💃 Your Risk Register Is Lying to You 🕺

Good morning, Salesforce Nerds! I came across a bonkers stat the other day …

Roughly 70% of large transformation programs fail to meet their objectives. McKinsey says it. BCG says it. The number barely moves year over year. 📉

Here's the part nobody puts on the kickoff slide: technology is rarely the killer.

Forrester found governance breakdowns cause more than twice as many transformation failures as technical issues. 😨 

At the end of the day, your Salesforce transformation will not die because Apex failed you.

It will die because nobody owned decisions, nobody measured the right things, and risky scope sailed through unchallenged. 💸

The good news? Every one of those failure modes is preventable.

Let's build the prevention system. 🎯

TABLE OF CONTENTS

💃 Your Risk Register Is Lying to You 🕺

• Decision Debt Compounds Daily

• Beware the Watermelon Status Report

• Shrink the Blast Radius

• Migration Is Where Dreams Die

• Kill Risky Scope Before It Ships

INDECISION IS A DECISION

DECISION DEBT COMPOUNDS DAILY

Every deferred architecture decision accrues interest.

Punt on the data ownership model in month two, and you'll pay for it with a six-week integration rework in month nine. ⏳

This is decision debt. It's technical debt's better-dressed cousin, and it lives in steering committees.

The classic anti-pattern: a monthly governance board that batches every decision, big or small. Delivery teams idle while a slide deck waits for a calendar slot. 💀 

The fix is a decision framework, not a bigger committee. Classify decisions by reversibility and blast radius.

✅ One-way-door decisions (org strategy, data model, integration architecture) get executive scrutiny.

✅ Two-way-door decisions get delegated to the teams closest to the work.

Then write every significant decision down. A lightweight Architecture Decision Record beats a perfect memory and a pointed finger. 🧾

Executive accountability means an exec's name sits next to every one-way door. Not a committee. A name. 🚦

GREEN OUTSIDE, RED INSIDE

BEWARE THE WATERMELON STATUS REPORT

Transformation status reports have a fruit problem. Green on the outside, red on the inside. 🍉 I’ll explain …

Lagging indicators create watermelons.

"Percent complete" and "milestones hit" tell you where the program was, filtered through people incentivized to sound confident. 😅 

Leading indicators tell you where it's going. So, instrument your delivery pipeline and watch these instead:

👉️ Deployment frequency - how often changes actually reach a production-like environment. Shrinking frequency means integration risk is pooling behind the dam.

👉️ Defect escape rate - bugs found in UAT or production versus earlier stages. A rising rate means your quality gates are theater.

👉️ UAT burn-down slope - not the count, the trend. A flat line three weeks before cutover is a fire alarm, not a detail. 🔥

👉️ Environment drift - how far sandboxes have wandered from production config. Drift is the silent assassin of "it worked in the full sandbox."

Tools like Gearset and GitHub give you most of this telemetry for free.

The governance move is making these metrics the official status, so a red trendline can't hide behind a green slide. 📊

If a program can't produce leading indicators, that itself is your risk signal. 🚨

SMALLER BOOMS, FEWER TEARS

SHRINK THE BLAST RADIUS

Big-bang releases concentrate risk into a single terrifying weekend. Phased rollouts spread it thin enough to survive. 💥

Blast radius containment means designing every change so its worst-case failure stays small. Roll out by business unit, region, or persona. Gate new behavior behind Custom Metadata Type feature flags so you can switch off a feature without deploying a fix.

Reversibility is the companion discipline, and it's a design requirement, not a hope. Every major change ships with a tested rollback path. ⏪

Tested is the operative word. A rollback plan that's never been rehearsed is a bedtime story for the steering committee. 🛟

// Feature flag via Custom Metadata: reversible by config, not deploy
if (Transformation_Flag__mdt.getInstance('New_Quote_Engine').Is_Active__c) {
    new QuoteEngineV2().price(opp);
} else {
    new QuoteEngineV1().price(opp);
}

Salesforce's Well-Architected framework treats resilience as an explicit pillar for exactly this reason.

Systems that assume failure recover from it. Systems that assume success don't. 🧯

YOUR DATA IS LYING

MIGRATION IS WHERE DREAMS DIE

Data migration is the quiet killer of Salesforce transformations.

It looks like a task on the plan. It behaves like a program of its own. 🗄️

The failure pattern is always the same: teams discover data quality problems during cutover, not before. Duplicate accounts, orphaned records, and legacy fields that meant three different things across two decades. 😬

Prevention has three moving parts.

1️⃣ Profile early - run data quality analysis in week one, not week forty. Every surprise you find late costs ten times what it costs early. 🔍

2️⃣ Rehearse the cutover - full volume migration dry runs into a Full sandbox, timed and scripted, at least twice. The first rehearsal always fails. That's the point. 🧪

3️⃣ Reconcile with math, not vibes - define record counts, financial totals, and relationship integrity checks before migration. Sign-off requires the numbers to match, and a named executive owns that signature.

SUBTRACTION IS A STRATEGY

KILL RISKY SCOPE BEFORE IT SHIPS

Here's the uncomfortable truth about that 70% failure rate: most doomed programs were doomed at scoping.

The riskiest features were visible from day one. Nobody had the authority, or the stomach, to cut them. ✂️

Prevention-oriented risk management means scope carries a burden of proof. 🧑‍⚖️ 

Every major work item answers three questions: What breaks if this fails? Can we reverse it? Does the business value survive contact with that risk?

Scope that can't answer gets cut, deferred, or descoped into a flag-gated pilot. 🛑

This is a governance muscle, and executives have to be willing to use it. A transformation that ships 80% of its scope on time and intact beats one that promises 100% and delivers a watermelon.

Manage risk by refusing to carry it because the cheapest defect to recover from is the one you never built. 🧠 

SOUL FOOD

Today’s Principle

and now....Salesforce Memes