Working With Us

For organizations that need to add to their technical capability, there are four traditional paths: hire a consulting firm, hire a development shop, hire a CTO, or kick the can down the road and do nothing. Each fails in its own specific way. We built IDS by taking each failure seriously and asking what the opposite would have to look like.

The answer is seven layers, from the first conversation about strategy to the system still running years later. Each is on offer separately in the marketplace. We believe they only work when one team holds them together.

In practice, very few mid-sized companies want to buy a “transformation”, and they shouldn't have to. Most start with one concrete need: a leadership team that wants to understand what AI means for the business, staff already using AI before anyone has set the rules, systems that don't talk to each other, or one workflow that eats a team's week. So you can start with any of the seven. Whichever one you start with, the same team stands behind all of them, and the next step is small enough to take right after the first win.

That means one team. Responsible for the entire pipeline. No gaps between what gets recommended and what gets built. No handoffs between firms that don't share context.
How Engagements Work

AI-Native Strategy

A roadmap that cannot be built is not a strategy. It is a document.

The reason consulting decks go unexecuted is rarely that the thinking was wrong. It is that nobody in the room had ever built the thing being recommended, so no one costed it properly, sequenced it realistically, or noticed the dependency that makes step four impossible.

We write strategy knowing we will be the ones implementing it. It is a hard discipline, and it is the point: it rules out the recommendation that sounds good and cannot be done.

In practice

We connect business objectives to a technology roadmap, identify where AI and data create real advantage rather than activity, and settle the practical questions early: which tools to license and for whom, and which workflow goes first.

What you get

A sequence you can act on, where each step produces something usable and you know why it comes in that order.

Leadership Training and Enablement

AI changes how teams are organized. The people who lead them have to see that first.

Training is the easiest place to start, and most of it stops too early. Teaching people to write better prompts is useful, but the return is limited, and adoption is rarely even. In one client, monthly usage ranged from 950 interactions at the top to 8 at the bottom: same firm, same licence, same tools.

The bigger prize is the leadership team. As AI agents take on a meaningful share of the work, the real question becomes how teams should be organized around them, and only leaders can answer it. Senior leaders need to see where AI changes the economics of the business. The people doing the daily work need to see what it does for their job. Train one group and not the other and you get a strategy nobody executes, or enthusiasm with nowhere to point it.

We teach both. And because we are also the team that builds, the training is grounded in the systems you will actually have.

In practice

Sessions for executive teams, boards and committees on where AI changes the business and which decisions need making now. Hands-on literacy for functional teams, built around their own workflows and the tools they already have. Change management as new systems go live, because a capability nobody adopts is indistinguishable from one that was never built.

What you get

Leaders who can make the organizational calls, and teams who get value from what you have already paid for.

AI Risk Audits, Governance and Acceptable Use Policy

Your people are already using AI. The question is whether anyone has decided how.

In most mid-sized companies, AI arrived before the policy did. Staff use it on personal accounts because the sanctioned tool is weaker than the free one they have at home. Client and company data goes into tools nobody has reviewed. Each department buys its own. Banning it doesn't work: it only moves the use somewhere you can't see. Ignoring it means the first governance conversation happens after the first incident.

We start with an audit of what is actually happening: which tools are in use, by whom, with what data, and what each vendor is allowed to do with it. Every tool then goes through the same review: what it can access, where the data goes, what the terms permit, and the conditions it can be approved under. Then we write the rules people will follow: an acceptable use policy permissive enough that your best early adopters keep going, and clear enough that everyone knows where the lines are.

Governance is not a binder on a shelf. It is a process that keeps running: a simple intake form when someone wants a new tool, a review that comes back in days rather than months, and a register of what has been approved and why. And because we are also the team that builds, the policy fits the systems you will actually run.

In practice

AI risk audits of current use, sanctioned and unsanctioned. Security and privacy reviews of individual tools, each with a clear rating and conditions of use. An acceptable use policy drafted for review by your legal function. A tool intake and approval process, a licence and seat strategy, and guardrails for which data can go into which tool.

What you get

An organization that can say yes to AI quickly, because it knows what it is saying yes to.

Workflow Planning and Architecture

Redesign the workflow before you automate it. Then put someone senior in charge of the technical decisions.

The fastest way to waste an AI budget is to make a process faster that should not exist at all. Before anything is built, we map how the work actually gets done: who touches it, where it waits, and which steps exist only because two systems never spoke to each other. Then we redesign it, and only then decide what software it needs. The workflow is the unit we plan, build and measure in.

