement adequate protection mechanisms for
the various types of systems that contain and man-
age information
The following key abilities are required and need to
be checked:
General purpose expertise in cybersecurity and
reference frameworks
Ability to embrace all SME-specific cybersecurity
issues
Ability to carry out organisational and technical
audits, according to a well-established methodol-
ogy that may be their own
The criteria are based on international standards
and inspired by European labels. These include the
NIST CSF (NIST, 2014), the 20 key criteria of the
Center for Internet Security (CIS, 2016) and similar
approaches undertaken in other countries (e.g. Cy-
berEssentials in Great Britain). These criteria cover
the main steps of a cybersecurity risk management
approach: identification - protection - detection - re-
sponse - recovery. The precise methodology is left to
the discretion of the auditor but to take into account
the 20 key criteria of the CIS. The criteria are also
only significant in their contextualisation in relation
to the risks incurred. The capability to perform risk
Methodology and Feedback about Systematic Cybersecurity Experts Auditing in Belgium
755
assessment is also evaluated during the interview pro-
cess based on case studies where basic risks and then
more complex risks are progressively injected.
2.1 General Interview Process
As a reminder the interview process considered
here concerns the validation of cybersecurity ex-
perts. We do not report about audits those experts
will carry out later inside SMEs, once their exper-
tise has been validated.
KIS works on a personal basis and not on a
company basis. So interviews are carried out in-
dividually, even if several experts from the same
organisation have applied. It is led by two special-
ists from the cybersecurity expert advice centre and
lasts for a maximum of two hours. After having wel-
comed and explained the KIS framework, the expert
is asked to give an overview of his professional train-
ing and experience.
KIS does not impose a methodology on the
provider, but checks the coverage of fundamentals
that guarantee a good mastery of cybersecurity within
SMEs. To do this, a few concrete scenarios are pre-
sented and serve as support for a dynamic discussion
to evaluate the following aspects:
Identification of risks in relation to the context of
the SME
Main strategies from prevention to recovery using
NIST CSF at top level
Use of basic controls, based on a detailed check-
list inspired from CIS20 but structured around
NIST CSF
2.2 Checklist
The check-list is organised as a spreadsheet divided
in seven main tabs. First an overview tab for filling
the administrative and evaluation information, then an
help task explaining the global structure and then five
tabs corresponding to the ve NIST CSF categories.
Those tabs are easy to fill using click control and have
room for comments. While an assessor is asking a
question, the other is filling and checking to prepare
more questions on issues that have not yet been cov-
ered by expressing them in the scope of the supporting
cases. An interesting way to use it is to upload it on
a collaborative platform so it can be filled collabora-
tively. So the sheet can efficiently help to both drive
and control the interview and allows the interviewers
to hop from a topic to another while keeping track of
everything.
As an example the Identify tab is depicted in Fig-
ure 2. Each tab is composed of a main categories
and then more detailed sections with checks organ-
ised by levels: basic, intermediate and advanced. As
the interview progresses, basic checks are first cov-
ered and then progressively more complex ones but
the discussion can also go more quickly deeper in de-
tail on some topics and already cover intermediate and
advanced topics. The coverage of all advanced topics
is not mandatory and left to the interviewers.
Note that the description remains short reminders
about topics to cover during the interview. Their for-
mulation is totally generic and technology-agnostic.
However the interview itself will generally introduce
some concrete scenarios (see next section) but also
Figure 2: Global summary of the ”Identify” tab of the KIS assessment spreadsheet.
ICISSP 2020 - 6th International Conference on Information Systems Security and Privacy
756
technological context. E.g. a Cloud storage may be
introduced as part of the infrastructure and will trigger
the need of more specific checks present in other parts
of the sheet. As a result, the filling will not be lin-
ear although a global progression from basic to more
advanced topics will be followed. So the sheet also
ensure that no important topics are overlooked.
2.3 Supporting Scenarios
Scenarios are not described in detail here for evident
reasons but are composed of two main cases: one
very basic, focus on a SME with basic needs of IT
infrastructure and one more specialised in order to be
able to ensure a wide coverage. Each case is explored
with a raising level of threats. The first case is typi-
cally inspired from a traditional domain with a limited
IT support: basic network, few workstations, config-
uration close to domestic use. It then develops the
scope of the business activities to increase its depen-
dencies on information technology through on-line
orders, transactions, more complex networks, remote
access, etc. A complementary case is also used to ex-
plore more specific problems such as high availability
and sensitive information, e.g. in health or logistics.
2.4 Outcome of the Assessment
Three types of outcome are possible:
Positive, possibly with some points of attention:
the expert is integrated in the KIS pool for three
years.
Positive with Conditions, i.e. with some improve-
ment points requiring follow-up. The expert is in-
tegrated into the KIS pool but will have to undergo
a control interview after one year.
Negative: the expert is not allowed to integrated
the KIS pool. The reasons can be technical but
also the lack of ability to conduct an audit, a lack
of general vision (even if the person can be very
sharp on a field) or a lack of cybersecurity scope
(e.g. pure GDPR consultants, see Section 4). The
candidate can also be redirected to other types of
services or be advised on how to improve. The
candidate can retry a new interview after a mini-
mal period to acquire the missing expertise.
