Open playbooks · Decision tools · Hands-on support
Find worthwhile opportunities, compare your options, and move from experimentation to practical adoption — with open playbooks and hands-on support.
Just want to learn? Read the open playbooks — 21 chapters, free, in 10 languages.
Start from your situation
Each situation calls for a different decision. Pick the closest one for the guide, tool and engagement that fit it.
Choose a starting point
“We've been told to adopt AI. Where should we start?”
What helps: A ranked shortlist with costs, dependencies and a credible first project.
Activate existing tools
“We bought AI tools, but useful adoption is patchy.”
What helps: Real workflows, working practice, controls and measured adoption.
Redesign a workflow
“This recurring process eats too much time.”
What helps: A redesigned workflow with measurable acceptance criteria.
Get a workflow approved
“We need approval before using AI with our information.”
What helps: A specific data-flow, risk and control assessment.
Fix a stalled pilot
“Our pilot works in a demo but isn't ready for staff.”
What helps: What's blocking it, and a practical remediation plan.
Compare products
“Which product should we buy?”
What helps: A requirements-led comparison tested against your own work.
How it works
Every engagement ends in a decision you can defend. Sometimes the right answer is to use software you already own and not commission custom AI — and we will say so.
Use the scorecard or the brief builder. You get a result straight away — no call required.
A fixed-scope sprint or review that ends in a decision: go, fix first, or stop.
Activation or implementation, accepted against criteria agreed before work starts.
Baseline versus after: effort, quality, adoption and running cost.
Work with us
Fixed-scope entry engagements with a concrete output and an agreed acceptance condition. Implementation and ongoing reviews follow only if they are justified.
Entry engagement
from £2,500
Find the workflows worth changing, price them honestly, and pick a credible first project.
Done when: The sponsor can make a go/no-go decision on the first project and has appointed an owner.
See scope & outputs →Entry engagement
from £6,000
Turn licences you already pay for into a few repeatable workflows your team actually uses.
Done when: Staff can run the agreed workflows with the required review step, and the follow-up measurement has been shared with the sponsor.
See scope & outputs →Entry engagement
from £4,000
Find out what is actually blocking your AI pilot from reaching staff — and the smallest fix that would unblock it.
Done when: The buyer understands what blocks deployment, the next investment required, and the evidence that would justify it.
See scope & outputs →An Opportunity Sprint ends with a brief like this for the recommended first project — written so you can take it to any supplier, including us.
Illustrative excerpt — not a client deliverable.
# Brief: Monthly client performance packs
Owner: Head of Client Operations
Volume: 40 packs/month · ~6 h each
Recommendation: AI-assisted drafting with analyst review; fix the data export first
Not recommended: autonomous agent; custom model
## Acceptance
- ≥ 95% of figures traceable to source
- Analyst review ≤ 90 min per pack
- No client data leaves the approved tenant
## Value (base case)
- 150 h/month released → reallocated to client reviews
- Cash saving: £0 unless hiring is avoided
## Open questions
- Who signs off the data-flow review?
Free decision tools
Each tool gives you the full result on screen, with no sign-up. Copy it, download it, or ask for an expert review.
~5 min
Score one workflow on value, feasibility, risk and readiness, and get a recommended approach — which can be “no AI needed”.
Open the tool →~4 min
Separate hours released from cash removed, add running costs, and see payback across low, base and high cases.
Open the tool →~10 min
Turn your workflow into a portable brief covering scope, systems, constraints, success criteria and purchasing needs.
Open the tool →Open source · free · 10 languages
The reference behind every recommendation. Start with the GenAI foundation, then master agentic AI. Apache 2.0, downloadable as PDF.
Playbook 1
11The foundation for business leaders
Playbook 2 · New
10From GenAI to autonomous agents
Evidence & methods
We separate vendor claims from test results, hours released from cash saved, and one engagement from an industry benchmark. Case studies appear only with the client's permission.
How we measure outcomes →Packt
7 AI workflows to save hours at work
Dipankar Sarkar
Also by the author
7 AI Workflows to Save Hours at Work Every Week · Packt Publishing · June 2026
Build smarter daily workflows with ChatGPT, Claude, and Perplexity. Seven practical workflows for automating emails, meetings, research, reports, task tracking, spreadsheets, and professional communication. Includes ready-to-use prompts and templates.
Practical insights on Generative AI, agentic systems, and AI-driven business transformation.
Aug 25, 2026
Why asking an LLM to 'return JSON' isn't enough, what schema-constrained generation actually fixes, and the validation layer you still need even when it works.
Read →Aug 18, 2026
Who should own GenAI adoption, what a Center of Excellence actually does day to day, and the operating model that keeps pilots from stalling at 'IT project.'
Read →Aug 11, 2026
Prompt engineering optimizes what you ask. Context engineering optimizes everything the model sees before it answers — and it's the harder, more durable skill.
Read →Jun 29, 2026
A practical, step-by-step guide to implementing Generative AI in your organization in 2026, from use-case selection to production, with ROI metrics.
Read →Jun 28, 2026
Agentic AI explained without hype: what it is, how it differs from GenAI, the autonomy spectrum, and why 2026 is the year agents go from demo to production.
Read →Jun 27, 2026
The Model Context Protocol is standardizing how AI agents call external tools. Here's what it is, how it works, and why every AI team should know about it in 2026.
Read →Generative AI is a class of artificial intelligence systems that create new, original content — text, images, code, or data — based on patterns learned from training data. Unlike traditional AI that analyzes and predicts, Generative AI produces novel output that closely mimics human-created content.
Agentic AI is an AI system built around an autonomous agent loop: the model receives a goal, reasons about the next step, takes an action (calling a tool, searching, writing code), observes the result, and repeats until the goal is met. Unlike a single prompt–response exchange, an agent runs over many cycles, maintains state, and can recover from failures.
The Model Context Protocol (MCP) is an open standard for exposing tools, resources, and prompts to AI applications. Open-sourced by Anthropic in 2024, it lets any AI model discover and call external tools uniformly — so you write the integration once and use it across all agents and models.
Successful GenAI implementation starts with identifying high-impact use cases, structuring clean data, building internal proofs-of-concept, measuring ROI, and ensuring security and regulatory compliance. The GenAI Playbook walks through each step with practical examples across five phases.
Secure AI agents by treating tool output as untrusted (defense against prompt injection), scoping tools per task with allowlists, requiring human approval for destructive actions, validating tool arguments, rate-limiting, isolating credentials, and logging every tool call for audit.
Generative AI can hallucinate, struggle with deterministic tasks, and raise data-privacy and compliance risks. Some workloads — precise calculations, regulated decisions, or tasks requiring guaranteed accuracy — are better served by traditional automation. Agents add scaffolding (planning, tool-use, self-check) that contains hallucination but doesn't eliminate it.
Disclosure:Paid engagements are contracted and delivered by Neul Labs, which Dipankar Sarkar founded and leads. Playbook recommendations stay vendor-agnostic, and we do not publish supplier rankings.