A content creator reviewing their performance

Product discovery

Validating the product before funding the build

A creator-tools startup needed to turn an AI coaching concept into a verified, fundable, and build-ready product plan. SDTC Digital delivered the specification, architecture, roadmap, prototype, and cost estimate through a ten-day discovery workshop.

Discovery workshop
10 days
Spec, architecture, plan, and cost
4 documents
Beta target
Around 3 months
Funding decisions
3 checkpoints

Project overview

GSD Creatives is developing an AI mobile app that acts as a personal social-media coach for content creators. Before committing to development, the company needed to verify platform assumptions, define the first release, and understand the investment required.

SDTC Digital ran a ten-day discovery workshop covering product scope, API capabilities, architecture, delivery, prototyping, and cost. Instagram and TikTok assumptions were checked against official developer documentation, revealing constraints before they became expensive development problems.

The final package gave the client a build-ready specification, an interactive prototype, and a funding plan designed to reach a demonstrable beta in approximately three months.

Scope
Instagram-first, TikTok companion
Phase 1 users
Individual creators
Phase 2 reserved
Agency console
Estimate form
Low / mid / high by workstream

Project details

Initial scope
Instagram-first, TikTok companion
Phase 1 users
Individual creators
Future phase
Agency console
Estimate structure
Low, mid, and high by workstream

The challenges

  1. 01The product needed to coach creators towards their goals rather than simply score them for brands.
  2. 02Assumptions about Instagram and TikTok capabilities needed verification against official API documentation.
  3. 03Multiple creator goals had to become one realistic plan rather than several competing workloads.
  4. 04The experience needed to avoid lost context, slow uploads, cluttered screens, and robotic interactions.
  5. 05AI-generated coaching had to remain grounded in verified creator data.

Project goals

The creator-tools startup needed an engagement that could:

  • Define a complete and buildable first release for individual creators.
  • Verify every platform-dependent feature before development.
  • Design a focused beta that could be demonstrated to investors within approximately three months.
  • Provide clear low, mid, and high cost ranges by workstream.
  • Separate deterministic calculations from AI-generated coaching.
  • Preserve a path towards a future agency product without rebuilding the foundation.

What we delivered

SDTC Digital converted the initial concept into a verified product package ready for fundraising, technical review, and development.

01

Functional specification and prototype

Features became user stories with acceptance criteria, mapped to interactive prototype screens.

02

Platform capability matrix

Instagram and TikTok capabilities were verified against official documentation before planning.

03

Technical architecture

The architecture defined mobile, backend, workers, database, integrations, and deterministic and AI layers.

04

Delivery roadmap and cost estimate

Six milestones structured the build around an early beta with role-based cost ranges.

05

Product roadmap and investor package

Future agency capabilities were separated from Phase 1, supported by investor-ready materials.

We chose them not just as a vendor but as a true partner to bring our concept to life. Their quality, work ethic, innovation, and client care have exceeded expectations - delivering on time, accommodating time zones, and providing the expertise to turn vision into reality. If you want a partner, not just a vendor, SDTC Digital is the right choice.
Frank CurtinCo-founder of Doc & Do

The User Journey

01

Connect creator accounts

The creator links supported social accounts and selects the outcomes they want to pursue.

02

Review current performance

The platform analyses available metrics and establishes a data-backed starting point.

03

Build the weekly plan

The strategy orchestrator combines multiple goals into one achievable set of actions.

04

Receive ongoing coaching

The app explains performance, recommends next steps, and updates guidance as the creator posts.

05

Track progress

The creator reviews completed actions, changing metrics, and progress towards selected goals.

The Thinking Behind the Product

  1. 01Verified platform access: Checked feature assumptions against Instagram and TikTok documentation.
  2. 02Code for calculations: Kept metrics, scores, and planning logic outside the language model.
  3. 03AI for coaching: Used AI to explain verified data and deliver personalised guidance.
  4. 04One coordinated plan: Combined multiple creator goals into actions that could support more than one outcome.
  5. 05Evidence-backed responses: Designed validation controls to reject coaching claims unsupported by available data.
  6. 06Protected AI workflows: Treated uploaded content and retrieved information as untrusted data rather than instructions.

Technology Stack

Specified during discovery. This engagement produced the architecture and the plan to build it.

Mobile application

  • React Native
  • Expo
  • TypeScript
  • pnpm
  • Turborepo

Backend and data

  • Node.js
  • NestJS
  • PostgreSQL
  • pgvector
  • Redis
  • BullMQ

AI and integrations

  • Anthropic Claude
  • Instagram Graph API
  • TikTok Login Kit
  • TikTok Display API
  • RevenueCat
  • KMS secrets

Testing and Delivery

10 days

Intensive discovery workshop covering product scope, platform capabilities, architecture, prototyping, delivery planning, and cost.

4

Complete package containing the functional specification, technical specification, delivery planner, and project estimate.

6 milestones

Phase 1 roadmap designed to reach a demonstrable, fundraising-ready beta in approximately three months.

3

Costed funding decisions covering the beta wedge, complete Phase 1, and low, mid, and high investment ranges.

Outcomes and Business Impact

  • A verified product concept, backed by four complete documents, specification, architecture, roadmap, and cost, ready for fundraising and development.
  • Unsupported Instagram and TikTok API assumptions were checked against real developer documentation and found wrong before engineering began; catching this in a 10-day workshop instead of mid-build was the difference between a pivot and a write-off.
  • A demonstrable beta was prioritised ahead of the complete Phase 1 build, engineered to land in three months rather than a six-month rebuild.
  • Transparent costs organised by role, workstream, and funding stage.
  • Clear separation between the creator product and the future agency platform, so Phase 1 decisions don't box in what comes next.
  • A complete package the client could use with investors or another development partner, giving them something to evaluate rather than a pitch deck promise.
  • The client credited the scope cut directly: "They told us what not to build. That saved us a quarter."
The interactive prototype
Weekly plan concept
Delivery plan and estimate

More work

Every engagement
Two colleagues in suits smiling at a phone screen by a sunlit office window

Transport

Turning every call into a booking.

Phone and web bookings answered by AI, written straight into the dispatch system, then rebuilt as a product other operators can buy.

A pharmacist reaching for a medication box on a dispensary shelf

Healthcare

Answering pharmacy calls without adding headcount.

An AI phone receptionist covering twenty stores around the clock, each with its own knowledge base and its data walled off from the rest.

A clinician on the phone at her laptop, medication jars on the desk beside her

Healthtech

Building an e-prescription platform around trust, access, and scale.

Clinic management and electronic prescribing with eIDAS-compliant signatures, and a pharmacy lookup that needs no account at all.

The cheapest thing you can do is check first.

Ten days of discovery routinely finds the assumption that would have cost a quarter. We run these as fixed-scope engagements.