Mustafa Peracha is Chief Consumer Officer at telecommunications company Ooredoo Qatar, where he supports the company's multi-million dollar AI adoption acceleration. With nearly three decades of experience, he shares 5 strategic AI lessons that brands of any size can learn from.
Over the past decade and a half, I've watched the Middle East's tech landscape balloon. It started with online shopping and apps, before an expansion of cloud services followed, bringing flexible, on-demand infrastructure to the region. Now we're in a third, pivotal wave: AI is essentially consolidating the power of the first two, making it much easier to leverage everything digital and cloud-based.
With AI now central to global innovation, it’s time to take our digital infrastructure and cloud to the next level, using them as necessary starting points for scalable AI deployment. The goal is to lead and shape the next era of customer engagement and operational efficiency, not just at Ooredoo, but across Qatar.
I’ve learned so much on our digital transformation journey so far. Here are five key lessons that really stand out:
1. We’re all at the same AI starting line
It’s a universal truth that big, established companies with a long history often struggle with change. Processes and culture can make them feel like they’re moving in slow motion compared to nimble startups. On the other hand, who wants to make an expensive first mistake? We definitely felt aspects of that during the early days of digital and cloud adoption. But AI has fundamentally rewritten the rules.
I genuinely believe that in the AI space, there is zero legacy. It is the great digital equaliser that should excite every leader and marketer. It doesn’t matter if you're working for a massive enterprise or a small, ambitious startup, everyone is starting at the exact same point in the AI race.
Ooredoo’s leadership sees AI implementation as a natural driver for action, not an internal hindrance. A company’s size and age no longer determine its competitive edge in AI, its agility and commitment do. We all have the chance to define our companies’ futures right now.
2. Productivity beats percentage increases
I believe one core benefit of AI is consistently undersold. Boardrooms focus on the commercial side, improving customer experience and driving revenue. While that’s important, the often-overlooked advantage is that AI is a massive productivity enabler, a benefit that’s often harder to track.
Take our recent Nojoom customer loyalty campaign. The entire creative process was streamlined by AI, from generating the master visual with Imagen 4 to creating multiple frames with Gemini 2.5 Pro, and finally, generating the video footage using Veo 3’s image-to-video feature. The whole ad was finished in eight hours, including post-production. Watch it here:
While the campaign exceeded audience reach and engagement goals, the bigger win was the productivity boost and the efficiency gained by our internal teams, freeing them up for strategy.
3. Responsible AI deployment rests on you, too
I believe that the way AI is deployed rests on the shoulders of the brands using these tools to engage people, not just the developers coding them.
Ooredoo Qatar’s AI approach is one of deliberate caution. We don't just open the floodgates. Instead, we adopt a 'sandbox approach'. That means running every untested AI model in a secure, isolated environment, completely cut off from our live customer data. This is a strategic mandate because, alongside protecting data privacy, our highest priority is protecting cultural identity. We can’t risk our customers feeling that AI is inadvertently diluting the unique, multinational culture of Qatar.
This challenge is amplified because current AI, by its nature, pushes a 'universal culture’. Whether the engine is running in London or Doha, it speaks a common language. This creates a critical friction point. How do we retain local relevance and cultural nuance, especially in the Middle East, while leveraging tools that are designed to be pretty generic across the globe?
Regional brands can ensure AI innovation is beneficial to their businesses and culturally sensitive by setting clear boundaries. It’s important to define the absolute, non-negotiable lines regarding cultural norms, and hold AI output accountable, vetoing anything that breaches these rules.
Brands should also open the AI tap gradually, moving carefully from limited, controlled experiments to full-scale deployment only after validation. And, finally, your AI strategy must be designed to amplify your brand's unique local voice and customer values, not relying on generic, one-size-fits-all messaging.
4. AI research is slowing you down
The saying, ‘Jack of all trades, master of none’, is very relevant in the AI space. The AI noise is overwhelming, and marketers often get obsessed with research, failing to pick a starting point. You can avoid the trap of perpetual learning without execution by picking an objective and AI tool, testing, learning, and scaling success.
For example, our marketing team faced a big hurdle recently: How to link online sales directly to revenue value to actually prove our Search campaigns were driving growth. So, we pivoted to AI-powered Value-based Bidding (VBB). This is a Google Ads smart strategy that automatically sets bids to maximise the total value you receive, not just the number of clicks.
Working with Google and our media agency, Croud, we ran tests that quickly confirmed VBB could maximise total revenue better than standard bidding. This move was a game-changer. We saw a 37% increase in Return on Ad Spend (ROAS) and 43% more conversions, leading to a 39% rise in prepaid revenue from Search. We're now scaling this successful VBB approach to other campaigns.
In another example, our team needed to unlock insights hidden in vast, messy customer feedback without spending a ton of money and time on analyst training. The first GenAI tool we used didn’t work, but then we found Google Agentspace, now part of Gemini Enterprise. It consolidates analytical workflows, connects different datasets, and uses large-language models (LLMs) to complete tasks with just a prompt. In our case, it took raw, unstructured data and quickly synthesised it into qualitative insights that immediately helped us figure out how to improve our customer experience. Our willingness to keep testing against that defined problem, even after the first attempt failed, is what led us to the right solution.
5. AI can help create new revenue streams
For marketing leaders in any industry, a huge question is always, ‘how do we generate extra cash beyond just selling our main products?’. The answer often comes down to monetising the digital assets you already own, a strategy the global telecom industry is addressing head-on through Application Programming Interface (API) monetisation. Think of an API as a secure digital messenger that lets two different software systems talk to each other.
Ooredoo Qatar tackled this by turning its core telecom assets into a new, highly valuable service. We’ve built trusted customer relationships and billing relationships over many years. With Google Cloud’s Apigee platform, which acts as a secure manager for APIs, we can now safely give partners controlled access to the functions they need. This means they can reliably plug into our system to securely verify a customer's identity or process a charge, successfully expanding our revenue streams outside of connectivity sales.
You don't have to be in the telecom industry to grow this way. Any brand can explore revenue expansion by identifying a unique asset and taking it to the world. Don't think of yourself as just a retailer, or a service provider. Identify the valuable, proprietary data or functionality your business owns. For example, customer loyalty data, specific logistics expertise, or unique verification processes. Next, transform that asset into a standardised, usable API, and figure out how AI can help you expand and improve on it.
In our case, AI optimises pricing, predicts partner demand, identifies the most valuable customer segments to offer to partners, and monitors API usage, making this revenue stream scalable.
