Before You Hire an AI Consultant, Answer These Five Questions

Hiring an AI consultant without a clear brief is not an investment. It is an expensive way to discover what you should have figured out before the first meeting. Across engagements in Vietnam, Singapore, and the UAE, I keep seeing the same pattern: smart founders, real budgets, and no clarity on what problem they are actually trying to solve. Here is the diagnostic I walk every SME through before they sign anything.

---

Question 1: What Is the Specific Business Problem — Not the Technology Wish?

Most founders arrive saying some version of "we want to use AI." That is not a problem statement. That is a procurement impulse.

A useful problem statement sounds like: "We lose roughly 30% of inbound leads because our sales team cannot follow up within the first hour." That is a problem. AI might be part of the solution. It might not.

Write your problem in one sentence before you brief anyone. If you cannot, you do not yet have a problem — you have anxiety about falling behind, which is understandable but not billable.

---

Question 2: What Data Do You Actually Own, and What Shape Is It In?

AI systems run on data. Not on ambition, not on budget, not on the consultant's demo deck.

A founder in Ho Chi Minh City once came ready to build a churn prediction model. When we opened their CRM, 18 months of customer interactions were split across three spreadsheets, maintained by different salespeople, with no consistent customer ID linking them. The first six weeks became data cleaning, not AI building — and the founder had not budgeted for that.

Before you engage anyone, answer these honestly: Where does your operational data live? How far back does it go? Can a competent analyst answer a basic business question from it within a day? If most answers are "we are not sure," your first investment is data infrastructure, not an AI consultant.

---

Question 3: Which Specific Process Are You Trying to Change?

AI is not a department. It is not a strategy. It is a capability that attaches to a process.

The best implementations I have seen across Southeast Asia and the Gulf started with a process someone already owned and understood deeply. A logistics company in Dubai automated freight document processing because one operations manager had mapped every step by hand before the consultant arrived — she knew where errors happened, how often, and what they cost. The project delivered in three months.

Briefs like "AI for operations" or "AI for customer experience" describe departments, not processes. Find the specific, repeatable task inside the department. That is your entry point.

---

Question 4: Do You Have an Internal Owner, or Are You Outsourcing the Thinking Entirely?

A consultant can build you a system. They cannot want it to succeed more than you do, and they cannot maintain it after they leave.

In Singapore, a mid-sized retail group hired an excellent AI firm to build a demand forecasting tool. The tool worked. Six months after go-live, the buying team had quietly stopped using it — no one internally had been assigned to champion it or integrate it into their workflow. Accurate and ignored.

Identify your internal owner before you hire. Give them real authority and real accountability. If you cannot name that person today, you are not ready to bring in external help.

---

Question 5: What Does Success Look Like in 90 Days — Not Three Years?

Contracts get signed on three-year transformation narratives, but relationships live or die in the first 90 days.

Define a near-term success metric before you engage. Not "we will be an AI-driven company" — something measurable. Response time to inbound leads drops from four hours to 20 minutes. Document processing errors fall by half. One metric, observable and achievable inside a quarter. This single discipline protects you from scope creep and the slow, expensive drift that kills most engagements before they deliver anything real.

---

What to Do Before You Brief Anyone

1. Write a one-sentence problem statement. Share it with two people who will tell you the truth. If they both immediately understand it, you are ready.

2. Run a 48-hour data audit. Ask your operations lead to pull every data source touching the problem — where each lives, how far back it goes, who controls it.

3. Name your internal owner now — before the proposal stage. Ask them what they would need to run this initiative once external support phases out.

The founders who get real results from AI engagements are not the ones with the biggest budgets. They are the ones who did the thinking first.

---

Frequently Asked Questions

Q: How do I know if I need an AI consultant or a data engineer first? A: If you cannot answer basic business questions from your existing data within a day, you need data infrastructure before AI. A consultant will spend your budget cleaning data instead of building solutions.

Q: What is a realistic 90-day AI milestone for an SME? A: One measurable improvement on one process — for example, cutting document processing errors by 50% or reducing lead response time from four hours to under 30 minutes.

Q: What happens if I hire an AI consultant without an internal owner? A: The system gets built and then quietly abandoned. Without an internal champion to interpret outputs and integrate them into daily workflows, even accurate tools get ignored within months of go-live.