Blog

You Can Write a UTM Convention, but GA4 Never Reads Your Table

Because UTM values are free text: the table you wrote is read only by you, while GA4 reads the same string by its own rules. Google's published default channel group logic checks source against a list of 819 sites and checks medium against fixed regular expressions, so for one identical link your table says referral and GA4 counts it as Organic Search. The convention document is only half the job. The other half is pulling out the violations with a query on a schedule.

46 Events, 14 Parameters We Chose Not to Send: How to Design an Event Taxonomy

Business questions, not event names. Once the questions are settled the event list follows, and the real decision after that is whether each value gets sent as a parameter or left out and recovered later through a join. In a sheet covering 46 events, 14 parameters were marked as not sent, and the reasons split five ways.

Impressions Are Not the Number in Your Ad Report: Measuring Off-Site Exposure Channel by Channel

Exposure is not one thing. It splits into paid, organic search, owned media, and earned. Ad platforms report only the first, and the other three each need a different tool. On top of that the word impression means different things per platform, since Google counts when the ad is served and Meta counts when it is on screen.

A Search Console Only Shows You Your Own Half: Counting 380 Naver SERP Slots Over Four Weeks

Search Advisor does report impressions, clicks and CTR. But it lists search keywords only for the top 30 by clicks, does not provide the exposure ranking, and no console answers how competing domains divide those slots. So I opened the search results for 77 keywords every week and counted the slots one by one, and out of 380 slots exactly 4 were ours, all of them near the top.

Why Ad Clicks and Site Sessions Never Match: Tracking Links and MMPs Explained

GA4 starts counting at the moment someone arrives on your site. Anyone who clicked the link but disappeared before arriving never shows up in any report. To measure that stretch, the link has to stop somewhere you control instead of going straight to the destination, and that is exactly what tracking links and MMPs do.

Backlinks Don't Work on Naver: You Build Search Authority With Data

Naver doesn't use external backlinks as a ranking signal. Backlinks aren't in the Search Advisor guide at all, authority builds through source trust (C-Rank), content quality (D.I.A.), and UGC instead. A practical breakdown of six steps for finding topics with search data, plus the four GEO elements that get you cited by AI.

Finding the True Owner of Branded-Search Conversions with GA4 Attribution, When PPL and Viral Reach Get Harvested as Search

How to trace the true source of branded-search conversions using GA4 attribution. Covers the structural limits of first-touch tracking and the dark funnel, and a ledger-versus-estimate framework for measuring PPL effects.

Getting Started with GA4: The 5 Reports to Check First and 5 Beginner Traps

A guide for anyone who feels lost after installing GA4 and doesn't know where to look. Covers the 5 reports to open first, from Realtime to Traffic acquisition and User acquisition, and the 5 traps beginners fall into most often.

Getting Started with GA4 Attribution: First Touch vs. Last Touch, Which Do You Trust?

A beginner's guide to GA4 attribution. Why user scope and session scope show different #1 channels, the four retired models and the data-driven attribution (DDA) that remains, the key event lookback window (30/90 days), and how to read first touch and last touch together.

Getting Started with GTM: Managing Tags with Google Tag Manager, No Code Required

How to manage tags with Google Tag Manager (GTM) without touching code. Covers the container, tag, trigger, and variable concepts, dataLayer, preview and version control, and the four mistakes beginners make most often.

How to Stop Losing UTMs: A Pixel-Free Ad Tracking Design Found on Claude's Site

A live test of why UTM parameters and gclid get lost in cross-domain conversions, and how Claude's site (claude.com) designed pixel-free ad tracking. Includes results from a Naver click ID (NaPm) experiment.

UTM Parameters, Fully Explained: How to Tag Your Links and 5 Mistakes Everyone Makes

A UTM parameter is a set of five tags appended to a link that tells an analytics tool like GA4 which channel, medium, and campaign a visitor came through. From the 3 required fields (utm_source, utm_medium, utm_campaign) to 5 common mistakes, all in one guide.

Measuring Ad Effectiveness: Reclaiming Channel Attribution with Search Volume and Time Series, Lift and Adstock Made Simple

How to measure ad channel effectiveness without individual tracking. Covers why a 0.15 correlation was an illusion, the baseline, lift, and adstock (advertising carryover effect) concepts, and the time-series data you need to start stacking today.

The Complete Guide to the Dark Funnel: 6 Ways to Trace the Customer Journey Hidden Inside "Direct Traffic"

What the dark funnel is, why so much traffic ends up logged as "direct," and 6 ways to track it, from UTM discipline to surveys, brand search volume, and experiments, laid out as an explainer guide.

Designing Marketing Measurement: Finding the Gaps in Your Company's Data with a Five-Layer Audit

How to self-diagnose where your company's marketing data is missing, using five layers (supply, demand expression, acquisition, conversion ledger, confirmation) and 15 questions. Includes a map of the whole series onto the five layers.

Google Search Console and Naver DataLab: How to See Search Data for Free

For free, you can see search interest (a relative value) in Naver DataLab, while actual monthly search volume (an absolute value) is checked in the Naver Search Ads Keyword Tool. Covers the principles behind Search Console, Search Advisor, DataLab, and Google Trends.

