Scope note: This page covers automation build — the implementation layer applied after the process is mapped and simplified. For the assessment that identifies automation opportunities, see C16 AI Tools Assessment. For process redesign before automation, see C17 Process Redesign.
"The most common automation failure is not a technical one — it is a sequencing one. The automation is built before the process is understood, so it encodes the existing inefficiency into a system that now runs automatically. The result is fragile: it breaks when any connected tool changes, it produces inconsistent outputs, and it gets abandoned within a few months. The fix is not a better automation. It is doing the process work first."
Freelance SEO & AI Consultant — London
The right order matters
Three stages — each a prerequisite for the next. Skipping a stage does not save time; it creates rework.
Map
Document how the workflow actually runs — not the textbook version. Identify where tasks get stuck, where handoffs fail, and where manual work is creating waste.
C17 Process RedesignSimplify
Remove unnecessary steps, consolidate duplicated work, and ensure the process produces consistent inputs before automation is applied. Automation locks in whatever the process currently does.
Part of the process redesign engagementAutomate
Automate only the clear, repetitive, high-value parts of a clean process — starting with the highest-ROI workflow and building up incrementally rather than trying to automate everything at once.
This is C18 — the build engagementWhat the build engagement involves
Four stages from scope confirmation to handover — each producing a specific output.
Scope confirmation
Before any build begins, the workflow is confirmed as automation-ready. This means the process is documented, simplified, and producing consistent inputs. If it is not, we address that first — either through C16 or C17 — rather than building automation against a moving target.
Platform selection
Zapier, Make, or n8n is selected based on the workflow requirements: volume, complexity, the tools involved, and whether the automation needs to run on your infrastructure or ours. Platform selection is not a preference — it is a technical decision made against the specific use case.
Build and test
The automation is built in stages: trigger confirmed, actions mapped, error handling added, edge cases tested. Every automation is tested against real data before handover — not just against the expected happy path.
Handover and documentation
You receive a working automation, a plain-language explanation of how it works, and documentation covering what to do if it breaks. The goal is an automation you can maintain without needing to call us every time a connected tool updates its API.
Zapier, Make, or n8n — how the choice is made
Platform selection is a technical decision based on the workflow, not a preference. Here is the honest comparison.
Simple, high-volume trigger-action workflows between popular SaaS tools
Higher cost at volume; less flexible for complex branching logic; limited data transformation
CRM updates, lead routing, form-to-spreadsheet, email notifications
Complex multi-step workflows with branching, data transformation, and error routing
Steeper learning curve; visual complexity grows quickly with workflow complexity
Multi-system data sync, conditional routing, API-to-database pipelines
Self-hosted or cloud workflows requiring full control, custom code nodes, or sensitive data
Requires more technical setup; self-hosted version needs infrastructure management
Internal tools, data pipelines with custom logic, privacy-sensitive workflows
What London SMEs automate first
Four workflow areas that consistently produce the highest ROI when automated — because they are repetitive, high-volume, and currently consuming disproportionate manual time.
Reporting
Weekly or monthly reports assembled automatically from multiple data sources — no manual spreadsheet work, consistent format every time.
CRM and lead management
New leads routed to the right owner, CRM records updated from form submissions, follow-up sequences triggered without manual intervention.
Internal approvals
Approval workflows that route requests to the right person, track status, and send reminders — without email chains or missed steps.
Data synchronisation
Keeping records consistent across two or more tools — removing the manual re-entry that creates version conflicts and data quality problems.
Fixed price per engagement, scoped before work begins. Price varies by workflow complexity and number of integrations. No hourly billing, no scope creep.
Not sure what to automate or whether your workflow is ready? The £999 assessment identifies the highest-value automation opportunities.
If the process needs mapping and simplifying before automation is applied, this is the prerequisite step.
For workflows that involve information retrieval, decision support, or knowledge management — the next layer after automation is in place.
Frequently asked questions
Do I need to have completed the AI Tools Assessment or Process Redesign first?
Not always — but the automation needs to be scoped against a process that is already working correctly. If you have a specific, well-defined workflow that you want automated and you are confident it is producing consistent inputs, we can scope the build directly. If you are unsure whether the process is ready, the AI Tools Assessment (C16) is the right starting point. Automating an unclear or inconsistent process produces fragile automations that break quickly.
How do you choose between Zapier, Make, and n8n?
Platform selection is based on three factors: the complexity of the workflow (simple trigger-action vs multi-step branching), the tools involved (whether they have native integrations or require API work), and your data requirements (whether the automation handles sensitive data that should not leave your infrastructure). Zapier suits simple, high-volume workflows between popular SaaS tools. Make suits complex multi-step workflows with branching logic. n8n suits workflows requiring custom code, self-hosting, or full data control.
What does the build engagement include?
The engagement includes: scope confirmation, platform selection, build and testing against real data, error handling, handover documentation, and a 30-day support window for issues arising from the delivered automation. It does not include ongoing maintenance — that is a separate retainer arrangement if needed.
How long does a build take?
Simple single-workflow builds typically take one to two weeks from scope confirmation to handover. Complex multi-workflow builds take three to six weeks. Timeline is confirmed at the scoping stage before work begins.
What happens if the automation breaks after handover?
The 30-day support window covers issues arising directly from the delivered build. Most post-handover breaks are caused by a connected tool changing its API or authentication method — these are covered during the support window. After 30 days, fixes are scoped as a separate engagement unless a maintenance retainer is in place.
Ready to automate the right things
Book a 30-minute discovery call. We will review your workflow, confirm whether it is automation-ready, and scope the build before any commitment is made.