×

Tag: ServiceNow

The ServiceNow Job Nobody Trained For Is The One Everybody Now Needs

Posted on by Dan Wells

In the last fortnight ServiceNow patched three vulnerabilities in its AI Platform that each scored a perfect 10.0, none of which required authentication to exploit. In the same window it expanded the AI Control Tower to govern AI across any system in the enterprise, whoever built it. I think those two events point at the same career opportunity, and at a profile the market is short of.

What happened

The advisory came out on 27 August. CVE-2026-18885 allowed arbitrary code execution. CVE-2026-18886 allowed privilege escalation through improper access control. CVE-2026-74820 was a SQL injection flaw against the underlying database. All three scored 10.0, none required authentication or user interaction, and all three were rated low complexity. A fourth, a sandbox escape, scored 8.7.

ServiceNow patched hosted instances directly and released hotfixes for self-hosted deployments across Xanadu, Yokohama, Zurich and Australia.

Separately, coverage in early September described the AI Control Tower being expanded to discover, observe, govern, secure and measure AI deployed across any system, not just ServiceNow’s own. I have not pinned the exact announcement date, so take the timing loosely. Alongside that, this year’s acquisitions tell a consistent story: Armis for asset visibility, Veza for identity governance, and Sweep, which ServiceNow confirmed as completed on 1 September, for a governance layer spanning ServiceNow, Salesforce and HubSpot.

Read those together

ServiceNow is becoming the place enterprises govern all their AI. Which makes the ServiceNow platform itself a very high-value target, and makes the person who secures it a very hard person to replace.

The role that is forming

There is a gap opening between two established professions and it does not yet have a settled job title.

Security teams own vulnerability management but frequently lack depth on the ServiceNow platform itself. Platform teams own the instance but sit inside IT service management, where security advisories are not the daily rhythm. Advisories land in the space between and both sides assume the other has it.

The people who can occupy that gap need three things at once: real hands-on platform knowledge, enough security literacy to read an advisory and work out what it touches in a specific configuration, and enough standing to force an out-of-cycle change when it matters. That is an unusual combination. My expectation is that scarcity of that kind gets priced eventually, though I am not going to pretend I can show you a published benchmark for a role that does not yet have a settled title.

It is also getting broader. When agents hold credentials and act autonomously inside your environment, they become identities that need governing like any other. The controls we built over two decades for human insider risk mostly have not been applied to them, because attention has been on the productivity gain. Somebody has to close that, and it will not be a pure security person or a pure platform person.

How to move towards it

1. Start with what you already have. If you are a platform person, you are closer to this than a security generalist is. Platform depth is the harder half to acquire. Security literacy on top of real ServiceNow experience is a shorter journey than the reverse.

2. Learn the machine identity side. The Zurich release documentation covers capability for securing machine-to-machine integrations, including the Vault and Machine Identity consoles. Check the release notes for what is in your version, then get hands-on. This is the concrete, learnable part of the agent governance problem, and it comes up far less often in conversations than the strategy layer does.

3. Read the advisories, properly. Not the headline. Go and understand which components each CVE touched and why. Doing that four times will teach you more about the platform’s attack surface than any course, and it costs nothing.

4. Build one governance story. How you handled access, auditability, or the decision not to automate something. Even small. Hiring managers are nervous about exactly this and hardly any candidate raises it unprompted.

5. Learn to talk about risk in business terms. The question that decides these interviews is some version of: an agent takes a wrong action against customer data at 2am on a Sunday, what happens? If you can answer that with a clear account of accountability, detection and rollback, you are ahead of nearly everyone.

A realistic word on timing

I am not going to tell you this is an easy pivot or that a certification unlocks it. It is a genuinely senior profile and it usually forms out of an architect or senior developer who has been pulled into governance work informally and then made it their identity.

What I will say, and I want to be clear this is my read rather than a measured finding, is that the demand side is visible in what ServiceNow is buying and building, while the supply side looks thin to me from where I sit. If that is right, it is a good window. If I am wrong about the supply side, the worst case is that you have gone deep on security and governance skills on a platform that is becoming the enterprise control plane for AI, which is not a bad place to be wrong.

If you are three or four years into ServiceNow and wondering where to go deep, this is the most defensible direction I can point you at.

Thinking about your next move? We speak to ServiceNow professionals across the UK, Europe and North America every day and we are happy to give you an honest read on your profile, whether or not you are actively looking. Start a conversation.

Sources: ServiceNow security advisory, 27 August 2026 · SecurityWeek, 31 August 2026 · ServiceNow newsroom

Three CVSS 10.0 Flaws, No Authentication Needed. Who In Your Organisation Is Applying That Patch?

Posted on by Dan Wells

On 27 August, ServiceNow published a security advisory covering four vulnerabilities in the AI Platform. Three of them scored a perfect 10.0. None of them need authentication, none need user interaction, and all three are rated low complexity to exploit. If you run ServiceNow self-hosted, the patch is your job, and the clock started two weeks ago.

