Things I am learning while building.
No content calendar theater. No generic motivation. I write when I have an observation, decision, framework, failure or experiment worth documenting.
Articles, by topic.
Every article answers one real question, links to the relevant expertise page, and is written from operating experience.
How much does an AI consultant cost in Dubai? Strategy vs implementation vs retainer
AI consulting in Dubai is priced three ways: a fixed diagnostic session, a scoped implementation project, or a monthly retainer. What each is for, what mine cost, and how to avoid paying for the wrong one.
AI consultant vs AI automation agency: which does your business need?
An AI consultant decides what should be built and why; an automation agency builds it. Most SMEs need the decision first.
AI agent vs chatbot vs workflow automation: what is the difference?
A chatbot responds. Workflow automation executes fixed steps.
The 90-day AI roadmap for an SME: what to do before buying more tools
Days 1–30: measure and decide. Days 31–60: automate one flow end to end.
What should a UAE business automate first? A practical decision framework
Automate the step that is repeated most, touched by the most people, and measurable — usually where data is copied by hand between a conversation and a system. A four-question framework from an operator.
10 business workflows UAE SMEs can automate before hiring more people
Before adding headcount, check whether the work is a process problem. Ten workflows — from lead intake to payment reminders — that growing UAE businesses can automate with the systems they already own.
How to choose an AI automation consultant in Dubai
Seven questions that separate an operator from a slide deck: what they have run, what they will say no to, how they handle approvals, what you leave with, and how they price. A checklist from someone who sits on both sides.
What should you measure before automating a business process?
Four numbers before any automation: how often the step runs, how long it takes end to end, how often it fails or gets redone, and how long the customer waits. Two weeks of measurement beats a year of arguing about ROI.
AI for CRM: lead qualification, follow-up and sales automation that a team actually uses
Where AI helps a CRM: qualifying inbound leads, drafting follow-ups, summarising threads and flagging quiet deals. Where it hurts: sending commitments without approval.
Start with the approval, not the feature
We nearly built the wrong automation because we started with what the software could do instead of who had to say yes.
WhatsApp + CRM + AI: a practical automation architecture for UAE businesses
The working architecture: official WhatsApp Business API, a shared inbox, the CRM or ERP as the system of record, AI for interpretation and drafting, and human approval for anything consequential. Built from running Go4WhatsUp.
Odoo + WhatsApp + AI: practical sales, support and operations use cases
Odoo Enterprise already sends and receives WhatsApp. The value is in what happens around the conversation: customer lookup, quotations, order status, invoice questions and AI-assisted replies with permissions.
WhatsApp is infrastructure, not a channel
What changes when you stop treating messaging as marketing and start treating it as the operating layer.
A real case study: problem → diagnosis → decision → implementation → outcome
How I write and run a case: the visible problem, the hidden constraint, the decision, what was actually built, and a factual outcome without invented percentages. Using a real WhatsApp and ERP engagement.
Revenue is not resilience
A growing top line hid a fragile business. What I now check before I believe a growth number.
What I write about.
WhatsApp / CRM / ERP
WhatsApp flows, Odoo, CRM integration, AI in customer operations.
WhatsApp automation consultingFounder / Execution
Growth, positioning, priorities, execution, operator lessons.
Business strategy consultingOne useful idea. One real story. One thing worth producing next.
What I am learning while building businesses, working with AI, solving execution problems, making decisions and occasionally getting things wrong.