10 Prototyping your MVP, with Bigger's Pablo Seoane

Title: Prototyping your MVP, with Bigger’s Pablo Seoane

Session context

• Format: Talk + live demo + Q&A
• Host: Louisa (Startmate)
• Speaker: Pablo Seoane (Co-founder & CEO, Bigger)
• https://www.linkedin.com/in/p-seoane/
• Audience: Launch Club founders (validation → POC → MVP build)
• Objective: How to scope, prototype, and build an MVP fast; practical frameworks and tech stack tips
• About the speaker
  • 15+ years across corporate, startups and software
  • Founded Bigger (matches founders with developers; helped 500+ startups)
  • Previously GM at a large tech company; founded subscription e‑commerce “Dating in a Box”
  • Non-technical → learned to ship software with teams; shares open-source scoping framework

Why prototype (and what counts)

• You only learn “does it work?” when users try it
• Prototyping reduces time-to-learning by cutting non-core features and shipping earlier
• Early adopters = people desperate for the problem who will try a non-perfect solution

MVP mindset: skateboard → bike → motorbike → car

Principle	Why it matters
Start with a “skateboard”	Solve the core job with the absolute minimum; get feedback immediately
Iterate in public	Each step adds comfort/speed only after user feedback
Evidence over assumptions	Validate problem, then demand, then build MVP for early users

Examples

• Airbnb v1: no payments, no map, conference-only
• Stripe v1: simple site; founders handled paperwork manually
• Dating in a Box: Wix landing page + Kmart games → bespoke boxes later
• Together (remote team bonding): PPT slide + roulette link → lightweight activity → app with Slack/Miro

Scoping framework (open-source) Pablo uses with 500+ startups

1. Problem dump → refine to one crisp problem statement
2. Personas
   • Buyer/admin (e.g., HR), facilitator/leader, participant/user
3. Golden path (end-to-end steps)
   • Login → add org/users → select experience → invite → run → collect feedback
4. User story mapping (what’s needed per step)
   • E.g., catalog, Miro/Slack integrations, invitations, analytics
5. Prioritise with MoSCoW
   • Must/Should/Could/Won’t for now; be ruthless on scope
6. Low‑fi mockups
   • Sketch/Canva/Figma; communicate flows and components without over-designing

YC-aligned build flow

• Validate problem → validate demand (landing + waitlist/preorder) → POC (manual OK) → MVP with early users → soft launch with paying users

What not to do vs what to do

Don’t do	Do instead
20-feature “MVPs”	Cut to the smallest path that solves the job
Months of stealth building	Ship in weeks; use landing tests and manual delivery
Premature brand/perfection	Learn with scrappy POCs; quality rises with each iteration
Avoid talking to users	Call early buyers; observe activation; capture objections verbatim

Build options (non-technical founder)

Path	Pros	Cons	When to choose
Do it yourself (no/low-code, AI)	Low cost; fast; retains equity	Time-consuming; limits/bugs	POC → early MVP; budget-constrained
Outsource (agency/contractor)	Faster start; senior oversight possible	Expensive; vendor selection risk	Need working MVP quickly; no tech lead yet
Hire tech co-founder/team	Deep expertise; long-term advantage	Hard to find pre-traction; opportunity cost	After early traction or if domain is complex

Guidance

• If non-technical and pre-traction: bootstrap POC, then outsource a tight MVP; later in-house or CTO
• Don’t burn cycles managing a dev you can’t manage; stay focused on users and distribution

Practical tech stack recommendations

• No/low-code for MVPs
  • Softr + Airtable/Google Sheets for data; Make.com for automations
  • Internal tools: Oracle APEX (free) for back-office workflows
  • Bubble: powerful, but steeper for non-tech; if you’ll hire a Bubble dev, consider going “proper code” instead
• AI-assisted builders (for POCs → code you can hand off)
  • Lovable, Replicate (preferred in Pablo’s tests), Bolt, Manus, Genspark
  • Expect bugs/hallucinations; debug with a human dev when stuck
• Prototyping/UI
  • Google Stitch (front-end prototyping; free), Figma, Figma Make (high-fidelity UI from prompts), Framer (site with optimized templates)
• Integrations he used
  • Slack for team invites; Miro for collaborative boards; plan Teams later

