Skip to content

Playbook: custom offers

Цель: конверсия inbound inquiry на high-value sale через custom offer с upscaled scope.

Base принцип: buyer который пришёл на $500 гиг уже prequalified. Главный барьер (выбор seller) уже пройден. Теперь задача — увеличить scope + цену до точки где это действительно полезно для buyer’а и прибыльно для Upready.dev.

Target conversion rate: 20-30% inbound → custom offer accepted на $1 500-3 500.

Custom offer в Fiverr — отдельный mechanism отправки client’у специфического предложения с собственной ценой, scope, timeline. Не привязан к существующему гигу. Создаётся через Fiverr chat → “Custom Offer” button.

См. Термин: Custom Offers.

Шаг 1: read the inbound (первые 60 секунд)

Section titled “Шаг 1: read the inbound (первые 60 секунд)”

Что buyer действительно говорит

Section titled “Что buyer действительно говорит”

Первое сообщение от buyer — почти всегда understated. Они пишут “hi, I’m interested in your Claude gig”, но hиже спрятано:

  • Real scope — почти всегда больше чем базовый гиг
  • Real timeline — часто urgent
  • Real problem — конкретная pain point, не generic “hey help me”

Твоя задача — draw it out уточняющими вопросами.

Первое сообщение от seller должно содержать:

  1. Acknowledge их запрос — “Yes, I can help with that”
  2. Показать expertise — ссылка на relevant past work / proof
  3. Ask 2-3 specific questions — чтобы понять real scope

Template:

Hi [Name]! Yes, we can add Claude AI integration to your Next.js app.
We've shipped similar work — here's one example: [Upready.dev
case study URL or gig screenshot].
Before I send you a scoped offer, I need to understand a couple things:
1. Is this an existing Next.js app or starting fresh? If existing —
Next 14 or 15, App Router or Pages?
2. What's the main use case — chat assistant, code assistant,
document Q&A, agent with tool use?
3. Do you need streaming, memory, tool calling, or RAG over your
own documents?
4. Any deadline on your side?
Once I understand these — I can send you a custom offer with
exact scope and pricing.

Key: четыре уточняющих вопроса чтобы real scope выяснилось. Без этого — guessing, under-quoting или over-quoting.

Шаг 2: упаковать custom offer (5-15 мин)

Section titled “Шаг 2: упаковать custom offer (5-15 мин)”

Base gig scope (что входит в гиг как advertised) +

Upscale additions (что часто нужно но не включено):

  • Tool use / function calling
  • RAG over documents
  • Streaming
  • Error handling / retry logic
  • Observability (Langfuse, LangSmith, custom)
  • Type safety
  • Tests / eval suite
  • Documentation
  • Deployment help
  • Handover call
  • Post-delivery support window (7-14 days)

Tier A gig: $500 base → $1 500-2 500 upscaled offer Tier B gig: $1 500 base → $3 500-7 000 upscaled Tier C gig: $8 000 base → $12-15K upscaled

Justification scheme:

  • “Your use case involves [specific complexity]” (production app, not demo)
  • “Production-grade implementation includes [list]”
  • “For your [industry], [security/compliance] is critical”
  • “Timeline pressure means [better architecture upfront]“

Поля Fiverr Custom Offer form:

  1. Title: краткое описание (60 symbols max)
  2. Description (main): detailed scope, deliverables, approach
  3. Price: USD number
  4. Delivery time: days
  5. Revisions: 2-3 (not unlimited)
  6. Expiration: 7 days по умолчанию
Title: Production Claude integration for [their product name]
Description:
Custom scope for your [their specific situation] — production-grade
Claude AI integration in your Next.js app.
Deliverables:
- [Bullet 1] — e.g., "Claude chat widget with streaming SSE"
- [Bullet 2] — e.g., "Tool use: can query your Supabase via RPC"
- [Bullet 3] — e.g., "RAG over your /docs folder with vector search"
- [Bullet 4] — e.g., "TypeScript types + error handling"
- [Bullet 5] — e.g., "Monitoring via Langfuse"
- [Bullet 6] — e.g., "30-minute handover Zoom call"
- [Bullet 7] — e.g., "7-day post-delivery bug fix window"
Tech stack:
- Anthropic Claude API (claude-3-5-sonnet-20250101 or latest)
- Next.js 15 App Router
- TypeScript strict mode
- shadcn/ui for chat components
- Prompt caching optimization
Not included (priced separately if needed):
- Vector DB setup (if existing infra not suitable)
- UI/UX redesign beyond component styling
- Multi-language support
Delivery: 7 days
Revisions: 2 included
Price: $2,500

Критично:

  • Каждый bullet — конкретен, не generic
  • “Not included” секция защищает от scope creep (см. scope-creep playbook)
  • Tech stack показывает quality signal

Шаг 3: отправить + мониторить

Section titled “Шаг 3: отправить + мониторить”

Response time target: < 30 минут от inbound message до custom offer sent.

Почему критично:

  • Buyer ищет активно — у него множество tabs открыто
  • Первый response wins — даже если не самый дешёвый
  • Fiverr ranking учитывает response time (badge “Responds within 30 min”)
"Just sent you a custom offer. Key points:
- Scope: [2-line summary]
- Price: $X
- Timeline: Y days
If anything doesn't match your needs — tell me what to adjust and
I'll revise. Offer expires in 7 days."