What was disclosed

  • CVE-2026-18885, CVSS 10.0. Code injection allowing arbitrary code execution, with potential access to and modification of data.
  • CVE-2026-18886, CVSS 10.0. Improper access control allowing an attacker to create or modify arbitrary data and elevate privileges.
  • CVE-2026-74820, CVSS 10.0. SQL injection allowing arbitrary SQL statements against the underlying database, and access to instance data beyond what was intended.
  • CVE-2026-6876, CVSS 8.7. A sandbox escape, also exploitable without authentication.

ServiceNow patched its hosted instances directly. Hotfixes were released for self-hosted deployments across the Xanadu, Yokohama, Zurich and Australia release families. The company says it has no evidence of exploitation and that the issues came out of its own research and responsible disclosure programmes.

The split that matters

ServiceNow states it patched hosted instances. Self-hosted deployments need the hotfix applied by you. Either way, confirm remediation status against your own instance list rather than assuming it, particularly if you have multiple instances or any inherited from an acquisition.

The gap between disclosure and patching

Jason Brown, director of counter fraud operations at iCOUNTER, made the point sharply in SecurityWeek’s coverage. Self-hosted organisations now have to find, schedule and apply the patch themselves, and in many organisations that takes weeks rather than days. During those weeks an unauthenticated attacker with a working exploit has a real shot at systems sitting next to HR records, vendor onboarding and finance approvals. His advice was not to wait for the normal patch cycle and to confirm it is applied this week.

That advice was published on 31 August. It is nine days old.

Why I’m writing about this on a recruitment blog

Because “find, schedule and apply” is not a technical instruction. It is a description of a person, and in a lot of ServiceNow estates that person does not clearly exist.

ServiceNow platform security tends to fall between two teams. The security function owns vulnerability management but often has limited depth on the platform itself. The platform team owns the instance but sits inside IT service management, where security advisories are not the daily rhythm. So the advisory lands, both teams assume the other is handling it, and the patch cycle quietly absorbs something that needed to be treated as urgent.

The organisations that handle this well have one named person who owns ServiceNow platform security specifically. Not security in general. Not the platform in general. The overlap. And that is a genuinely awkward role to fill, because it needs someone who can read a ServiceNow advisory, understand what it touches in your configuration, and have the authority to force an out-of-cycle change.

Four questions worth asking this week

1. Are we hosted or self-hosted, and does everyone in the room agree? It sounds trivial. In organisations with multiple instances and a history of acquisitions, it frequently is not.

2. Who received the 27 August advisory, and what did they do with it? Not who should have received it. Who did.

3. Do we have a route to apply an emergency ServiceNow patch outside the normal change window? If the answer involves a CAB meeting three weeks out, you have found your problem.

4. Is ServiceNow platform security in anyone’s objectives? If it is not written down as someone’s responsibility, it is nobody’s.

The wider pattern

This is the third time this year the same shape has appeared. ServiceNow ships capability at pace, and the operational burden of running it safely lands on customers who have not resourced for it. The AI Platform is now the layer holding your workflow logic, your identity data and your audit trail. Three unauthenticated 10.0s in that layer is a serious thing, and the fact that it was found and fixed responsibly is genuinely reassuring. What is less reassuring is how many organisations will not have applied the hotfix by the end of this month.

If you are self-hosted and you cannot say with confidence that the patch is in, that is today’s job, not this quarter’s.

Need someone who owns the overlap between ServiceNow and security? It is one of the harder profiles in the market and we recruit for it regularly. We can tell you what it pays and how long it takes before you commit to a requisition. Talk to us.

Sources: ServiceNow security advisory, 27 August 2026 · SecurityWeek, 31 August 2026 · BankInfoSecurity, 1 September 2026

Do you need a ServiceNow developer, technical consultant or architect?

Posted on by Dan Wells

Roughly a third of the ServiceNow roles that land on my desk have the wrong title on them.

Not slightly wrong. Wrong in a way that costs the client six weeks and a rejected offer.

The job spec says developer. The interview questions are about integration design and platform governance. The salary band is set at developer level. The candidates who could actually do the work take one look at the money and pass, and the ones who fit the money can’t answer the questions.

Here is how we help clients sort it out before the advert goes live.

The three roles, in plain terms

Developer. Builds what has been specified. Flows, scripts, UI, catalogue items, integrations that someone else has designed. Give a good ServiceNow developer a clear ticket and they will close it. Ask them to sit in front of a CFO and challenge the business case for a Now Assist rollout, and you have put them in the wrong room.

Technical consultant. Sits between the client and the build. Runs the workshops, translates what the business says it wants into what the platform can actually do, then either builds it or hands it to developers. This is the role most people mean when they write “developer” in a job spec. It needs the technical depth of a developer plus the willingness to be in a room with stakeholders who disagree with each other.

Architect. Owns the shape of the platform. Instance strategy, data model, integration patterns, upgrade path, what goes on ServiceNow and what does not. An architect’s value is mostly in the things they stop you doing. If you have one instance, one module and a two-person team, you do not need one yet.