3 ANALYSIS
This section reports on our analysis of 25 evaluations
carried out between March and November 2019 (most
of them over the past four months). The results are
anonymised and presented as aggregated statistics on
different dimensions of the questionnaire (consider-
ing the ve NIST CSF phases and the global matu-
rity level). We also look at the most and less prob-
lematic (type of) controls, e.g. controls that experts
tend to forget despite their importance. Before going
into those details the sample is characterised based on
their main domain activities.
3.1 Domain-level Analysis
Our current data set is composed of 25 experts coming
from 21 different companies. Most of the companies
are of very small size as described in Table 1. For the
smallest companies, the domains of activity closely
match the (single) expert domains of expertise. Some
intermediate companies are organised as network of
experts focusing on cybersecurity. The bigger compa-
nies have a wider range of activities including a pool
of cybersecurity experts.
Table 1: Distribution of company size.
Size # companies
1 10
2-5 7
6-10 2
>10 2
Total 21
Table 2 presents the overall split across the main
domains of activities. Its also reports about the var-
ious evaluation outcomes (either full accept, condi-
tional accept or reject as described in Section 2). Note
that some experts may have more than one expertise,
so the total does not add up to our total number of
experts here.
Table 2: Distribution of experts across domains.
Domain Total # full
accept
# cond.
accept
#
reject
devops 7 3 3 1
cybersecurity 6 6 0 0
web developer 6 3 2 1
GDPR expert 6 4 0 2
IT audit/strategy 5 4 1 0
Interesting point is that quite a large number of
devops and web developers applied for KIS but they
often remain too technical. Some lack a wider organi-
sation level perspective enabling risk analysis or sim-
ply auditing capabilities. Without too much surprise,
experts focusing fully on cybersecurity were all ac-
cepted. The point with them was to make sure they
adopt a broad and staged view to deal with the SME
Methodology and Feedback about Systematic Cybersecurity Experts Auditing in Belgium
757
context. Another interesting category is IT auditor
and consultants in enterprise architecture. Those have
very good risk assessment capabilities and are usually
able to master the required level of technical expertise
in cybersecurity.
3.2 Phase-level Analysis
The phase level analysis is based on the five NIST
CSF phases. Table 3 provides the percentage of can-
didates in each of our four categories of readiness for
applying some control, from fully spontaneous to to-
tally discarded. We also provide a global maturity in-
dex for each phase which is computed using the fol-
lowing weighted means resulting in a perfect score of
10 (everything spontaneously checked) and a worst
score of 0 (everything discarded):
30.#{spontaneous} + 20.#{questioned} + 10#{basic}
3.#{audited}
Table 3: Distribution of experts across domains.
Phase Spont. Quest. Basic Disc. Score
Identify 60% 10% 23% 11% 7,4
Detect 62% 15% 22% 6% 7,9
Protect 78% 5% 15% 7% 8,6
Respond 60% 3% 21% 21% 6,9
Recover 53% 6% 24% 21% 6,5
The resulting score is the higher for the protect
category, followed by detect and identify while later
respond and recovery phases are less well investi-
gated. Note there might be some bias due to the fact
that more time tend to be spent on earlier phases dur-
ing the interviews.
When digging further in the answers (not detailed
here) the top ve categories of controls points are re-
lated to our top two scores:
Awareness (protect)
Access control and identity management (protect)
Data security (protect)
Protection technologies (protect)
Response to anomalies (detect)
While the bottom five categories of controls points are
more mixed across categories:
Recovery planning (recover)
Communication (recover)
Maintenance (protect)
Risk management (identify)
Response to anomalies (respond)
4 SOME LESSONS LEARNED
At this stage, we are still learning a lot from the on-
going interview process. However, we could already
extract a few interesting points worth being shared.
Overall Reactions of Experts. When launching
our initiative, we were a bit concerned about reac-
tions of experts who would not pass the evaluation,
especially coming from an unregulated context where
self-proclaimed IT experts could start working with
SMEs. In the end, in most cases, the process revealed
quite smooth. Most experts without track records
have actually turned down the interview request and
were removed from the KIS pool. Some experts came
to have a try but knowing about their limits and more
eager to listen to recommendations. As the goal is to
help SMES, if a good potential is detected, our goal is,
of course to propose an improvement path and another
evaluation can be scheduled later on. For the experts
that have all the competencies, it is also rewarding as
they get some recognition and they also see that some
clean-up is done at the benefit of the end-user SMEs.
We were also surprise with some elaborated and docu-
mented methods some experts had developed, includ-
ing company awareness-raising actions for some of
them.
Responsability Issue. A interesting point is to see
how companies are dealing with the fact that the KIS
is granted to an individual expert and not to a com-
pany. Two alternative behaviours can be observed: ei-
ther a single expert is labelled for the company. This
does not prevent other experts to be involved but all
reports must be endorsed by the validated expert who
engages it responsibility on the quality of the work de-
livered to the SME. Another option is to send multiple
experts to the evaluation. This second option gives
more visibility on the real number of practising ex-
perts. However, the process still takes some time and
effort to complete so all potential experts of a com-
pany are not expected to apply for it.