Зачем: re-engage buyer, summary легче читать чем offer body, reduces friction to accept.

A. Accept immediately (10-15% случаев)

Section titled “A. Accept immediately (10-15% случаев)”

Buyer принимает сразу — start work.

B. Counter на lower price (30-40% случаев)

Section titled “B. Counter на lower price (30-40% случаев)”

Strategy 1 — hold firm:

"I understand, but the pricing reflects the production-grade scope we
discussed. If you need a smaller budget, I can reduce scope:
- $1 500: [less features, e.g., no RAG, no monitoring]
- $1 000: [minimal MVP, e.g., just chat widget with streaming]
Let me know which fits."

Strategy 2 — value add instead of discount:

"Instead of discount, I can include [extra deliverable] at same price:
- Free 30-day post-delivery support (worth $500)
- Extra revisions (3 instead of 2)
- Free 1-hour consultation after delivery
Which would help most?"

Never: просто снижать цену. Подрывает positioning и создаёт precedent для всех следующих clients.

C. Questions about scope (30% случаев)

Section titled “C. Questions about scope (30% случаев)”

Нормально. Answer specifically и offer to revise scope если needed. Не защищаться.

D. “Let me think” / no response (20-30% случаев)

Section titled “D. “Let me think” / no response (20-30% случаев)”

Follow up через 3 days:

"Hey, just checking in. Offer expires in [X] days. If you're still
evaluating, happy to discuss any questions or tweak the scope.
If not the right fit, no worries — just let me know so I know."

Second follow up через 5 days (ONE day before expiration):

"The offer expires tomorrow. Last chance — want me to adjust scope
or extend timeline? Otherwise I'll assume not a fit and we can
revisit later."

После expiration — let it go. Не re-engage если нет signal.

Normal — 20% buyers просто disappear. Не тратить energy.

Inbound: “Hi, I saw your $500 Claude gig. I have a small Next.js app and want to add a chatbot.”

Уточняющие вопросы → buyer отвечает:

  • Next 15 App Router
  • Support chatbot for their SaaS (e-commerce analytics)
  • Нужен context от user’s data в Supabase
  • Timeline: hard deadline в 2 weeks

Real scope:

  • Chat widget (advertised in $500 gig)
  • + Tool use to query Supabase (buyer’s data = tool call)
  • + Streaming for better UX
  • + Supabase RLS check в tool handler (security)
  • + Type safety + error handling
  • + Langfuse observability
  • + Post-delivery handover + 7-day fix window

Custom offer: $2 500, 10 days, 2 revisions.

Result (typical): ~25% acceptance rate at this price point, assuming good fit.

Inbound: “Need help with Supabase permissions, looks like RLS is broken.”

Уточняющие вопросы → buyer отвечает:

  • Production SaaS в active использовании (not demo)
  • 12 tables
  • Multi-tenant architecture (critical for their business)
  • Recently failed pen test

Real scope:

  • RLS audit (advertised in $400 gig)
  • + Full production rewrite (не просто report)
  • + Test suite for все permission paths
  • + 30-day monitoring setup
  • + Incident response plan
  • + Documentation для их team

Custom offer: $3 000, 14 days, 3 revisions.

После accept:

  1. Thank buyer в chat, confirm timeline
  2. Request access — GitHub, Supabase, Anthropic key, design files
  3. Set up project в internal Notion (tech team brief)
  4. Tech team starts work async
  5. Daily updates в Fiverr chat (“progress update — done X, working on Y, ETA Z”)
  6. Deliver через Fiverr submit (не email / GitHub private link без notification)
МетрикаTarget
Inbound inquiries на week3-5
Custom offers sent2-3 per week
Acceptance rate20-30%
Avg custom offer value$1 800-2 500
Orders from custom offers1-2 per month
Gross revenue$2 000-5 000

Это главный канал рокет-fuel для premium pricing без organic traffic.

Симптом: все inbound inquiries — “I have $50-100, can you do X”.

Диагностика: positioning в гиг description недостаточно premium, или attracting wrong audience через Tags / Keywords.

Fix:

  • Review gig titles — убрать anything sounding “cheap” (“quick”, “simple”, “basic”)
  • Remove tags типа “budget” / “cheap”
  • Highlight “enterprise” / “production” / “premium” в description
  • Raise Basic tier price (если $400 привлекает tire-kickers, set $800)

Симптом: 10+ custom offers sent, 0 accepted.

Possible causes:

  1. Price слишком high для market
  2. Scope unclear, buyers не понимают value
  3. Competition (buyer compared и выбрал cheaper)
  4. Buyer seriousness — просто info-seeking

Fix:

  1. Lower price только если pattern clear (не single data point)
  2. Better scope writing — bullets, specific tech, measurable outcomes
  3. Ask в reply почему не accept: “Just curious — was it price, scope, or something else? Helpful for us to know.”

Buyer просит multiple revisions custom offer

Section titled “Buyer просит multiple revisions custom offer”

Нормально первые 1-2 revisions (уточняем scope). После 3rd — красный флаг:

  • Buyer сам не знает что хочет
  • Или торгуется нечестно
  • Или tire-kicker

Protocol: на 3rd revision request — ответить:

"Happy to refine, но before я send another version — can we get на
15-минутный Zoom call чтобы align на exact scope? Это будет faster
than много revision cycles."

Если buyer declines call — likely not a real buyer. Move on.