LiveSold

1. One-liner

AI copilot that auto-captures live-stream comments, generates invoices, and syncs inventory for Filipino and SEA micro-sellers.

2. Trend signal — why now?

Live selling is the dominant commerce format in Southeast Asia, and the Philippines is ground zero. The numbers are absurd:

• 73 million livestream shoppers in the Philippines alone (Truelogic, 2025)
• 350,000+ active sellers on Shopee Philippines as of end-2024
• 47% of Filipino entrepreneurs sell daily via live social media, spending 14 hours per week on live sessions (ScienceDirect, 2024)
• Philippines social commerce projected to grow from $28.4B (2025) to $96.4B by 2034 at 14.5% CAGR
• TikTok Shop, Shopee, and Lazada live selling expected to represent 25-30% of SEA’s online GMV by 2026
• Filipino sellers managing 100+ daily orders spend 3-5 hours/day on manual order-pulling, inventory updating, and label printing (BigSeller, 2026)
• CommentSold (US) processes $1.2B+ annualized GMV from comment-to-checkout automation — proving the model works

The problem: US tools (CommentSold at $49-149/mo + 3-5% commission, SoldLive, Dibsly) don’t support SEA marketplaces, SEA payment rails, or SEA price points. BigSeller and OneCart handle inventory but don’t do live-stream comment capture. Nobody owns the “comment → invoice → inventory sync” workflow for SEA live sellers.

Provenance:

• Signal 1: 73M PH livestream shoppers + 47% of sellers doing daily live selling, spending 14hrs/week — ScienceDirect + Truelogic — 2024-2025
• Signal 2: CommentSold proves comment-to-checkout model at $1.2B+ GMV, but is US-only, Shopify-only, priced at $49-149/mo + commission — Tracxn — 2026
• Signal 3: SEA social commerce projected at $186.5B by 2030 (31.4% CAGR), with live commerce becoming “structured and operational” per GlobeNewsWire APAC report — GlobeNewsWire — April 2026 Category: Geographic arbitrage + Platform shift

3. The opportunity

In the US, CommentSold turned the “comment SOLD on a Facebook Live” gesture into a $1.2B GMV software business. They acquired Popshop Live and dominate US boutique live-selling.

But here’s the gap: CommentSold doesn’t work for SEA. No Shopee integration. No Lazada. No TikTok Shop SEA. No GCash or Maya invoicing. No support for COD (cash on delivery), which is still 40-60% of SEA e-commerce. Pricing starts at $49/mo — that’s ₱2,800, which is a week’s minimum wage in the Philippines.

Meanwhile, Filipino live sellers are drowning. They go live on Facebook or TikTok, comments flood in — “mine po!”, “order!”, “pa-reserve” — and they’re manually writing down names in a notebook or spreadsheet. After the stream, they spend 2-4 hours matching names to orders, DM-ing each buyer for payment, and updating stock counts across Shopee and Lazada. Overselling is constant. “Joy mining” (fake order comments) wastes hours of follow-up.

BigSeller handles inventory sync for free — but it’s an ERP, not a live-selling tool. It doesn’t watch your stream and capture orders. Nobody in SEA is doing what CommentSold does in the US.

LiveSold fills that gap. A mobile-first app that watches your live stream, captures order comments using keyword detection + AI, auto-sends payment links via Messenger/SMS (GCash, Maya, or COD), deducts inventory in real-time across connected platforms, and flags joy miners by scoring buyer history.