The question that usually settles it

When a client is not sure which one they need, we ask this: who decides how the work gets done?

If the answer is “we tell them what to build”, it is a developer.

If the answer is “we want them to work that out with the business”, it is a technical consultant.

If the answer is “we want them to tell us what we should be building at all”, it is an architect.

That one question resolves most of it in about thirty seconds on a call.

What goes wrong at each level

Hiring a developer when you need a consultant. The most common one. You get someone who builds exactly what the ticket says, and the tickets are wrong because nobody has run a proper workshop with the business. Six months in, the platform works and nobody uses it.

Hiring a consultant when you need a developer. Less damaging, more expensive. You have paid a premium for stakeholder skills you are not using, and the person gets bored. They usually leave inside a year, and their exit interview says “the work wasn’t what I expected”.

Hiring an architect too early. This one comes up when a business has bought a lot of ServiceNow and panicked. A strong architect with no team to direct and no scale to manage will spend three months writing standards documents and then start looking. Architects want complexity. If you cannot give them any, someone else will.

Not hiring an architect when you should have. Usually visible about two years in, when you have four integrations built four different ways, customisation nobody can explain, and an upgrade everyone is afraid of. That is the expensive version.

What this means for your job spec

A few things we ask clients to be specific about before we take a role to market:

Who writes the requirements? If the answer is the person you are hiring, say so in the advert. It changes who applies.

How many people will they work alongside? A consultant in a team of twelve is a different job from a consultant who is the entire ServiceNow function. Both are legitimate. Candidates need to know which one they are walking into.

Which modules, honestly. “ITSM plus a bit of HRSD” is a real answer and it is fine. “Full platform” when you mean ITSM makes experienced people suspicious.

What does the first six months look like? Greenfield implementation, BAU support, and rescuing a bad implementation attract very different people. The third one, done honestly, attracts more people than you would expect. Plenty of good consultants enjoy a mess.

Is the title negotiable? Sometimes the work is architect-level and the internal band is not. Say it early. We would rather find someone who wants the scope more than the title than lose a candidate at offer.

Where partners and customers differ

If you are a ServiceNow partner, your consultants need client-facing polish and the ability to move between three accounts in a week. Your developers need throughput.

If you are an end customer, the same titles mean something different. Your consultant will spend more time on internal politics than on discovery workshops, and your developer will be closer to the business than a partner-side developer ever gets.

Candidates know the difference. When a customer-side spec reads like a partner-side spec, it reads as though you have copied it from somewhere. Which, usually, you have.

Getting it right first time

Most of the mis-hires we get asked to fix were not skills failures. The person could do the job in the spec. The spec just described a job the business did not need.

Twenty minutes scoping the role properly before it goes live saves a lot more than twenty minutes.

If you are about to open a ServiceNow role and you are not certain which of the three it is, send me the spec and I will tell you what I think. No charge and no pitch attached.

ServiceNow Eyes $7 Billion Acquisition of Cybersecurity Startup Armis

Posted on by Dan Wells

ServiceNow is reportedly in advanced discussions to acquire Armis, a San Francisco-based cybersecurity startup, in a deal valued at approximately $7 billion. This acquisition would mark one of ServiceNow’s most significant strategic moves as the workflow automation giant continues to expand its security capabilities and strengthen its position in the enterprise technology market.

About Armis

Founded in 2015 by Yevgeny Dibrov and Nadir Izrael, both Israel Institute of Technology graduates with backgrounds in the Israel Defence Forces software units, Armis has become a leader in cybersecurity asset management. The company’s flagship Centrix platform provides real-time visibility, risk assessment, and protection across an organisation’s entire digital attack surface.

Armis has demonstrated impressive growth, reaching $300 million in annual recurring revenue earlier this year, up from $200 million the previous year. The company was last valued at $6.1 billion following a $435 million funding round in November 2024. Its client roster includes major organisations such as Colgate-Palmolive, United Airlines, and NHS South Wales.

Strategic Implications

This acquisition aligns with ServiceNow’s aggressive expansion strategy in the security space. Earlier this month, ServiceNow acquired identity security startup Veza for an undisclosed sum, and in March, the company purchased AI firm Moveworks. The potential Armis deal represents ServiceNow’s commitment to building a comprehensive security portfolio that complements its workflow automation platform.

For ServiceNow professionals and organisations invested in the ecosystem, this acquisition signals continued innovation and expansion of security capabilities within the platform. As the deal nears completion, industry watchers expect the integration of Armis’s technology to enhance ServiceNow’s security offerings and create new opportunities for implementation and specialisation.

The deal is expected to be announced within days, pending final negotiations.

Let's get started

Talk to one of our consultants todays, if you're looking to register a vacancy and apply for a role.

Upload your CV

Submit CV

This field is for validation purposes and should be left unchanged.
Name(Required)
Consent(Required)
Max. file size: 2 GB.