Professional

Context

When nCinga was folded into the group, I inherited the integration on the product side. Two product teams, two roadmaps, two overlapping customer bases, and a deadline to make it look like one company by [milestone].

Approach

• Ran a two-week product audit across both orgs — what was shipping, what was actually being used, what was quietly on life support
• Made the call to sunset [X features] and consolidate around [Y core surface]
• Rebuilt the combined roadmap around the merged customer segments rather than the merged teams
• Sat with the nCinga leads individually to walk them through what was changing and why, before publishing anything

Outcome

[Roadmap consolidation complete within Z weeks. Zero customer churn in the transition window. Retained X% of the nCinga product team through the merger.]

What I learned

The hardest part of PMI isn't the product decisions. It's that the people you're merging with have already told their team what the future looks like, and you're asking them to say something different next Monday. Give them room to change the story themselves.

Context

A GCC client wanted to know how much AI compute demand would materialize regionally over [X] years, and what share could be captured domestically versus routed through hyperscalers abroad. This was before "sovereign AI" was a common phrase.

Approach

• Built a bottom-up model across training, inference, research, and enterprise-inference demand pools
• Layered in supply-side constraints: fab capacity, export controls, regional power availability, data center build timelines
• Cross-referenced demand against announced regional projects to identify overcapacity and undercapacity by [year]
• Wrote the recommendation around where sovereign compute buildout was economically defensible versus symbolic

Outcome

[Recommendation delivered to client. Model became internal reference asset for adjacent engagements. Subsequent industry data validated the directional call.]

What I learned

Forecasting markets that don't exist yet is mostly about being disciplined about which numbers you trust and which you treat as narrative. The useful output is rarely the number itself — it's the sensitivity analysis around what would have to be true for the number to be wrong.

Context

By [month year], Zilingo — the fashion supply-chain platform I'd been at for [X] years — had entered a very public period of crisis. I was leading product for [line]. Every incentive around me said pause: freeze the roadmap, wait it out, protect the balance sheet.

The call I made

I argued for the opposite. This was exactly the moment to ship the [product] we'd been building. Enterprise customers were watching to see whether we were still a going concern, and nothing signals going concern like shipping.

Approach

• Replanned the launch around the two customers I trusted to anchor it
• Restructured pricing so the customer commitment felt derisked from their side — [specific mechanic]
• Coordinated across engineering, GTM, and legal in a company where the org chart was rewriting itself weekly
• Wrote the internal narrative for the launch so leadership had one story to tell externally

Outcome

[X customers signed within Y weeks. $Z ARR contribution. A% retention through the crisis window.]

What I learned

The instinct to pause during turbulence is almost always right for the balance sheet and almost always wrong for the product. The two customers who trust you enough to sign in the worst quarter of your company's life are the two you keep for the next decade.

Context

[Two sentences on what you were solving.]

Approach

• [What you did]
• [What you did]
• [What you did]

Outcome

[Metrics.]

What I learned

[One or two honest sentences.]

Context

Solvei8 (Zilingo's B2B SaaS arm) had grown as a sales-led org: enterprise deals, long cycles, high-touch onboarding. The economics worked but didn't scale. My job was to figure out whether — and how — we could turn on a product-led motion without breaking the enterprise business paying for it.

Approach

• Segmented the customer base to isolate which use cases were self-serviceable and which structurally weren't
• Rebuilt onboarding around a first-value moment measurable in days rather than months
• Introduced [specific PLG mechanic — usage-based tier, in-product trial, free-forever slice]
• Kept the enterprise motion intact by drawing a bright line between the two funnels internally, so sales didn't feel undercut

Outcome

[Signup-to-active conversion moved from X% to Y%. CAC on the PLG cohort was Z% of the SLG cohort. Enterprise pipeline held.]

What I learned

SLG-to-PLG transitions fail when they're framed as replacements. They work when framed as parallel funnels that trade leads back and forth — and when someone senior is willing to defend the sales team's numbers publicly through the transition.

Context

Amazon UK's seller onboarding funnel had a drop-off between [step X] and [step Y] that had been sitting on the dashboard for months, treated as a fixed cost of the funnel. I got assigned to look at it as a data question.

Approach

• Wrote the SQL myself rather than requesting a pull, because the shape of the query kept changing as I understood the funnel
• Segmented the drop-off by seller category, geography, and product line, and found the drop was concentrated in [specific segment]
• Traced the cause to [specific mechanic]

Outcome

Recommended [change]. [Post-change conversion moved from X% to Y%, worth an estimated $Z in annualized GMV once rolled out.]

What I learned

Aggregate funnel numbers hide almost everything worth knowing. The fastest way to run a real analysis is to write the queries yourself — every layer of intermediation costs you a day and a hypothesis.

Context

[Two sentences on the org and what you were brought in to do.]

Approach

• [What you did]
• [What you did]
• [What you did]

Outcome

[Deliverable, decision, or metric.]

What I learned

[One or two sentences.]

Context

[Two sentences.]

Approach

• [What you did]
• [What you did]
• [What you did]

Outcome

[Deliverable or metric.]

What I learned

[One or two sentences.]