4. Target market

• Primary customer: Solo live sellers and micro-entrepreneurs in the Philippines, age 25-45, running 3-7 live sessions per week on Facebook Live and/or TikTok, handling 20-200 orders per session. Typical monthly GMV of ₱50K-500K ($900-$9,000). They sell fashion, beauty, home goods, and food.
• Why they buy: They lose 2-4 hours after every live session to manual order matching, DM chasing, and inventory reconciliation. They oversell regularly, leading to refunds and negative reviews. They can’t grow because the manual workflow caps their throughput.
• Rough TAM reasoning: 350,000+ active Shopee PH sellers. If 47% do daily live selling, that’s ~165,000 live sellers in PH alone. Add Indonesia (larger e-commerce market), Thailand, Vietnam, Malaysia — conservatively 500,000-750,000 active live sellers across SEA. Even 1% penetration at ₱799/mo = significant revenue.
• Why now for them: DTI’s Internet Transactions Act (fully enforced June 2025) requires transparent pricing, merchant ID display, and complaint mechanisms — pushing sellers toward structured tools. TikTok Shop’s 2025-2026 ASEAN push is onboarding massive numbers of new live sellers who need tooling. Platform commissions (Shopee: 4-8%, TikTok: 1-5%) are rising, making efficiency critical to margins.

5. Product sketch (MVP)

• Comment capture engine: Connects to Facebook Live / TikTok Live via API; watches comments for “mine”, “order”, product codes, and variations in Filipino (Tagalog) and Bahasa; creates a real-time order queue during the stream
• Joy miner scoring: Flags commenters with history of non-payment, auto-deprioritizes their orders; shows seller a trust score per buyer
• Instant invoicing: After the session (or in real-time), auto-sends payment links via Facebook Messenger or SMS — GCash, Maya, PayMongo (cards), or COD booking
• Inventory sync: Connects to Shopee, Lazada, and TikTok Shop seller accounts; deducts stock in real-time as orders are confirmed; prevents overselling
• Mobile dashboard: The entire workflow runs on the seller’s phone — stream management, order queue, payment status, inventory view
• Post-stream recap: Summary of the session — items sold, revenue, unpaid invoices, inventory adjustments — exportable as a simple report

6. AI angle — what’s load-bearing

AI is doing the hard work here, not decorating a dashboard:

1. Comment parsing in mixed language: Filipino live-selling comments are a chaotic mix of Tagalog, English, shorthand, and emoji. “Mine po yung blue XL”, “order ko po #3”, “pa-reserve ng 2pcs” — this requires NLP that understands Filipino commerce slang, not just keyword matching. Off-the-shelf models (GPT-4o-mini, Claude Haiku) can handle this at pennies per session with the right prompting.

2. Joy miner detection: Pattern recognition across buyer history — flagging accounts that comment “mine” but never pay, new accounts with no purchase history, rapid-fire comments on high-value items. This is a lightweight classification problem that gets better with data.

3. Smart inventory allocation: When 3 people comment “mine” on the last unit, AI decides allocation based on timestamp + buyer trust score, then auto-notifies waitlisted buyers.

Without AI, you’d need the seller to manually read every comment, judge intent, and track order priority. That’s exactly the bottleneck they’re stuck in today.

7. Localization angle

This is the localization play. The entire product concept is geographic arbitrage — taking a proven US model (CommentSold) and rebuilding it for SEA realities:

• Language: Filipino/Tagalog comment parsing is non-negotiable. Bahasa Indonesia for expansion. English as a secondary language. None of the US tools handle this.
• Payment rails: GCash (94M users), Maya, PayMongo, COD via J&T Express / Ninja Van. US tools use Stripe/Shopify Payments — useless in SEA.
• Pricing: ₱499-999/mo ($9-18/mo) vs. $49-149/mo for US tools. This is 5-10x cheaper, matched to SEA wallets.
• Platform ecosystem: Shopee + Lazada + TikTok Shop, not Shopify + Amazon. Totally different API landscape.
• Cultural norms: COD is 40-60% of PH e-commerce. “Joy mining” is a Philippines-specific problem. The product must handle both.
• Distribution: Facebook Groups for sellers (massive in PH), TikTok seller communities, J&T/Ninja Van partner channels.

8. Business model — path to $1M-$5M ARR

• Pricing: ₱499/mo ($9) Starter (1 platform, 500 orders/mo), ₱999/mo ($18) Pro (3 platforms, unlimited orders, joy miner scoring), ₱2,499/mo (~$45) Business (team accounts, analytics, API access)
• ACV: Blended ~$150/year (weighted toward Pro tier as sellers grow)
• Rough math to $1M ARR: 6,700 customers × $150/year = $1M ARR. That’s 1.3% of the estimated 500K SEA live sellers. Achievable in 18 months.
• Rough math to $5M ARR: 33,000 customers at $150 ACV, or fewer customers at higher ACV as the Business tier grows. Indonesia expansion doubles the TAM.
• Expansion path: Transaction fees on payment processing (0.5-1% spread), premium analytics for brands working with live sellers, white-label for agencies managing multiple live sellers.

