3x Founder | Xoogler | Design Thinking | Follow for daily tips on how to launch, grow, fundraise for and steer your startup.

TL;DR: I bring 15+ years of digital innovation, consultancy, startups and strategic thinking experience to help your startup thrive.

Are you looking to:

→ IDEATE: Zero in on innovative ideas that could disrupt the landscape?
→ BUILD: Build and validate concepts, FAST, using iterative design?
→ GROW: Fully launch and scale using end-to-end go-to-market planning?
→ RAISE: Nail upcoming rounds with a killer fundraising strategy and prep?

Let's chat.

━ 

ABOUT ME:

→ Advisor: Helped 20+ ventures launch, iterate, scale, fundraise, and unlock growth using innovation
→ Founder: Zero to 970k MAUs in 2 years (100% organic)
→ Google: Portfolios of $50M-$100M ARR, focused on bizdev, digital consulting and partnerships
→ Strategist: 100+ strategy and GTM workshops with the biggest global brands
→ Innovation: Stanford d.school bootcamp, IDEO.U, AJ&S, Design Sprint Master, and more.

━ 

Want daily strategies about how to run your startup?
→ Follow me on Linkedin.

Want some (out of the box) advice about your venture?
→ https://gxjzup7dc3p.typeform.com/to/I9QZmwcg

Want to know what other people think of Ruya Advisory?
→ https://testimonial.to/ruya-advisory/all

Majd Alaily

I'm a tech entrepreneur who builds MarTech solutions - I launched 6 products and successfully sold one of them. With +10 years in digital analytics and marketing automation, I combine hands-on software development with deep marketing experience. Previously, I worked at Google as a Programmatic Ads Specialist and gained experience as a performance marketer on the agency side.

My background combines coding, marketing, and analytics. This mix of skills enables me to create user-friendly solutions that connect technology and marketing effectively.

Ahmed Ali

Big update for Google Ads API.

Performance Max channel breakdown is now supported in the new v23.

No more workarounds to pull these insights.

It took Google a long time to bring this... advertisers have been asking forever.

When Google initially launched PMax, they had zero intention of giving this level of transparency.

It fits their bigger vision:
give advertisers less control,
move budget wherever the system wants,
and avoid the questions that come with visibility.

Because the moment you see channel breakdown, you start asking:

Why is spend moving here?
Why is this channel getting most of the budget?
Why did performance drop even though budget went up?

But… here we are.

If you work with PMax reporting/audits, v23 just made your life easier.

Vibe coding for marketing is the new default.

The "Technical Marketer" role is being rewritten.

Coding used to be a barrier for marketers. You had an idea or a task, but you didn't have the Python or SQL skills to build it.

Now? The wall is gone.

We have AI agents that turn a "vibe" into a workflow, a script, or a dashboard. Whatever you need, you just describe it.

This is why coding agents (like Claude Code) are currently the best general agents. Because they have access to the file system and the command line, the agent has full freedom to actually build the solution, not just talk about it.

But for marketing, you still need a specific setup:

1- Guard-railed tools: To ensure data accuracy.
2- Human-in-the-loop: To approve and provide the strategy that AI lacks.

The barrier to entry has officially collapsed. You don’t need to know Python; you just need to know how to describe what you want.

If you’re still doing things the "old way," you aren't just slower....you're playing on hard mode for no reason.

The era of the technical marketer is shifting: It’s less about writing code and more about directing systems.

We ran 1,000+ ad-ops tasks through Gemini 3, GPT-5.2, and Claude Opus 4.5.

This wasn’t copywriting. Real ads operations:

- Creating campaigns + ads with the right structure
- Root-cause analysis (CPA spikes, spend drops, tracking issues)
- Optimization recommendations that can turn into actions (not just text)

We judged each model on 2 things:
- Performance (accuracy + reliability)
- Cost (how expensive it is to get to a usable result)

Here’s what we learned 👇

GPT-5.2
Most consistent for ads automation, especially when it comes to respecting platform schemas and producing clean tool outputs.

It follows instructions very closely, which sometimes means it can overcomplicate simple tasks and spend more time than necessary.

Gemini 3 Flash
Fast and reliable for straightforward tasks, especially for the price.
If you have repetitive tasks, like optimizing search terms, it’s perfect because it stays consistent. The tradeoff: it can jump to conclusions and make more assumptions.

Gemini 3 Pro
Top-tier reasoning without heavy tool usage. It plans really well. It does make more tool-calling mistakes than GPT-5.2, but it usually catches them and corrects itself

Claude Opus 4.5
God-tier for complex tasks that require deep analysis + coding (root-cause analysis is where it shines). The main drawback is cost, often you can get a similar outcome with models that are ~5x cheaper.

If I had to pick:

1- One model for most ads tasks → GPT-5.2 (best value for reasoning + reliability)

2- For repetitive, straightforward tasks → Gemini 3 Flash

We support all of these in Markifact (and you can bring your own API keys).

What’s your experience been with these models?

Amazon Ads has an API that feels like it was built specifically to discourage anyone from building on top of it.

If you want to pull a simple report on how much was spent yesterday, a single report can take up to 3 hours to complete.

And somehow, with every API update, it gets slower. I’m not even joking.

The Amazon team says this is “intended behavior” because it’s an offline API, meaning you make the call… and go to sleep.

It’s obvious they’re pushing Amazon Marketing Stream, but it’s not a fit for third-party apps.

I was hesitant to add Amazon Ads to Markifact. But our users kept asking for it, so we shipped it.

It seems the only practical way to work with it is a pipeline:
schedule the report → wait → store it (BigQuery / Sheets) → then query your stored data.

It still burns compute hours, because you have to poll the report every few minutes.

Anyone found a better way to work with Amazon Ads reporting?

taplio

The Battle

Majd Alaily

Ahmed Ali

Followers

Posts

Engagements

Likes

Performance ScoreAhmed Ali's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Majd Alaily

Ahmed Ali

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Ahmed Ali's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreAhmed Ali's LinkedIn Performance