Case studies

Systems built for real businesses, with the numbers that moved.

14 case studies from 210+ clients in 12+ countries. Each one names the client type, the stack and the number that moved, then links to the full write-up: the problem, what we built and what changed.

Book a free consult Scan the results table

Results ledger

Every case, in one table.

The award winner first, then the strongest results. Each row names the client type, the system, the stack and the number that moved, and links to the full write-up.

No.ClientSystemStackResult
012026US used-book retailer, 4 storesE-commerce / BooksDaily SEO Publishing for Shopify: 3 Scenarios, No Duplicates1st place, Make Scenario of the MonthSeptember 2026
  • Make.com
  • OpenAI
  • Shopify
  • Google Sheets
119 articles indexed, first run took articles with no shop links from 94% to 0%
022025State-Contracted Transportation ProviderTransportation Invoice Automation: 30 Days → 7 Days
  • Make.com
  • Retool
  • GoHighLevel
  • Claude
Invoice processing time down 77%, 30 days to 7, 31 hours a week back
032026Property-law specialist, AustraliaLegalAI Contract Review SaaS: 1,000 Pages, 94.9% Accuracy
  • Next.js
  • Supabase
  • Make.com
  • OpenAI
  • Stripe
  • Vercel
94.9% best-run accuracy on contracts of 100 to 1,000 pages, every fact page-cited
042026Faceless astrology account on ThreadsCreator / ContentCreator Content Engine: 5M+ Threads Views on Claude, Notion and n8n
  • n8n
  • Claude
  • Notion
  • Playwright
  • Threads API
5M+ Threads views, 1.15M in the first 27 days, followers 120 to 615
052024DistributorRetail and Supply ChainAI Vendor Data Extraction
  • Make.com
  • OpenAI
  • Telegram
  • Airtable
  • Shopify
10+ hours a week saved, error rate under 1%
062025US professional services firmError-Tolerant Document Generation: 50+ a Day
  • Make.com
  • Airtable
50+ documents a day, 75% of a bad batch still ships on time
072026Elevator and field-service company, USAElevatorsmonday.com ERP Audit and Rebuild: 25 Automations
  • monday.com
  • Make.com
  • Slack
0 live automations found by the audit, 25 delivered across 9 boards
082026Audiovisual event services, USEvents / Local ServicesYelp AI Lead Agent on Zapier and Claude
  • Zapier
  • Claude
First reply in minutes, follow-up at 12 hours, owner takeover in one click
092024Premium Travel BrandMulti-Store Sales Intelligence
  • Make.com
  • BigQuery
  • Google Sheets
Weekly and monthly reports, 1 to 2 days of manual work per report to 0, 100K to 250K records a cycle
102024Alok DaiyaFinance / TradingTrading Signal Automation
  • Make.com
  • TradingView
  • Telegram
Alerts delivered in under 2 seconds, from 2 to 5 minutes by hand
112024AstroVedansh, IndiaAstrology ReportsAstrology Report Studio: Order to Delivery on Make.com
  • Make.com
  • Razorpay
  • Google Sheets
  • Google Slides
  • Flutter
7 automations around a 198-module core, 0 delivery failures at scale
122024Apexal CollabCrypto Intelligence Hub
  • Make.com
  • CoinMarketCap API
  • Google Sheets
  • Social Scrapers
New listings scanned every 6 hours, 1,000+ tokens, 98% accuracy
132026Editorial teamMedia / EditorialFive-Stage AI Article Pipeline with Human Gates
  • Make.com
  • Airtable
  • OpenAI
  • Claude
  • Slack
Per article 6 to 9 hours to 60 to 90 minutes (client-reported), one stage re-run, not the article
142026Senior hiring expert, personal brandRecruitment / ContentAI Voice Reels Pipeline with ElevenLabs and Apollo for a Hiring Expert
  • ElevenLabs
  • Apollo
Reels without a recording session once the voice is cloned, time saved not yet measured

Every figure is the one its case study publishes. Where a number came from the client rather than our own logs, the row says client-reported.

Featured

Four builds, before and after.

Chosen for the size and clarity of the figure, and so each shows a different kind of build: an operations rebuild, a contract-review SaaS, a content engine and a Vision AI catalog intake, on four different stacks.

  1. 77%less invoice processing time

    Transportation Invoice Automation: 30 Days → 7 Days

    State-Contracted Transportation Provider

    • Make.com
    • Retool
    • GoHighLevel
    • Claude
    Problem
    Manual paper-based workflows delaying state reimbursements.
    Built
    A unified digital operations platform connecting GHL, Retool and NocodeBackend.
    Result
    Invoice processing time down 77%, 30 days to 7, 31 hours a week back
    Read the case study
  2. 94.9%best-run accuracy

    AI Contract Review SaaS: 1,000 Pages, 94.9% Accuracy

    Property-law specialist, Australia · Legal

    • Next.js
    • Supabase
    • Make.com
    • OpenAI
    • Stripe
    • Vercel
    Problem
    Lawyers cannot sign an AI summary that has no clause behind it.
    Built
    A multi-tenant SaaS with Make.com as the document-processing backend.
    Result
    94.9% best-run accuracy on contracts of 100 to 1,000 pages, every fact page-cited
    Read the case study
  3. 5M+Threads views, September 2026

    Creator Content Engine: 5M+ Threads Views on Claude, Notion and n8n

    Faceless astrology account on Threads · Creator / Content

    • n8n
    • Claude
    • Notion
    • Playwright
    • Threads API
    Problem
    No data on which format, hour or shape earns reach, and manual posting broke cadence.
    Built
    Discover, draft by format, schedule, publish, review weekly, kill what dies.
    Result
    5M+ Threads views, 1.15M in the first 27 days, followers 120 to 615
    Read the case study
  4. 10hmanual entry saved weekly

    AI Vendor Data Extraction

    Distributor · Retail and Supply Chain

    • Make.com
    • OpenAI
    • Telegram
    • Airtable
    • Shopify
    Problem
    Unstructured chat noise delaying catalog updates.
    Built
    An AI-powered intake engine that uses Vision AI to parse chat images and messages.
    Result
    10+ hours a week saved, error rate under 1%
    Read the case study

What clients say

Four clients, in their own words.

5 star average across 83 Fiverr and 14 Topmate reviews, with more on Upwork and LinkedIn.

  • “Prem is the best! great work will hire again and would refer 100/100!”

  • “He solved my use case in a way that I didn't think of that saved me a ton of money. I'll definitely work with him again.”

    Rob LevinConsulting. Review on Topmate.

  • “Straight up solved my workflow instantly. I explained in layman terms and he executed perfectly. I have worked with Prem 3 times now. Basically whenever I open Make I just hire him, it's not worth wasting my time clicking around.”

    Jack CraigeonRepeat client, 3 projects. Review on Topmate.

  • “The system turned messy contract documents into clean review-ready summaries while keeping final legal judgement with us. It saves hours before we even start the actual review.”

Check every claim before you book

Hire us for the same

The buyer page behind each case.

Every buyer page prices its scope in writing and lists the cases on this page that prove it.

Your process could be the next row.

A free 30-minute working session with Prem on Google Meet. Bring the process as it runs today. You leave knowing whether it is worth automating, on which platform and roughly what it would cost, whoever builds it.

Book a free consult Send the fit-check instead

Not booking today

Answer your own questions first.