9. Go-to-market wedge — first 100 customers

1. Facebook Groups infiltration: There are dozens of large Filipino seller communities on Facebook (e.g., “Online Sellers PH”, “TikTok Shop Sellers Philippines”) with 50K-200K members each. Post demo videos showing the manual vs. LiveSold workflow. Offer 30-day free trial. Target: reach 5,000 eyeballs, convert 2% = 100 trial signups, convert 30% = 30 paying.

2. TikTok seller program partnerships: TikTok Shop runs ASEAN MSME programs. Position LiveSold as a recommended tool. Even without an official partnership, sellers enrolled in these programs are the highest-intent segment.

3. J&T Express / Ninja Van co-marketing: These couriers serve every live seller in PH. They have seller newsletters and Facebook groups. Offer a co-branded integration — “Book your COD pickup inside LiveSold” — and get distribution through their seller base.

4. Live demo streams: Do a live selling session yourself using LiveSold, streaming on TikTok PH. Show the product working in real-time. This is the ultimate product demo for this audience — they live on these platforms.

5. Micro-influencer seeding: Give the tool free for 6 months to 20 mid-tier live sellers (10K-50K followers) who stream daily. Their followers see the tool in action. Word of mouth in tight seller communities is powerful.

10. Build complexity — justification

Medium. The comment capture engine requires API integration with Facebook Live (Graph API) and TikTok (Seller API), plus a lightweight NLP layer for Filipino comment parsing. Payment integrations via PayMongo (aggregates GCash + Maya + cards) simplify the payment side. Shopee/Lazada/TikTok Shop seller APIs are documented and accessible. Mobile-first React Native or Flutter app is standard. The main complexity is reliable real-time comment streaming and the joy miner ML model, which can start as simple heuristics. A team of 2-3 can ship v1 in 12-14 weeks.

11. Gating checklist

Gate	Pass?	Note
Legal in target market	✅	Standard SaaS tool, no regulated data. DTI Internet Transactions Act actually pushes sellers toward tools like this.
Ethical — no harm / dark patterns	✅	Helps legitimate sellers operate more efficiently. Joy miner flagging is protective, not discriminatory.
Market exists (evidence above)	✅	350K+ active Shopee PH sellers, 47% doing daily live selling, $28.4B PH social commerce market.
1-5 person team can build this	✅	2-3 engineers, 12-14 weeks to v1. Standard web/mobile stack with API integrations.
Launchable with <$50K / ₱40L	✅	₱15-25L ($2,700-$4,500) for 3-month build + initial marketing. Infrastructure costs near-zero at low volume.

12. Feasibility score

Axis	Weight	Score	Notes
Problem intensity	20	17/20	Sellers lose 2-4 hours per session to manual matching. Overselling causes refunds and bad reviews. Real money lost daily.
Demand evidence	15	12/15	CommentSold proves the model at $1.2B GMV in the US. 350K+ active PH sellers with 47% doing daily lives. No localized equivalent exists. One strong signal gap: haven’t found explicit “I’d pay for this” quotes from PH sellers specifically.
Build feasibility	15	12/15	APIs exist for all platforms. PayMongo handles payments. NLP for Filipino comments needs light tuning but LLMs handle it well. 12-14 week timeline is realistic.
Distribution clarity	15	13/15	Facebook seller groups, TikTok seller programs, courier partner channels are all concrete and free. The demo-via-live-stream approach is uniquely powerful for this audience.
Revenue mechanics	15	11/15	Pricing at ₱499-999/mo is affordable for the target but ACV is low ($150/year). Need volume. Transaction fee upsell helps but isn’t in v1.
Time to first revenue	10	7/10	12-14 weeks to build, then 30-day free trials. First revenue at ~week 16-18. Not instant but acceptable.
Defensibility	10	6/10	Soft moat via Filipino language NLP tuning, joy miner scoring data, and seller workflow lock-in. A well-funded competitor (or BigSeller) could copy. Speed and community matter.
Total	100	78/100

13. Qualitative modifiers

Founder-fit tags

technical-heavy — The founder needs to build real-time comment streaming, NLP pipelines, and multi-platform API integrations. Domain expertise in live selling helps but isn’t required — the product tells the story.

