Do not automate
The expected benefit does not justify the investment. We say so in writing and explain why.
Entry engagement
AI Workflow Opportunity Sprint
Find the right automation opportunity before investing in a full build.
A focused discovery engagement that examines one important business workflow: its bottlenecks, automation feasibility, required integrations, risks and the recommended implementation path.
Discuss Your WorkflowPricing shared after the workflow review
- Focus
- One important business workflow
- Typical duration
- 5-7 working days
- Format
- Remote discovery and analysis
- Pricing
- Shared after initial workflow review
What the Sprint examines, in order
- 01
Current process
- 02
Bottlenecks
- 03
Opportunity
- 04
Architecture
- 05
Roadmap
How the Sprint works
How the Sprint runs
- 01
Submit the workflow
You describe one process that costs your team time, and the systems it touches.
- 02
Initial fit conversation
A short call to confirm the workflow is a sensible candidate before anything is scoped.
- 03
Discovery and workflow mapping
We speak to the people involved and document the process as it actually runs.
- 04
Feasibility and architecture assessment
We test which steps can be automated dependably and design the boundary around them.
- 05
Findings and implementation roadmap
You receive the written findings, the recommendation and a sequenced path forward.
Deliverables
What you receive at the end
Eight deliverables are included in every Sprint. The ninth is optional, separately scoped and never assumed.
- 01
Stakeholder discovery
Interviews with the people who run and depend on the workflow.
- 02
Current workflow map
Every step, owner, system and handoff as the process actually runs.
- 03
Bottleneck analysis
Where work waits, is re-entered or is reworked, and what that costs.
- 04
Automation feasibility assessment
Which steps can be automated reliably, and which should not be.
- 05
Data and integration requirements
The records, permissions and systems any solution would depend on.
- 06
Recommended solution architecture
A proposed system boundary with components, data flow and fallbacks.
- 07
Risk and human-review requirements
Where a named person must stay accountable for a decision.
- 08
Implementation roadmap
A sequenced delivery plan with explicit exclusions.
- 09Optional · scoped separately
Optional working prototype
Optional and separately scoped. Not included by default.
If the conclusion is that automation is not worth building for this workflow, that is a valid outcome, and we will say so in writing.
After the Sprint
What happens after the Sprint?
Three outcomes are possible, and two of them do not involve us building anything.
Improve the existing process
Workflow or ownership changes are recommended before any software is built.
Proceed to implementation
The opportunity holds up, and we move into a separately scoped build engagement.
Questions
Before you request a Sprint
Is the Sprint paid?
Yes. It is a paid discovery engagement. Pricing depends on the size and complexity of the workflow and is confirmed in writing after the initial workflow review, before any work begins.
How long does it take?
Typically 5-7 working days once the initial fit conversation is complete and the relevant people are available. Larger or multi-system workflows may need longer, which we confirm before starting.
What information do you need?
A description of the workflow in your own words, access to the people who run it, examples of the documents or records involved, and a list of the systems it touches. Access is scoped and agreed in writing.
Is the prototype included?
No. A working prototype is optional and separately scoped. If a prototype would materially reduce uncertainty, we say so in the findings and quote it separately.
Are we required to hire Vision Bridge for implementation?
No. The findings, workflow map and roadmap are yours. There is no obligation to proceed to implementation unless a separate engagement is officially confirmed.
What happens if automation is not recommended?
You still receive the workflow map, bottleneck analysis and reasoning, plus any process changes we would recommend instead. A negative recommendation is a valid and useful outcome.
Next step
Bring one workflow
We will look at it properly before anyone commits to a build.