Everything you need
to know before starting.

We price by engagement, not by the hour. Before any work begins, we scope exactly what will be delivered and agree a fixed investment for that defined scope. You know the number before you commit. There are no retainer requirements and no open-ended billing.

Our rates are structured to be substantially below Tier-1 consulting firms, typically around 40% below, while maintaining the same standard of senior delivery. That difference comes from our structure: you get co-founders, not account managers and junior analysts.

Scope changes are handled through a straightforward change process. If new information surfaces that materially changes what is needed, we discuss it openly, agree the revised scope, and issue an updated Statement of Work before proceeding. There are no hidden charges for natural clarification or refinement within the original scope, the boundary between scope refinement and scope change is something we establish clearly at the outset.

This depends on the scale and complexity of your situation, which is exactly what the Discovery Call is designed to establish. As a rough guide: if you have a specific, bounded problem to diagnose, start with the Starter or Professional package. If you need tools or automation deployed and operational, Solutions Engineering is more appropriate. If your operation needs rebuilding end to end, including intelligent automation across multiple functions, Transformation is designed for that.

We will always recommend the package that fits your situation, not the most expensive one available. If a Starter engagement is the right answer, that is what we will propose.

Basic automation executes a fixed sequence of steps when triggered, it follows a script and cannot adapt. Intelligent automation uses AI to handle variability: reading unstructured inputs, making contextual decisions, drafting outputs that require judgement, and routing exceptions that fall outside a standard rule.

In practice, this means AI solutions can handle tasks like triaging inbound enquiries, extracting structured data from documents, drafting routine communications, and escalating edge cases to the right person, all without a human in the loop for every step. It is the difference between automating a fixed process and automating a function.

Yes, and this is one of the most important things we tell every client who comes to us wanting to automate. Automating a broken process does not fix it; it makes the errors faster and harder to trace. The sequence that actually works is: understand the process, fix what is wrong with it, stabilise it, then automate the parts that should not require human attention.

This is why our Business Analysis & Optimization practice sits alongside Intelligent Automation, not separate from it. In practice, most of our automation engagements begin with a process review, not because we insist on selling additional services, but because skipping that step produces automation that breaks quickly and is expensive to maintain.

The strongest candidates are tasks that are repetitive, rule-governed, and time-consuming, but not necessarily simple. Good examples include: processing and classifying inbound documents or emails, generating first-draft communications or reports from structured data, extracting and populating information across systems, routing tasks and approvals based on defined criteria, monitoring for conditions and triggering alerts or actions, and compiling status updates from multiple data sources.

As a general test: if a competent new employee could be trained to do it in two days by following a clear set of rules, an AI solution can almost certainly do it. If the task genuinely requires deep contextual judgement or creative problem-solving, it is probably better kept with a person, at least for now.

We are tool-agnostic and select based on what is right for the specific situation, your existing technology environment, your team's technical capability, and the nature of the task being automated. We work with leading AI model providers and with workflow automation platforms that integrate with the tools you are already using.

We do not have commercial partnerships with specific vendors that would bias our recommendations. If a simpler, cheaper solution solves the problem adequately, we will recommend that over a more sophisticated one. Our job is to deploy what works, not to implement technology for its own sake.

Deployment timelines depend significantly on the complexity of the task, the condition of the underlying process, and how well it integrates with your existing systems. A focused, well-defined automation for a stable process can be designed and deployed in 2, 4 weeks. A more complex, multi-step agent that spans several systems and requires process redesign as a prerequisite will take longer, typically within an 8, 12 week Solutions Engineering engagement.

We scope the timeline as part of the proposal, so you always know what to expect before work begins. We do not commit to timelines we are not confident in.

Every automation we deploy includes full documentation: how it works, what triggers it, what it produces, and how to modify it. We also provide a maintenance guide so your team, or any technically capable person you work with in the future, can understand and update it without depending on us.

The 30-day post-delivery support period is specifically designed to catch first-run issues. Beyond that, we offer optional retainer support for clients who want ongoing assistance. But our goal is to leave you with automation you own and understand, not automation that locks you into a dependency on us to keep it running.

Data handling is addressed explicitly as part of every automation engagement. We design with data minimisation in mind: AI solutions only access the data they need to perform the specific task they are built for. We discuss data residency, access controls, and any applicable regulatory requirements (including UAE data protection frameworks) before any automation touches your systems.

We do not use your business data to train any models, and we implement access controls so that automation credentials are scoped to the minimum required permissions. If you have specific compliance requirements or data governance policies, we work within them, not around them.