Resources
What we have learned, written down.
What we have learned putting AI into business workflows: why projects stall, what adoption actually costs, and the numbers behind both.
These guides cover workflow design, adoption and measurement, with industry-specific examples and the research behind the numbers we quote. They explain what to check before building and how to tell whether a change helped.
They are written to be useful whether or not you ever hire anyone: each one gives you a practical way to examine the problem with your team.
Why projects stall
-
Measure lead follow-up before you automate it
Most sales teams have the same quiet leak: an enquiry arrives, someone means to call, two days pass. Three measurements that cost nothing find it.
-
Why your team stopped using the AI tool you bought
Adoption is rarely a training problem. People stop using a tool when using it costs more attention than not using it — and that is fixable.
By industry
-
AI legal document review with source-linked lawyer verification
A source-linked legal document review workflow covering scope, question lists, extraction, chronologies, lawyer verification and quality measures.
-
AI client intake and conflict-check preparation for law firms
A controlled law-firm intake workflow covering enquiry capture, party data, conflict-search preparation, lawyer review, matter opening and quality measures.
-
Law-firm AI ROI: capacity, write-offs and matter economics
Calculate law-firm AI ROI using realised capacity, write-offs, fixed-fee margin, full ownership costs and professional-quality release conditions.
-
Accounting-firm AI ROI: capacity, margin and client turnaround
Calculate accounting-firm AI ROI using engagement capacity, turnaround, rework, contribution margin, full operating costs and professional quality controls.
-
AI client document collection for accounting and tax firms
A controlled workflow for accounting client documents: request lists, secure intake, classification, missing-evidence follow-up and accountant review.
-
AI bookkeeping exception triage and month-end review
A controlled AI workflow for bookkeeping exceptions and month-end review: source evidence, reconciliations, review queues, integrations and quality measures.
-
AI customer service and complaint triage for financial services
A controlled AI workflow for financial customer service: request classification, complaint and vulnerability signals, human review and quality measures.
-
AI loan document processing and underwriting support
A controlled workflow for loan documents: classify and extract evidence, detect exceptions, prepare cases, integrate systems and preserve human decisions.
-
Financial-services AI ROI: capacity, quality and return
Calculate financial-services AI ROI using capacity, backlog, rework, realised financial benefit, full costs and customer and control guardrails.
-
AI maintenance triage for property managers
A practical AI maintenance workflow for property managers: safer intake, human escalation, work-order routing, contractor approval and measurable outcomes.
-
AI AML alert triage and investigation support
A practical AI workflow for AML alert triage: evidence gathering, investigator review, case integration, controls and measures for investigation quality.
-
AI customer onboarding and KYC document processing
A controlled AI workflow for financial customer onboarding: document intake, evidence checks, human review, system integration and quality measurement.
-
AI lead response and follow-up for estate agents
How estate agencies can automate lead capture, reviewed replies, CRM routing and follow-up while measuring response quality, appointments and opt-outs.
-
AI for food waste reduction: measure causes before predicting them
Use structured food-waste records and AI-assisted analysis to find preventable patterns, test changes and measure value without hiding the causes.
-
AI menu engineering: measure item profitability before changing prices
Combine restaurant sales, recipes and channel costs to review menu-item contribution, test decisions and avoid misleading margin recommendations.
-
AI reporting for multi-location restaurants: trusted variance briefs
Standardise restaurant data, explain location variances and create source-linked operating briefs without hiding missing or inconsistent records.
-
AI restaurant inventory and purchasing: a controlled workflow
Connect counts, recipes, invoices and demand to prepare restaurant purchase suggestions with manager approval and measurable stock outcomes.
-
AI labour planning for restaurants: forecast, schedule and review
Turn restaurant demand forecasts into reviewed role coverage while respecting availability, skills, local rules and manager approval.
-
AI demand forecasting for restaurants: from sales data to action
Build a reviewable restaurant demand forecast for prep, purchasing and labour, with clean inputs, manager overrides and useful accuracy measures.
-
Restaurant AI ROI: build a business case from operating evidence
Calculate restaurant AI ROI from a measured baseline, realised margin or cost effects, implementation expense and quality guardrails.
-
AI candidate submission packs for recruitment agencies
How to turn CVs and recruiter notes into accurate, branded candidate submission packs with AI, ATS integration, human approval and measurable time savings.
-
AI interview scheduling for recruitment agencies
A practical guide to AI interview scheduling: ATS and calendar integration, recruiter control, time zones, rescheduling and measuring coordination effort.
-
Recruitment automation ROI: which workflow should you automate first?
Calculate recruitment automation ROI using recruiter time, full costs and contribution margin. Compare workflows and plan a measurable pilot.
-
AI employee offboarding automation: tasks, access and handover
Coordinate employee offboarding across HR, IT and managers with approved triggers, verified access changes, handover records and exception handling.
-
AI for HR data quality: employee records and HRIS integration
Find missing or conflicting HR records, reconcile system updates and propose reviewed corrections without silently changing employee facts.
-
AI employee onboarding automation for HR teams
How to automate employee onboarding with HRIS integrations, approved checklists, human review and measurable readiness for the first day.
-
AI HR document automation: drafts, approvals and delivery
Prepare HR documents from approved templates and verified employee data, with version control, human approval and reliable delivery records.
-
AI HR policy assistants: accurate answers from approved documents
Build an HR policy assistant around approved sources, employee permissions, effective dates, citations and a clear route to human help.
-
AI HR service desks: employee requests, routing and resolution
Use AI to classify employee requests, prepare summaries and route HR cases while preserving confidentiality, ownership and measurable resolution.
-
HR automation ROI: choose a workflow and measure the return
Measure HR automation ROI with workflow baselines, full costs, adoption and service quality. Includes a worked example and pilot decision criteria.
-
AI candidate communication and follow-up for recruitment agencies
A practical guide to AI candidate follow-up: applicant replies, interview scheduling, WhatsApp and email coordination, ATS records and measurable results.
-
Candidate rediscovery: make your existing database useful
A practical guide to candidate rediscovery: search existing ATS records, check relevance, refresh availability and measure interviews and placements.
-
AI CV matching and shortlisting for recruitment agencies
How recruitment agencies can use AI CV matching with recruiter review, ATS integration and practical measures of shortlist quality and time saved.