Someone also has to own the technical decisions, and they cannot work for the vendor. A development shop builds what you specify. That sounds like service until you notice it makes you the architect: choosing platforms, approving designs, assessing security exposure, arbitrating between vendors who each have an interest in the answer. Those decisions have ten-year consequences, and they usually land on a COO or a CFO, because there is no CTO.

We take that seat. As embedded fractional CTO and fractional CISO, we own the architecture, select the tools, and carry the risk assessment, on your side of the table. And because every workflow is designed against one model of your business, each new piece connects to the last instead of becoming another stand-alone tool.

In practice

Workflow mapping and redesign, data architecture design, platform and tooling selection, security architecture, and technical management of whatever else you have running.

What you get

A plan for each workflow worth building, and a senior architect and security lead whose incentive is your system still working in five years.

Legacy Modernization

AI needs something solid to stand on.

Every mid-sized company has a backlog: the ERP upgrade that keeps slipping, the acquisitions that arrived with their own systems, the Access database and the Excel workbook that run more of the business than anyone likes to admit. None of it is glamorous, and all of it is in the way. Point an AI agent at six systems that disagree on what a customer is, and you get six confident answers.

Clearing that backlog is often the most valuable first project there is. It pays off on its own, because leadership can finally see one version of the numbers. And it lays the foundation that every later AI build depends on.

In practice

Integrating and normalizing data across ERPs, accounting packages, spreadsheets and homegrown databases. Cleaning and restructuring document stores so AI tools you already license answer from the right source. Dashboards that replace hand-reconciled reports, and modernizing the systems the next build will sit on.

What you get

One trustworthy view of the business, and a foundation you can build AI on.

New AI Builds

The people who designed it have to be the people who build it.

This is the conviction we would defend hardest, because almost the whole industry is organized against it.

Every handoff between design and construction loses context: the constraint that was discussed and never written down, the reason a field is nullable, the edge case someone raised in week two. You cannot document your way out of it. The only fix is that the gap does not exist.

It is also what makes a fractional CTO workable rather than ornamental. A senior technologist with no engineering team produces memos. We bring the team.

Every build also adds a layer to YourWorkMap, our model of how your business works, so the second project starts further ahead than the first.

How YourWorkMap works is on Our Technology

In practice

Document pipelines, proposal and report generation, monitoring and intelligence products, integrations and AI applications, built by our engineers under the same senior oversight that designed them.

What you get

Production systems built around your data and your judgment, and no codebase you are left holding alone.

Maintenance, Hosting and Deployment

Software needs a relationship, because systems drift.

Getting a system into production is its own discipline: security review, access controls, monitoring, and the handover to the people who will use it every day. We deploy inside your own cloud environment wherever we can, so sensitive data stays under your control. Where it makes more sense, we host and run it for you.

A project has an end date. A system does not. It drifts as the business changes, as a supplier changes a format, as a regulation moves, as the model underneath it is deprecated. Firms that sell projects are structurally finished at the moment you most need them. We think the engagement should be shaped like the problem, which is ongoing.

As the build completes, the retainer steps down to a lower-cost maintenance and fractional-CTO level. It is designed to shrink. That is what a partnership looks like when it is working.

In practice

Deployment into your cloud environment, hosting and monitoring, maintenance and extension of what we have built, continuous strategic and technical advice, and senior judgment available when conditions change rather than when a new contract is signed.

What you get

A technology partner that is still there, and still accountable, after the interesting part is over.

How Engagements Work

No long-term lock-in.

Engagements are structured so you build confidence before scaling investment.

No large upfront commitment.

You buy access to all seven layers and use what you need. Most clients start with one or two and add the others as the work calls for them.

The workflow is the atomic unit.

Handing everyone an AI licence can make people more productive with nothing to show on the bottom line. We help you pick one workflow at a time to redesign and automate, for measurable ROI.

How to start. Every engagement begins with a monthly retainer. At its smallest, that covers strategy and architecture (layers 01 and 04): senior strategic and technical advice, in effect a fractional CTO and CISO, for a fraction of what hiring them would cost. Where a project is small enough, we build it within the same retainer.

A long-term relationship. We want to be your technology partner for the long run. So we start small, on a retainer. Together we choose the workflows worth automating, one at a time, and scale the engagement in proportion to what they return. When the build work is done, we expect the retainer to step back down to the right size for ongoing strategy and CTO advice, training, and running and maintaining the systems we built.

See What This Looks Like in Practice

Specific examples of work performed: what the problem was, what we built, and what changed.