{ } AI View What is this?
Behind the Numbers

Systems I've built.

Every stat on this site comes from somewhere. Here's a look at the internal tools and frameworks I've created to track, measure, and automate my work.

mkdm agent

>

Conversations may be logged to improve service. Privacy policy

I believe in building systems, not just doing tasks. Over the years, I've developed a suite of internal tools that help me work more efficiently and give clients complete transparency into results.

These aren't off-the-shelf solutions. Each system was built to solve a specific problem, whether that's tracking every deliverable across clients, automating content production, or ensuring nothing falls through the cracks.

01

WorkTrack Connect

The project database that powers every stat you see on this site.

Since February 2025, I've been meticulously tracking every project, deliverable, and outcome in a centralized database. Action items, recap threads, and client work are all cross-referenced and searchable.

But I didn't stop at February. I retroactively populated the database by scraping my Google Drive and email archives, so historical project data going back years is now organized and queryable.

When you see stats like "50+ clients served" or specific ROI figures, they're pulled directly from this system. Not estimates. Not guesses. Actual tracked data.

Client WorkGoogle DriveEmail ArchivesWorkTrackConnectCross-ReferenceAction ItemsRecap ThreadsWebsite Stats50+ clients101% ROI

Data flows from multiple sources into a single queryable database

100% Projects tracked since Feb 2025
2x Weekly automated updates
Years Of retroactive data recovered
02

Blog Production System

AI-assisted content production with human strategy at the core.

BPSv1 transforms keyword research into published, indexed blog posts through a 15-phase automated pipeline. It integrates with Notion, WordPress, SEMrush, Google Gemini, and Google Search Console.

But here's what makes it different: AI handles the heavy lifting (research aggregation, draft generation, image creation) while I maintain editorial control over strategy, voice, and quality. Every piece gets human review before it goes live.

The system includes hub-spoke architecture for topic clusters, automatic internal linking, entity optimization, and quality gates that block publishing until standards are met.

01-02Research+ Setup03-04Content+ GEO Opt05-07ImagesAI + Upload08-10QA ReviewHuman Check11-14Publish+ IndexLivePerplexitySEMrushGeminiManualWordPress+ GSC

15-phase pipeline with human QA gate before publishing

15 Phases from keyword to indexed
7+ Active client integrations
Human Strategy and editorial control
03

Daily Planning

Automated morning briefings that aggregate tasks from everywhere.

Every morning at 5 AM, this system pulls tasks and communications from Slack, Motion, Gmail, Google Calendar, and Microsoft Teams. It categorizes everything by priority (client servicing, sales, business development) and generates a comprehensive daily plan.

No more checking six different apps to figure out what needs attention. One document, every morning, with everything organized and prioritized.

SlackMotionGmailCalendarTeams5 AMAggregatorSmart PrioritizationCategorizationDaily PlanP1Client ServicingP2SalesP3Business DevP4InternalP5Admin.md fileReady by 5:01 AM

6 platforms converge into one prioritized daily plan

6+ Platform integrations
5 AM Daily automated run
Smart Priority categorization
04

TechSEO Autopilot

Automated technical SEO monitoring across all clients.

When SEMrush site audits arrive via email, this system automatically processes them, extracting issues, creating action items in WorkTrack Connect, and generating implementation instructions for each finding.

It uses a hub-and-spoke architecture: one central system handles all the logic, while client-specific configurations keep data organized. Issues are tracked through resolution and automatically logged in recap threads.

SEMrush Auditvia EmailAutopilotExtract IssuesWorkTrackAction ItemsResolutionTrackingRecapLogHub-and-spoke: one system, multiple clients

Email-triggered processing through resolution

Auto Email-triggered processing
4+ Clients monitored
Full Issue-to-resolution tracking
05

Reminder System

Simple cross-device reminders that actually sync.

A lightweight tool for creating reminders that sync across all Apple devices via iCloud. Supports recurring reminders, date shortcuts ("+3" for three days from today), and multiple reminder lists.

Sometimes the best systems are the simplest ones. This one just works.

./reminder"Task" +3 10:00iCloudSyncMaciPhoneiPadWatch

One command, all devices

All Apple devices synced
+N Date shortcuts supported
Recurring Daily, weekly, monthly, yearly
06

Monthly Reports

5-phase workflow for client performance reporting.

Client reports go through a structured pipeline: data collection from Google Ads, GA4, Search Console, and SEMrush; analysis and interpretation; narrative drafting; personal touch and strategic context; and finally, publication to Notion.

Each phase produces artifacts in the filesystem, allowing for review and refinement before the final report reaches the client. The system integrates directly with Google Ads API and Notion for seamless data flow.

Phase 1Data CollectionGA4, Ads, GSCPhase 2AnalysisROI, TrendsPhase 3DraftingNarrativePhase 4Personal TouchMatt's VoicePhase 5Publish→ Notion1_collected_data.md2_analysis.md3_draft_report.md4_final_report.mdNotion Page90-120 min per client

Each phase produces reviewable artifacts

5 Phases per report
90-120 Minutes per client
Direct API integrations
07

Why I build

I build these systems because I believe in transparency and precision. When I tell a client their campaign generated 101% ROI, I want that number to come from tracked data, not a spreadsheet I put together at the end of the month.

These tools also free up time for the work that matters (strategy, creative thinking, and actually talking to clients) instead of getting buried in administrative overhead.

If you're curious about how any of these work, or if you'd like to see the data behind a specific stat, just ask. I'm proud of what I've built.