How Conversion Tracking Actually Works: What the Pixel, Click ID, and Server-Side Conversion API Each Do

The Meta Pixel is JavaScript that observes visitor behavior and sends it to ad platform servers. Click IDs like gclid and fbclid match a click to a later conversion, and server-side methods like the Conversions API bypass the browser to recover conversions cookies would otherwise miss.

Cookies, Sessions, Events: The Complete Basics: What GA4's Numbers Are Actually Counting

A GA4 session is a bundle of interactions cut off after 30 minutes of inactivity, and the user count isn't a count of people, it's a count of browser-level cookies (client_id). A ground-up rundown of the basics: cookies, sessions, events, and GA4.

The Complete Guide to Computer Specs for AI: From Cloud to Local LLMs (Windows & Mac)

What computer do you actually need to run AI and LLMs? Cloud AI like Claude and ChatGPT needs no graphics card, just 32GB of RAM. Running models locally is a different story: a usable 32B model needs 24GB of VRAM or 48GB of Mac unified memory, and 70B and up realistically needs 128GB on a Mac.

The Knowledge Warehouse That Organizes Itself When You Talk: Building an AI-Managed PKM Second Brain | How It Works & Setup Guide

A personal knowledge management (PKM) system where AI, not you, classifies, summarizes, links, and updates your notes. You view it in Obsidian; the AI works from one rules file. A guide from folder structure to a 30-minute setup.

I Thought My AI Was Learning Me, Until I Opened Five Weeks of Logs

Does bolting memory onto an AI make it smarter? Across five weeks of session logs, only 15% of skills and 4% of reference docs were ever reopened. 96% of the knowledge was a write-only grave. A hands-on log of redesigning AI memory to wire in resurfacing first.

Free Custom Domain Email: Complete Guide to Cloudflare Email Routing + Gmail

How to use a domain email like hello@mcpy.kr without a server or paid subscription. Step-by-step from Gabia nameserver delegation to Cloudflare Email Routing (MX, SPF, DKIM, DMARC), Gmail Send mail as, app passwords, and spam filters.

A Second Brain That Only Accumulates Is Already Dead: Building a Self-Evolving Knowledge System in a Day

Notes pile up but become useless because there is no circulation. I spent a day with AI grafting four circulation loops and automated health checks onto a knowledge vault, and the first checkup found 22 silently broken files. A hands-on build log.

When AI Can't Render Korean Text: Background from AI, Text from Code

AI image generators break Korean text, logos, and exact pixel dimensions. The fix: generate only the background scene with AI (Claude delegates to Codex imagegen), then layer Korean text and logos via HTML with web fonts, captured to PNG with a headless browser. A build log for automating blog illustrations and OG cards.

Fixing Invisible OG Images: From Keyword Research to Automated Social Cards

Why a polished card disappears in chat previews, the 9px math, gaze cueing, and contrast, plus a complete pipeline from keyword research and page titles to SERP checks, card production, and spreadsheet-driven operations.

Why the 1200px Card You Made Became an Unreadable Blob in KakaoTalk

Why a carefully crafted OG image turns into a blurry smear in KakaoTalk (Korea's dominant messenger app). The 9px math of canvas vs. render, how gaze-cueing works, URL cache invalidation, and why 7:1 WCAG contrast is the real minimum, all traced back to primary sources.

How Christianity Spread: Dissecting History's Most Successful Growth Marketing

A case study dissecting how early Christianity grew from roughly 1,000 followers to 35 million over 350 years, analyzed through nine growth marketing levers: addressable market, value, price, channel, promotion, segmentation, people, process, and growth engine. Historical figures are academic estimates.

The First Button of Any AI Marketing Agent: A Brand Brief, Not a Smarter Model

A brand brief is the single shared input for market-research, positioning, growth, and content agents. Without it, every agent fills in its own assumptions which then contradict one another. The secret to a good brief is honestly marking unknown fields as TBD rather than guessing. Includes a 10-section template and a real example.

Making Hundreds of Notes Fast for AI: I Tried Ontology, Then Found Something Lighter

Hundreds of notes piled up but AI still had to open every file to answer a simple question. I thought a full ontology would fix it. A lightweight index layer built from existing frontmatter turned out to work 2.9-13.3x cheaper. This is the build log.

I Logged In and Every User Property Was undefined: GTM and GA4 from Setup to Revenue Analysis

Can You Still Retarget Users Who Withdrew Marketing Consent: A Pixel Gating Design Story

How to Make Your Ads Learn Revenue: An Offline Conversion Tracking Case Study

Kakao Share Thumbnail Not Showing? Here's the Fix (JavaScript SDK 2025)

Learn why Kakao Share thumbnails don't appear even with a valid imageUrl, and fix it using scrapImage to convert your URL into a KAGE CDN URL. Step-by-step guide.

How to Generate CRM User Manuals in 20 Minutes with Claude Code Sub-Agents

Learn how Claude Code sub-agents can analyze your CRM codebase and auto-generate user manuals and visual guidebooks in just 20 minutes. See the real-world results.

GEO Generative Engine Optimization: The Complete 2025 AI Marketing Strategy Guide

Learn what GEO (Generative Engine Optimization) is, how it differs from traditional SEO, and discover 5 proven AI marketing strategies to get cited by AI search engines in 2025.