Quality and engineering hygiene (even for MVPs)

• Ask: what’s your testing approach? Don’t ship to production without basic tests
• Have rollback/monitoring; simple logs and alerts beat blind shipping
• Story: a “3‑month plan” bug charged customers 3×; testing would have caught it

Services-as-Software (Do things that don’t scale)

• Deliver value manually with your own lightweight tooling; users don’t need to see your backend
• Great to validate workflows, pricing, and objections before user-facing app is “ready”

Selected Q&A highlights

Topic	Takeaway
Tech vs business co-founder	You need both “build” and “sell”; solo is OK to start, but aim to complement skills
Outsource vs internal	Pre-traction: outsource a tight MVP while you find users; in-house/CTO later
Picking agencies	Offshore for budget; use a due-diligence checklist (Pablo to share)
No/low-code/AI framework	Prompting quality first; iterate prompts in a free LLM then use paid tools; bring a dev to debug
Real-time mobile app via no-code	Start with Softr + AI tools (Replicate, Lovable); consider POC without an app; only hire when blocked
Bubble stance	Not “anti-Bubble”: just note learning curve; if paying a Bubble dev, consider going to code for scale
Tool overload	Pick one and ship; don’t get stuck comparing tools
Extra tools	Figma Make praised for higher-fidelity outputs; Framer for sites; Stitch free for quick UI

Benchmarks and pacing

• Ship POC in days; MVP in weeks
• Early-stage: optimize time-to-learning, not completeness
• Validate with actual use and, where possible, paid pilots or trials

MVP action template (fill this in)

Area	Baseline (now)	2-week goal	Weekly target	Success criteria
Problem	Clear JTBD + personas?	One crisp problem statement + golden path	5–10 user calls validating steps	Users agree steps lead to desired outcome
Product	POC status	Clickable mock + POC (manual OK)	1 build + 1 test + 1 feedback loop	Users complete core flow; fewer manual patches
Demand	Landing/waitlist/preorders	Simple page + 2 value prop variants	2–3 creative tests; 10–20 calls	CTR/CR to pre-order/booking hit target
Tech	Tooling choice	Decide stack (Softr/Make/AI) + integrations	Ship one integration (e.g., Slack or Miro)	Integration supports end-to-end “golden path”

Practical metrics you can choose

• Activation: % who complete the core action within first 7 days
• Usage: DAU/WAU for the core experience; repeat task completion
• Retention: Day‑7 and Week‑4 return rates; cohort curves
• Love/value: “Very disappointed if removed” %, qualitative quotes
• Demand: Preorders/paid pilots/LOIs; MRR; # paying customers

Prototyping playbook

Step	Tactic
Demand test	Framer/Figma Make/Notion + Stripe checkout/preorder; run small ad tests
POC	Manual delivery via Google Slides + Miro/Slack; services-as-software backend
MVP	Softr + Airtable + Make; one integration (Slack/Miro); basic analytics & feedback loop
Iterate	Weekly: change → ship → observe → talk → decide

Action items

#	Action
1	Write the one-line problem statement and identify buyer, facilitator, participant personas
2	Map the golden path; create a user story map; run MoSCoW to cut scope
3	Draft low-fi mockups (Figma/Stitch/Figma Make); choose one stack to start
4	Build a manual POC this week; schedule 10 user sessions; instrument feedback
5	Ship a 2–3 screen “good but small” MVP in 2–3 weeks; integrate one key tool (Slack/Miro)

Workshop logistics and follow-ups

• Pablo to share:
  • Open-source Miro/Mural “design sprint” scoping framework
  • Agency/vendor due-diligence checklist
  • AI tooling session video from Bigger’s tech lead
• Tools mentioned: Softr, Airtable/Sheets, Make.com, Oracle APEX, Lovable, Replicate, Bolt, Manus, Genspark, Google Stitch, Figma, Figma Make, Framer, Slack, Miro, Microsoft Teams (later)
• Connect with Pablo on LinkedIn (“Pablo Bigger”) for resources and questions