Most sales teams don’t lack effort. They lack a shared process. Salesforce’s State of Sales report, 2026 found sales professionals spend only 40% of their time selling. Without a process, a salesperson asks loose questions, the buyer struggles to see the value, and sales management can’t tell which deals are real. MEDDPICC fixes that with a structured checklist and clearer questions. AI makes each step faster, from meeting prep to the proposal. Below are four stages, with a worked example on one Dubai company using only public sources: research prospects that match your ideal client profile (ICP), map the decision-making unit, prepare MEDDPICC questions for each stakeholder, and build a proposal with total cost of ownership, cost of inaction, return on investment (ROI), payback and internal rate of return (IRR).
Table of Contents
Why a structured process matters
MEDDPICC is a deal qualification checklist. Each letter is something your team needs to know before a deal can be forecast with confidence:
- Metrics: the measurable result the buyer wants.
- Economic buyer: the person who can release the budget.
- Decision criteria: how options will be judged.
- Decision process: the steps and people between interest and a decision.
- Paper process: contracts, legal review and procurement.
- Identify pain: the problem that makes acting now worth it.
- Champion: the insider who sells for you when you are not in the room.
- Competition: the alternatives, including doing nothing.
For sales management, the payoff is transparency: every answer is logged against its letter as confirmed, assumed or missing, so the pipeline shows what is known and what is still a guess.
Two excuses come up often. The first is “I don’t have time.” AI does the first draft of the research, so your time goes on the right conversations. The second is “AI can do all of this for me.” It can’t. AI drafts the research, the questions and the numbers; the salesperson still asks, listens and checks every figure. That is the method I teach in my AI sales training with MEDDPICC.
Our demo company is Salik, Dubai’s toll gate operator, listed on the Dubai Financial Market. I have no relationship with Salik, and this example is for illustration only.
Step 1: Use AI to research prospects that match your ICP
Describe your best clients without naming them, and ask the AI to build an ICP: industry, size, emirate or free zone, buyer roles, trigger events and disqualifiers. Then paste in a prospect’s latest results release and ask it to score each criterion as fit, gap or unknown, quoting the supporting line and inventing no figures. Claude can search the web and work with files you upload, and Grok Bot can research accounts and score contacts with intent. Apollo lists buying signals such as intent, job changes and website visits, and its AI research helps qualify prospects. Lusha tracks job changes, funding rounds, hiring surges and new leaders.
Say your ICP is companies building new digital payment or partnership lines. Salik’s H1 2026 results release shows triggers: ancillary revenue reached AED 17.2 million, up 98%, and the concession fee rose from 22.5% to 23.1%. What the filing cannot tell you, such as current vendors, goes on the unknown list for Step 3.
Step 2: Map the decision-making unit before the first meeting
Next, find out who you need to speak to. Paste in the governance report and executive page, and ask the AI to suggest a likely MEDDPICC role for each leader, with a one-line reason, marked as a hypothesis. Apollo and Lusha both hold B2B contact and company records, and Lusha’s Buying Group ranks contacts as decision maker, champion or end user.
Salik’s 2025 annual report lists five executive roles: Chief Executive Officer, Chief Financial Officer, Chief Technology Officer, Chief Legal Counsel, and Director, Support Services. Our hypotheses, not Salik’s: the CFO as economic buyer, the CEO as owner of the metrics, the CTO as technical evaluator, and the Chief Legal Counsel as owner of the paper process. Family-owned and government-linked firms often look different, as I cover in selling to family businesses in the Gulf.
Step 3: Meeting prep: MEDDPICC questions for each stakeholder
Ask the AI for two MEDDPICC questions for each stakeholder. Each should quote one figure from the filing, ask for something only the buyer knows, and carry its MEDDPICC letter. One example for each of four stakeholders:
CFO (Economic buyer, Decision criteria): “The concession fee went up from 22.5% to 23.1%, and the interim dividend pays out about 100% of first-half profit. What payback does a new spend need to show, and who sets that bar?” (source 1, source 2)
CEO (Metrics): “Ancillary revenue almost doubled to AED 17.2 million, about 1.2% of first-half revenue. What does it need to reach before the board sees it as a second engine?” (1.2% is our calculation: 17.2 divided by 1,412.0; source 1.)
CTO (Decision criteria, Competition): “You already run parking payment partnerships with Dubai Airports, Valtrans and Parkonic, and electric vehicle (EV) charging with Schneider Electric and Vcharge. Where would a new system have to integrate, and what would rule a supplier out?” (source 1)
Chief Legal Counsel (Paper process): “Dubai Airports was a 10-year agreement; Shamal and the Dubai Integrated Economic Zones Authority (DIEZ) began with memoranda of understanding (MoUs). Which paper would a new supplier sign first, and who reviews it?” (source 1)
Step 4: Use AI to build the business case
The last stage is the proposal. Claude can draft account briefs and proposal decks in your template. Tag each input “from buyer”, “from filing” or “assumption”, and ask it to calculate total cost of ownership (TCO), ROI, payback, net present value and IRR, showing every formula. Cost of inaction is the net benefit forgone for each month the decision waits.
An illustrative model. Every input is an assumption, not a Salik or PKM figure, and every result is our calculation:
- Inputs: one-time cost AED 1,000,000; running cost AED 200,000 a year; benefit AED 700,000 in year 1 and AED 900,000 in years 2 and 3; discount rate 10%.
- Three-year TCO: AED 1,600,000. Total benefit: AED 2,500,000. ROI: 56.25%.
- Payback: about 1.71 years, or about 21 months.
- Net present value at 10%: about AED 559,000. IRR: about 37.7%.
- Cost of inaction: about AED 41,700 a month (year 1 net benefit of AED 500,000 divided by 12).
Then ask for conservative, base and upside projections. Benefits 20% lower give an IRR of about 17.7%, and 20% higher about 56.3% (our calculation). A clear business case also beats discounting to win deals.
What to check before you trust AI output
AI tools can state wrong figures with confidence. Check every number against its source, and delete anything you can’t find. Keep client emails, contracts, pricing and sales records out of public AI tools unless your company has approved them.
Stakeholder maps contain personal data. The UAE’s Personal Data Protection Law, Federal Decree-Law No. 45 of 2021, prohibits processing personal data without the owner’s consent, except in some cases. Companies in the DIFC and ADGM follow their own rules: the DIFC Data Protection Law, DIFC Law No. 5 of 2020 and the ADGM Data Protection Regulations 2021. Keep maps to published roles and store names only in approved systems. This is general information, not legal advice.
Agree one set of prompts per stage and a rule that no proposal goes out until a person has checked every figure. Train your sales managers first; my sales manager training in Dubai covers this.
Book a free 30-minute call
Book a free 30-minute call with me, and we’ll talk through your team’s sales goals. No cost, no obligation.
Related: Corporate sales training · AI sales training