SiriusOne:

Turning Competitor Ads into AI-Powered Market Intelligence

How Adlensa went from a single-channel prototype to a production-ready competitor ad monitoring SaaS covering Meta, Google, LinkedIn and TikTok — with a working MVP in 7 days and commercial launch readiness in 10 weeks.

Client & Project Overview

Adlensa is a competitor ad monitoring platform for marketing teams, agencies and founders. A user enters their company website, and Adlensa identifies the competitors that matter, brings together their ads from the Meta Ad Library, Google Ads Transparency Center, LinkedIn Ad Library and TikTok Ad Library, and uses AI to explain what those competitors are saying, which creative formats they rely on and how their messaging changes over time.

Business Problem

Competitor advertising research is still largely manual. Every major ad platform publishes its ads in a separate transparency library with its own search logic, filters and formats. Marketers open four tabs, copy ads into spreadsheets and try to spot patterns by hand. For Adlensa’s target customers, a single competitor analysis typically took around eight hours of analyst time — and was out of date within weeks.

“We knew exactly what we wanted Adlensa to be and who it was for. What we needed was a team that could turn those decisions into production software at startup speed — and tell us honestly when a shortcut would cost us later. SiriusOne did exactly that.”

Daniil

Chief Product Officer & Co-Founder, Adlensa

Tech Stack

AI & intelligence: Anthropic Claude models for competitor research, advertising analysis and conversational insights

Data sources: Official public ad transparency libraries: Meta Ad Library, Google Ads Transparency Center, LinkedIn Ad Library, TikTok Ad Library

Backend: Python API services, relational data storage, caching and asynchronous processing

Frontend: React web application with a responsive product interface

Cloud: AWS: managed compute, storage, networking, monitoring and security services

SaaS capabilities: Authentication, team workspaces, Stripe billing, usage controls, exports, transactional email

Engineering & delivery: Infrastructure as code, automated testing, CI/CD, AI-enabled SDLC

#AdTech

#SaaS

#GenerativeAI

#AISDLC

#AWS

Turning Competitor Ads into AI-Powered Market Intelligence - slide 1

Turning Competitor Ads into AI-Powered Market Intelligence - slide 2

Turning Competitor Ads into AI-Powered Market Intelligence - slide 3

Project Timeline

The first working MVP was ready in week one. Over the next nine weeks, the team iterated it into a production-ready, multi-tenant SaaS — ten weeks end to end.

Experts Lead. AI Accelerates. The speed of this project came from SiriusOne’s AI-enabled software development lifecycle — plan, build, test, review, release, operate. AI tools accelerated design exploration, code scaffolding, test generation and code review. Every change still passed automated quality checks and review by an engineer before it reached production, and releases were gated by the Product Owner’s acceptance.

That combination is what allowed a team of five to ship a usable MVP in seven days and a production SaaS in ten weeks, without trading quality for speed.

Duration

10 weeks

Effort

~1,100 hours

From prototype to MVP

Week 1

A working end-to-end product on a real user flow: company website in, competitors and their ads out.

Core intelligence experience

Weeks 2–5

Expanded coverage to all four ad platforms, improved result relevance, and introduced competitor discovery and AI-generated insights.

Production SaaS foundation

Weeks 6–8

AWS infrastructure, secure authentication, team workspaces and separate development, staging and production environments.

Commercial launch readiness

Weeks 9–10

Plans and billing, usage limits, exports, onboarding, monitoring and operational safeguards.

Team involved

Tech Lead / AI Engineer

Technical architecture, AI capabilities, backend engineering and the engineering delivery approach.

UI/UX Designer

Product experience, prototypes, design system and commercial user journeys.

Frontend Developer

The customer-facing web application: research, insights and account-management workflows.

DevOps Engineer

Cloud infrastructure, deployment pipeline and production environments.

Project Manager

Delivery planning and release coordination with the Product Owner.

Solution Overview

SiriusOne turned Adlensa into a working SaaS platform that replaces four disconnected ad libraries and a spreadsheet with a single, structured research workspace.

How We Worked: The Client Owns the “What”, SiriusOne Owns the “How”. Many outsourced product builds stall for a simple reason: the development partner ends up deciding what to build. Engineers know how to build software well, but they rarely hold the market insight needed to decide which features matter, for whom and at what price.

In this engagement, product ownership stayed firmly with Adlensa. SiriusOne owned architecture, engineering, design execution and delivery — and brought options and trade-offs to the table rather than making product decisions on the client’s behalf.

In practice, this meant:

one prioritised backlog, owned by the Product Owner and reviewed every week;
regular demos of working software on a staging environment, with go/no-go decisions made by Adlensa;
product decisions recorded alongside technical ones, so the team always knew why a feature existed.

AI-assisted competitor discovery

Users start with their own website. Adlensa suggests relevant competitors and pre-fills their profiles on each ad platform, so research no longer depends on a manually prepared competitor list.

Multi-channel competitor ad monitoring

Ads from Meta, Google, LinkedIn and TikTok are collected in one place, so marketers can review competitor campaigns and creatives side by side instead of switching between libraries.

AI insights on messaging and creative

Powered by Anthropic’s Claude models, Adlensa summarises competitors’ positioning, offers, hooks and creative patterns, compares approaches across brands, and lets users ask follow-up questions in a conversational interface.

Production-Ready SaaS

Secure sign-in, team workspaces, usage-based plans with Stripe billing, exports on every plan, transactional email, monitoring and separate environments for development, staging and production.

Results

7 days

From validated prototype to a working MVP.

10 weeks

To a production-ready, multi-tenant SaaS, ready for commercial launch.

4 ad platforms

Meta, Google, LinkedIn and TikTok — in one competitor research workspace.

8 h → min

Competitor analysis reduced from around eight hours of manual work to minutes.

Similar

implemented cases:

View all cases

Similar

implemented cases:

AI-Powered Customer Support Automation

SiriusOne designed and deployed a multi-channel AI agent system combining voice AI, text automation, and a real-time agent copilot — reducing average handling time by 52%, increasing first-contact resolution to 89%, and enabling the existing team to process 3.4x more interactions per agent per day.

Tech Stack: Cloud: AWS, AI/ML: GPT-4o · RAG · Voice AI · intent classification, Voice: Twilio · WebRTC, Backend: Python · FastAPI · PostgreSQL · Redis, Security: PII masking · RBAC · audit logging