The GDPR Effect. GDPR was and remains a great
incentive for raising awareness about cybersecurity
inside SMEs. A downside is that some consulting
companies active in GDPR also position themselves
on cybersecurity based on their data protection focus.
However addressing GDPR is not enough to cover
the whole scope of activities required for protecting
SMEs from cybersecurity threats. In the end, the point
revealed a bit touchy: without banning GDPR experts,
they can apply provided they are able to address the
full scope of a global cybersecurity audit. As it has
ICISSP 2020 - 6th International Conference on Information Systems Security and Privacy
758
only a partial overlap with GDPR, a pure GDRP con-
sultant will not pass the checking process.
Collecting Evolution Needs. The interview pro-
cess was also the opportunity to get feedback from
the field. Although our assessment grid was designed
to fit SME maturity and was validated before start-
ing our interview campaign, some checks proved too
advanced like forensics analysis or direct cooperation
with local CERT. Other checks may need more de-
tailed breakdown like making sure the security pol-
icy matches the company purpose (after identifying
both). This evolution is planned on an annual basis
and will be discussed with an advisory board involv-
ing cybersecurity professors from local universities
and with all the interested experts part of KIS with the
support of a local cybersecurity cluster. In addition to
help us improving our criteria, those meetings also
help to define the path to increase the maturity level
of SMEs engaged in an cybersecurity improvement
process while keeping attracting news SMEs through
specific awareness-raising actions. Last but not least,
we expect this will also be the opportunity to share
some good practices between experts.
5 CONCLUSION & NEXT STEPS
In this paper, we reported about the ongoing evalua-
tion process of cybersecurity experts carried out us-
ing the Keep IT Secure framework at the regional
level in Wallonia (Belgium). We showed how the
framework is aligned with our national perspective
and international standards such as NIST cybersecu-
rity and CIS20 while also providing a path to ISO27K.
Based on those strong references, we designed an au-
dit methodology for validating the expertise of cy-
bersecurity companies that will help end-user SMEs.
The resulting evaluation grid is used in a role-playing
game that allows the advise centre to check how well
an expert covers the full spectrum of key activities and
controls when dealing with an SME case. The process
does not impose a specific methodology but follows
the expert methodology. This enables to assess how
effective it is and to make some recommendations.
Second, we also reported about the analysis of the
interesting data set collected during our interviews.
Thanks to the systematic use of our check-list, we
could perform a quite interesting analysis. Although
it requires some extensions both in size and scope,
we could already point out interesting characteristics
and some lessons learned. We believe our approach
could interest other countries dealing with the prob-
lem of providing a reliable expert network to help
SMEs tackle the cybersecurity threats.
Our future work is to update our analysis based on
more audits. At this point an estimated 60% of active
cybersecurity companies have been covered. We also
plan to evolve our framework based on the collected
feedback and to make it available more widely for
those interested in sharing similar approaches. At a
more global scale, we are working on the interconnec-
tion of our work with emerging certification scheme
at the Belgian federal level (see Figure 1). We are also
providing feedback at European level through specific
projects like SPARTA and organisations like ECSO.
ACKNOWLEDGEMENTS
This research was partly supported by Digital Wal-
lonia and the SPARTA H2020 project (nr. 830892).
We also thank Infopole, DGO6 and companies of the
Walloon cybersecurity cluster.
REFERENCES
CCB (2016). Cyber Security Guide for SME. http://www.
ccb.belgium.be/en/guide-sme.
CIS (2016). CIS Controls V6.1. https://www.cisecurity.org/
controls.
Digital Wallonia (2018). Keep IT Secure. https://www.
digitalwallonia.be/keepitsecure.
FINCSC (2018). Finnish Cyber Security Certificate. https:
//www.fincsc.fi.
ISO (2013). Iso/iec 27001 information se-
curity management. https://www.iso.org/
isoiec-27001-information-security.html.
Keeper Security (2018). 2018 State of Cybersecurity in
Small and Medium Size Businesses study. https:
//start.keeper.io/2018-ponemon-report.
Muller, P. et al. (2015). Annual Report on European SMEs
2014/2015. European Commission.
NCSA (2018). Stay Safe Online - Cybersecurity Awareness
Toolkit for SMB. National Cyber Security Alliance.
NIST (2014). Cybersecurity Framework. https://www.nist.
gov/cyberframework.
Ponsard, C. and Grandclaudon, J. (2018). Survey and guide-
lines for the design and deployment of a cyber security
label for smes. In 4th Int. Conf. on Information Sys-
tems Security and Privacy (Revised Selected Papers),
Funchal, Madeira, Portugal.
Ponsard, C., Grandclaudon, J., and Bal, S. (2019). Survey
and lessons learned on raising SME awareness about
cybersecurity. In 5th Int. Conf. on Information Sys-
tems Security and Privacy, Prague, Czech Republic.
UK Gov. (2016). Cyber essentials. https://www.
cyberaware.gov.uk/cyberessentials.
Methodology and Feedback about Systematic Cybersecurity Experts Auditing in Belgium
759