💫 Ramplify: August Afterword
Advice on advisors, Quantum breakthrough, Google Prompt Energy
(What’s Driving Investor Focus Now,” alongside Alister Coleman, Carolyn Breeze GAICD, and Matthew Browne, brilliantly hosted by Chris Ellis.)
👋 Fresh off S2S Summit 2025: Most startup advisors are expensive comfort; the right ones are force multipliers by Andrew Poesaste
On the panel, I discussed advisors, and it seemed like there was a missing link for founders on how to navigate that world. So, I thought I’d put together a founder-first playbook to help you pick, pay, and manage advisors without burning your equity.
Two recent examples from our portfolio tell the story. The first founder brought on a former product leader for hands-on help - weekly check-ins, metrics frameworks, and roadmap feedback. For 6-12 months, this advisor was gold. Then their usefulness dwindled as the company scaled. The founder cut the vesting short at 1/3 of the original schedule. Both parties walked away happy.
The second founder landed a prominent NASDAQ executive who opened their network 2-3 times per year and gave razor-sharp GTM feedback. Less frequent, but worth their weight in gold when it mattered. Expectations managed from day one.
Bottom line up front: Your advisor should de-risk a stage, not rubber-stamp feelings.
How to pick and trial the right people
Purpose matters more than pedigree. You're hiring expertise to solve defined problems, not collecting impressive names. The best advisors ask uncomfortable questions that push you back to first principles—exactly what you need when you're drinking your own Kool-Aid.
Share your chessboard first: current stage, 12-month outcomes, key bottlenecks, what you've tried, open questions. This becomes your advisor's brief.
Source 3–5 candidates who've solved your exact problem multiple times
Listen for pattern depth - can they spot failure modes you haven't considered?
Pass immediately on generic playbooks, name-dropping without substance, or promises to "open doors" with no names
Run 30–60 day trials with 1–2 finalists; agree upfront on 1–2 concrete artefacts (hiring scorecard, pricing test plan)
There's a reason and a season for every advisor. Bring them in to solve specific problems, then honestly reassess.
Structure the engagement
Equity: 0.10%–0.25% at seed for hands-on help (that's $10K-50K on a $20M exit, $100K-500K if you hit unicorn status). Monthly vesting over 3–4 years with a 3-month cliff. No front-loading.
Agreement: Standard advisor doc with IP assignment, confidentiality, and conflict disclosures. Cash-only when the scope is concrete and near-term.
SLA (not vibes): Cadence (2×45 min/month), 48-hour async response, single channel, clear outcomes by next milestone.
Quarterly review: Renew, reshape, or cut early. Unvested equity returns to the pool -once vested, it's gone forever.
Signals
Green: Specific experience with your exact problem; crisp failure modes to avoid; willingness to be measured; energy that speeds your learning.
Red: "I'll open doors" with no names; asks >0.25% at seed; guaranteed outcomes; recycled generic decks; context-blind playbooks.
The game is afoot: Choose wisely, structure thoughtfully, don't be afraid to make changes when relationships stop delivering value. If you're building something that needs this kind of strategic thinking, we should chat.
🗞️ Tech News:
Is this the breakthrough in Quantum everyone’s been waiting for? ‘Neglectons’ + Ising anyons for universal quantum computing. USC-led researchers revived a once-discarded theoretical particle—the neglecton—and showed that pairing a single, stationary neglecton with Ising anyons enables universal topological quantum computation using braiding alone (no magic states or extra measurements). The work reframes “mathematical garbage” from non-semisimple TQFTs as the missing piece for fault-tolerant quantum logic, with peer-reviewed results tied to the underlying arXiv/Nature Communications study.
“Those discarded objects turn out to be the missing piece… like finding treasure in what everyone else thought was mathematical garbage.”
Google turns Gemini into an AI ‘school system’ — free for Workspace for Education. Teachers can assign safe, teacher-led AI experiences (custom “Gems,” NotebookLM notebooks) directly in Classroom, while students get auto-generated quizzes, visual explainers and higher usage limits - all with admin controls and data protections, at no extra cost. It’s a big push to put AI tutoring, content creation and assessment drafting into everyday workflows across schools.
The State of AI 2025 - Bessemer Venture Partners. Bessemer maps today’s AI startup landscape into two winning archetypes: “Supernovas” racing to ~$100M ARR in ~1.5 years (often with thin/negative gross margins) and “Shooting Stars” compounding to ~$100M over 4 years with stronger unit economics. They predict agentic AI moving into the browser layer (the new operating surface), 2026 as generative video’s breakout year, and a wave of incumbent M&A - while urging founders to build private, trustworthy evals, “systems of action,” and memory-driven moats.
How much energy does one Gemini prompt use?. Google published a per-prompt accounting for AI inference: a median Gemini Apps text prompt uses ~0.24 Wh energy, 0.03 g CO₂e and ~0.26 mL water, with efficiency gains of 33× (energy) and 44× (emissions) over 12 months.
“The energy consumed per median prompt is equivalent to watching television for
less than nine seconds.”
