The useful answer is a cost model, not a headline figure
A narrow workflow using well-connected cloud tools can be simpler than a business-critical process spanning legacy systems, sensitive data and many exceptions. Any price given before the workflow is understood is an assumption.
A sound estimate should separate discovery, process design, implementation, software, testing, training and ongoing ownership. That makes the investment easier to challenge and the result easier to maintain.
The six cost components
1. Discovery and process mapping
This identifies the real workflow, people involved, data used, exceptions, volume and current performance. Skipping it can make the build cheaper initially and far more expensive later.
2. Process and SOP improvement
Automation should not preserve avoidable steps. Time may be needed to clarify responsibilities, simplify approvals and create a dependable procedure before anything is built.
3. Implementation and integration
Cost rises with the number of systems, quality of their integrations, data transformation, exception handling, user interfaces and security requirements.
4. Software and usage charges
Consider licences, premium connectors, transaction limits, storage, messaging and AI usage. A low-cost tool can still become expensive if every user needs a separate licence or volume grows sharply.
5. Testing, training and rollout
The system needs realistic examples, edge cases, user acceptance, documentation and a controlled launch. These are part of delivery, not optional extras.
6. Maintenance and ownership
APIs change, staff move, processes evolve and errors need monitoring. Decide who owns alerts, access, updates, documentation and supplier relationships.
What makes a project more expensive?
| Lower complexity | Higher complexity |
|---|---|
| One clearly bounded workflow | Multiple connected departments |
| Modern tools with reliable integrations | Legacy systems or manual exports |
| Structured, consistent inputs | Unstructured documents and variable data |
| Few, well-defined exceptions | Many judgement-based decisions |
| Low-risk internal administration | Sensitive data or customer-critical outcomes |
How to compare value
Start with the current cost of the problem: staff time, delays, lost enquiries, avoidable errors, rework and the cost of adding capacity as the business grows. Then compare that with the full first-year and ongoing cost of the proposed change.
Recovered capacity becomes valuable when it supports more sales, better service, increased delivery, improved quality or avoids future hiring. Use the Veliq ROI calculator with conservative assumptions.
Questions to ask any provider
- What discovery and process mapping are included?
- Which assumptions could change the scope?
- What software and usage fees will we pay directly?
- How are exceptions, failures and manual overrides handled?
- Who owns the accounts, data, documentation and finished workflows?
- What testing, training and post-launch support are included?
- How will success be measured after launch?
A strong proposal should be specific about the workflow and honest about uncertainty. If the provider starts with software features before understanding the work, slow the decision down.