Key assumptions to validate (3-5)

1. Assumption: Filipino live sellers will pay ₱499-999/mo for automation they currently do manually. How to test: Survey 50 active live sellers in 3 Facebook groups. Ask: “How many hours do you spend on post-stream order matching?” and “Would you pay ₱499/mo to cut that to 10 minutes?”

2. Assumption: Facebook Live and TikTok APIs provide sufficient comment stream access for real-time capture. How to test: Build a proof-of-concept that connects to a test Facebook Live stream and captures comments within 2 seconds.

3. Assumption: LLM-based Filipino comment parsing is accurate enough (>90% order detection) at viable cost (<$0.01/session). How to test: Collect 500 real live-selling comments from public streams, run through GPT-4o-mini with a tuned prompt, measure precision and recall.

4. Assumption: Sellers who start with one platform (Facebook Live) will connect Shopee/Lazada for inventory sync within 60 days. How to test: Track platform connections during beta; if <30% connect a second platform in 60 days, the inventory sync feature is less valuable than expected.

Risk flags

1. [Platform dependency]: Reliance on Facebook Live Graph API and TikTok API. Both can restrict access or change terms. Mitigant: screen-based comment capture as fallback, and being useful even without API access (manual order entry with smart suggestions).

2. [Low ACV]: $150/year average means high volume is required. A competitor with enterprise pricing (BigSeller, Ginee) could add live-selling features as a loss leader. Mitigant: transaction fee revenue layer, aggressive community lock-in.

3. [Copy risk]: BigSeller already has a free tier with inventory sync. If they add comment capture, LiveSold’s differentiation shrinks. Mitigant: move fast, own the Filipino-language NLP and joy-miner scoring, build community features that create switching costs.

14. Structured verdict

Score:                  78/100
Verdict:                GO
Confidence:             Medium
Best-fit builder:       Technical founder familiar with SEA market or Filipino co-founder for GTM
Time to revenue:        16-18 weeks
Capital to launch:      ₱15-25L ($2,700-$4,500)
Top 3 assumptions to validate first:
  1. Filipino live sellers will pay ₱499-999/mo for post-stream automation — survey 50 sellers in FB groups
  2. Facebook Live API provides reliable real-time comment access — build POC
  3. LLM-based Filipino comment parsing hits >90% accuracy at <$0.01/session — test on 500 real comments
Kill criteria:
  - Abandon if <20% of surveyed sellers express willingness to pay ₱499/mo
  - Abandon if Facebook Live API blocks comment access for third-party tools
  - Abandon if comment parsing accuracy stays below 80% after tuning

15. Next step — 1-week validation sprint

• Day 1-2: Join 5 large Filipino seller Facebook groups. Post a survey targeting live sellers: “How many hours do you spend matching orders after a live session? Would you pay ₱499/mo to automate it?” Aim for 50 responses. Simultaneously, build a quick POC: connect to Facebook Live Graph API, capture comments from a public live-selling stream, parse for order intent.
• Day 3-4: Analyze survey responses. If >40% say they’d pay, proceed. Record 3 public live-selling sessions and collect 500+ real comments. Run them through GPT-4o-mini with a tuned prompt. Measure: what % of “mine”/“order” comments does the model correctly identify? What’s the false positive rate?
• Day 5: Decide go/no-go based on: (a) ≥40% WTP signal from survey, (b) ≥85% comment parsing accuracy on real data, (c) Facebook Live API confirms reliable comment streaming. If all three pass, start building. If any fail, identify whether it’s fixable or fatal.