🛠️ How to Startup:
Getting started with usage-based pricing (for AI features) — A super-practical playbook from Stage 2 Capital’s Liz Christo (with pricing operator Kyle Hart) on shifting beyond flat subscriptions. TL;DR: don’t default to pure consumption—offer optionality (credits/tiers alongside fixed plans), treat pricing like a product hypothesis, validate with sales pilots and interviews (not just surveys), and future-proof your billing so you can evolve as usage explodes. Great if you’re rolling out AI features with spiky, uneven adoption.
“I like optionality in pricing wherever possible… it puts the decision in the customer’s hands.”
Startup org design: design power centres intentionally — Molly Graham (ex-Facebook/Quip/CZI) argues early companies must explicitly choose where decision-making authority lives (e.g., Product, GTM, Operations) and revisit that design as they scale. Clues that your “power centres” are mis-set include decision latency, cross-team priority collisions, and “two bosses” ambiguity. Fixes include single-threaded ownership, clear escalation paths, and time-boxed mandates—before you ossify into slow matrix structures. Pair this with the broader “stay functional longer” founder-mode philosophy popular in product-led companies.
Self-awareness over grind: Kim Hansen on leading for the long game — A candid note from Cake Equity’s CEO argues that founders don’t win by projecting invincibility; they win by building self-knowledge, creating tight feedback loops, and closing the gap between how they see themselves and how others experience their leadership. Useful reminder to treat mental fitness like any other core system in the company — deliberate, measured, and designed to last.
“The answer isn’t grinding harder. It’s knowing yourself better.”
“The more honest I became about my limitations, the stronger I became as a leader.”
👕 VC in Australia/New Zealand:
Canberra elevates AI to a national priority. Following the Productivity Roundtable, Industry Minister Tim Ayres is fast-tracking an economy-wide AI Capability Plan (compute & data centres, skills, local capability, diffusion), previously slated to land by end-2025. Before proposing any new AI laws, Treasury will run a “gaps analysis”, amid competing pushes from business (minimal new rules) and unions/academics (tighter guardrails and worker protections).
“We need to build sovereign capability while making sure AI’s benefits diffuse across the real economy — and manage the risks as we go.”
New Zealand’s tech quietly surges: 6.1× growth since 2019. A new Dealroom/NZGCP report finds NZ’s ecosystem value up 6.1× since 2019, outpacing peers including Australia, Ireland, Israel, and the Netherlands. With 8 unicorns and 3 decacorns, it ranks #9 globally despite comparatively low VC funding. 2025 is already the second-highest year on record for VC raised (NZ$1.2B+), boosted by FNZ’s ~NZ$773M (US$500M) growth round.
🤩 Portfolio News
Humble, honest, happy, hungry, and high-performing - these five H’s are the secret sauce behind Sendle’s thriving team culture. The AFR sat down with Sendle Co-founder James Chin Moody to explore how he motivates a remote-first workforce, why the order of the five H’s matters, and the daily routines that keep Sendle seamlessly operating across time zones.
Quantum Brilliance is a small but mighty contender in quantum hardware. In the latest 2025 Quantum Market Sizing Report from The Quantum Insider, Quantum Brilliance was named as the third largest vendor based on share of total systems sold. Although heavy hitters like IBM (lead in deal value) and IQM (lead in unit volume), Quantum Brilliance is showing great commercial traction as the quantum computing industry begins the commercialisation phase.
Goterra is igniting change and propelling the conversation on the circular economy at the Circular Future Forum. Catch Olympia Yager (Founder and CEO of Goterra) speaking in Orange from 3-5 September! The forum brings together innovators, policymakers, businesses, councils, and community leaders to unlock practical solutions for a circular economy and a net-zero future.
💼 Portfolio Job Board
Customer Success Manager at Insightech
Location & type: Sydney or Hybrid, Full Time
What they do: Insightech is a leading digital experience analytics company empowering teams to drive exceptional digital experiences through actionable insights and real-time data.
What you’ll do: Key responsibilities include maintaining strong customer relationships and supporting account growth and customer success.
Senior Infrastructure Engineer at Skedulo
Location & Type: Fortitude Valley, QLD (Hybrid) & full-time
What they do: Skedulo creates cloud-based software to allow any company in any industry to schedule, manage, engage, and analyse their deskless workforce.
What you’ll do: You will design, build, and maintain scalable, secure infrastructure and Kubernetes environments with a focus on automation, observability, incident response, and cross-team collaboration.
Do you have a job you’d like us to promote? Add it to Hatch and share the link with us.
See all jobs across the Rampersand Portfolio.
🌏 Rampersand Travel Diary:
Andrew will be in Melbourne the week of September 6th. Reach out for coffee.
🎉 Upcoming Events:
Startmate Alumni Demo Day. Thu 11 Sep 2025, 5:45–8:00 pm AEST — ICC Sydney — FREE — Andrew & Tas are attending.
Thanks for being part of the Rampersand Community! You can stay updated with even more news and info on our LinkedIn and X/Twitter pages.