Product Discovery for a B2B Social-Media Screening SaaS Platform

DBB Software ran a Product Discovery engagement that turned a manual, spreadsheet-driven service (~20 screenings/month, 3–5 days each) into a buildable plan for an automated, multi-tenant SaaS platform sized for 500+ screenings/month with a target turnaround of under 24 hours.

Industry

Technology

Service

Product Discovery

Team

DBB Software Architecture Team

Project State

Completed

Country

United Kingdom

About the Client

The client is a UK firm operating in the social media screening and digital reputation market that produces standardized, consent-based screening reports for B2B customers, including HR departments, compliance teams, and recruitment agencies.

The Client's Initial Request

The client engaged DBB Software for a Product Discovery: produce a System Design Document that the client could fund and build against while surfacing and resolving the biggest unknowns before development was committed.

Assess the Inherited Prototype

Determine whether the existing prototype could be salvaged or had to be rebuilt, with a clear, evidence-based verdict.

De-Risk the Unknowns First

Identify the assumptions that could risk the project and define a way to validate them before the build.

Translate the Manual Process Into Scoped Requirements

Convert a hands-on workflow into concrete functional and non-functional requirements with an explicit MVP vs. later split.

Resolve Build-vs-Buy

Evaluate and recommend a technology for every major component, with documented rationale and exit paths, including where to override the client's stated infrastructure preference and why.

Make It Executable

Provide a phased delivery structure, team composition, and a risk register with mitigations.

The Deliverables

DBB Software delivered a complete System Design Document that both specifies the platform and de-risks it before a line of production code.

Legacy-Prototype Reuse Assessment

Reviewed the inherited prototype across 7 components (architecture, API integration, scoring algorithm, keyword database, report template, infrastructure, data models) and decided to replace them. The prototype lacked the core business logic and used an incompatible stack. This replaced an open question with a defensible decision and removed false reuse expectations from later estimates.

A Pre-Build Research & Validation Phase

Designed a ~1-week research phase split into 3 blocks (API access validation, scoring-algorithm formalization, finalization with a go/no-go decision). The phase isolated the 2 risks most likely to sink the project (third-party data access and an un-formalized scoring algorithm) and surfaced 5 pending client deliverables (keyword spreadsheets, scoring documentation, sample reports, API credentials, brand assets) required before development could commit.

Scoped Requirements With an Honest MVP Split

Converted the manual workflow into 34 prioritized functional requirements traced to business requirements, with an explicit MVP-versus-later split and a documented deferral table flagging 5 items that need explicit client sign-off (CSV bulk upload, recurring monitoring, sponsor notifications, target re-entry, auto-invoicing) before scope was locked.

Evaluated Technology Stack & Architecture

Delivered a per-component evaluation across 6 build-vs-buy decisions (backend, frontend, database, authentication, cloud, AI/NLP), each with 3–4 alternatives compared and a documented exit path. The recommended stack is a single-language stack on a managed cloud in a UK data-residency region, structured as a 9-module modular monolith serving 3 user portals (customer, admin, target form) from a single SPA, with managed specialist services owning the highest-risk capabilities rather than building them.

GDPR-First, Multi-Tenant Design With a Resilient Pipeline

Specified a GDPR-first architecture because the platform handles personal data of screening subjects. The pipeline is designed for resilience, with retries, queue-based processing, a keyword-only fallback when the AI classifier is unavailable, and confidence thresholds that route uncertain classifications to human review. Delivered with 55+ measurable NFRs across 7 categories (performance, scalability, security, availability, observability, compatibility, email deliverability), 20 risks across 3 registers (technical, project, business) with mitigations, and 23 monitoring alerts specified for operations.

Results Achieved

A Buildable, Bounded Plan

Delivered a System Design Document the client can build from, with a prioritized MVP-versus-later split and a sign-off-flagged deferral table for change control.

The Make-or-Break Unknowns Resolved First

Assessed the inherited prototype, and isolated two assumptions most likely to sink the project into a pre-build research phase with go/no-go gates.

Compliance Built In

The platform handles personal data of screening subjects, so multi-tenant isolation, PII encryption, audit logging, and data-retention rules are part of the architecture itself.

Build-vs-Buy Resolved

Every major component has a recommended technology with a rationale, trade-offs, and a documented portability path, so the decision stays with the client.

Have a Product Idea but No Technical Plan Yet?

Our Product Discovery turns a vision into a clear plan to build it through scoped requirements, the right technology choices, a sound architecture, and a roadmap that surfaces and de-risks the make-or-break unknowns before they cost you.

Book a call