Short answer: A typical engagement runs audit, design, build, handover across three to six weeks. Costs range from low four figures for a single automated process to mid five figures for a full rebuild with data migration. Welsh SMEs may also be able to access business support through Business Wales and, in some cases, funding towards digital adoption — worth checking before you commission.
Phase 1 — Audit (week one). Mapping how the operation actually runs. Interviews with everyone who touches a core process, screen-shares of real work, a tool inventory and a data quality assessment. Output is a current-state map and a ranked list of findings. Your time commitment: four to six hours.
Phase 2 — Design (week one to two). Process architecture before tooling. What the entities are, how they relate, what states exist, who owns each transition, where the source of truth sits. Tool selection follows from this rather than preceding it. Output is a target-state design you sign off before anything is built.
Phase 3 — Build (weeks two to five). Data cleaning and migration, database structure, automated workflows, integrations, reporting. Data work usually takes longer than automation work, which surprises people.
Phase 4 — Handover (final week). Documentation, training, and the playbook. This phase is where engagements are won or lost. A working system without documentation is a dependency.
The honest variable: data quality. A quote given before anyone has looked at your actual records carries a large assumption.
Welsh SMEs have access to business support that's worth investigating before commissioning private work:
Programmes and eligibility change frequently, so check current availability directly rather than relying on any published summary. A consultant who mentions this unprompted is generally acting in your interest.
Define the measures before you start and take a baseline. Useful ones: hours spent on the target processes, error and rework rate, cycle time from enquiry to invoice, and how long it takes to answer a basic operational question. Retrospective estimates of "time saved" are unreliable in both directions.
Not technically. Local presence helps most during discovery, and there's an accountability argument for working with someone in your business community.
Typically five to fifty people. Below five, targeted point solutions usually beat a rebuild. Above fifty, you're into a different scale of project.
Yes, and it's often better. Audit and design first, then build in phases with each phase delivering something usable.
Decode & Grow works with founder-led businesses across Wales and the UK. Get in touch.