David D. Newell · Blog

Getting Started with Claude Code – for marketers!

Thu, 08 Jan 2026 00:00:00 GMT

Claude Code is Claude running in your terminal so it can work with files and repeatable workflows. This guide is written for marketers who are curious, perhaps a little apprehensive about technical tools used by software engineers, and want practical wins fast, without “learning to code.”

Terminal is a non-graphical way that you can tell your computer to do things. This guide helps you learn how to use this interface that is essentially keyboard-only. You can still use the mouse to select things, but all of the inputs are pasted or typed, then hit “enter” to run the command.

As we go through this guide, you’ll see text formatted like this which indicates something that you should copy or type into the Terminal, then hit enter to run it. Sometimes these commands will be broken out as a blockquote with one or more commands (each one will be on a separate line):

command1
command --option

Commands are instructions that you’re giving the computer. Hit enter to have the computer follow that instruction.

Let’s get started!

Bold Prediction: sometime in the course of 2026, this guide will become obsolete because a non-Terminal interface supporting work with local files will become available for non-technical users.

Please share your clever uses of Claude Code on social media using #MarketingWithClaude!

What you’re doing (and what you’re NOT doing)
Quick Start (~15 minutes)
Terminal Basics: just the 6 commands you need
Folder Structure
Context: how Claude stays oriented
How to give Claude your files
Model Context Protocol (MCP)
Debugging without tears
Marketing example prompts
The prompt formula that works

What you’re doing (and what you’re NOT doing)

You’re not “learning to code.” You’re learning:

A tiny bit of command-line actions (like learning a few phrases in a foreign language before traveling to a new country)
How to hand Claude your local files (like data exports from analytics tools, creative images, message content, etc.)
How to ask Claude to analyze, summarize, compare, and draft things for marketing work

Quick Start (~15 minutes)

Step 1) Open Terminal

Mac: Applications → Utilities → Terminal
Windows: Windows Terminal (or PowerShell)

Terminal is the old-school interface to interact with your computer. It’s often used for things that don’t have or need a graphical interface. In the Terminal, you’ll be using your keyboard almost exclusively.

Tip: when you’re typing a file or folder name in Terminal, you can hit Tab

Step 2) Create a working folder for Claude

In Terminal, type these commands (or copy/paste one at a time) and hit enter after each line:

mkdir -p ~/ClaudeMarketing
cd ~/ClaudeMarketing

You just created a folder and navigated to it! We need to know how to create folders (aka directories) and navigate through them so that we can have Claude run in that folder.

mkdir stands for “make directory”
cd stands for “change directory”

Note: you can also use your file explorer (like Finder on a Mac) to create folders, but you’ll still need to navigate to it in the terminal

Step 3) Install Claude Code

Follow the instructions in the Claude Code overview.

Step 4) Start Claude Code

Type or copy/paste, then hit enter:

claude

Follow the login prompts given. This command opens Claude in the Terminal, which, once signed in, is where you can enter prompts, just like using Claude on the web at claude.ai.

Step 5) Ask Claude something low-stakes

Try this:

Explain what you can do for a marketing team if I give you data exports from GA4 or Amplitude. Give me 10 examples.

You’re officially started. 🎉

Terminal Basics: just the 6 commands you need

If you only learn six commands, you can do 90% of what you need:

1) `pwd` — “What folder am I in now?”

pwd

pwd stands for “present working directory”

2) `ls` — “What’s in this folder?”

ls

ls stands for “list”

3) `cd` — “Go into this folder”

cd FolderName

cd stands for “change directory”

Tip: as you start to type the folder name, you can hit Tab to quickly fill in the rest of the folder name.

⚠️ Note: if a folder name has spaces, it needs to be “escaped” by adding \ before space, so the computer doesn’t confuse it as another option in the command. Using Tab to fill in the folder name, these escaping slashes will be added automatically.

To go up one level to the parent folder:

cd ..

One dot refers to the current folder, two dots to its parent. Use slashes in the folder structure to go further. The parent’s parent is ../..

4) `mkdir` — “Make a new folder”

mkdir CampaignQ1

5) `cp` — “Copy a file”

cp ~/Downloads/ga4_export.csv .

That last dot . means “copy into this current folder,” like mentioned above.

cp stands for “copy”

6) `mv` — “Move and/or rename a file”

mv ga4_export.csv ga4_260107_traffic.csv

mv ga4_export.csv inputs/ga4/ga4_260107_traffic.csv

mv stands for “move”

Don’t worry about memorizing these commands. Keep this section open the first few times and copy/paste the commands you need.

Folder Structure

Claude gets dramatically more helpful when your files are organized. Think of your folder as a project room full of evidence.

The marketer-friendly project folder template

Create one folder per initiative (campaign, experiment, research study). You can create these folders in Terminal or in your file explorer (Finder or Windows Explorer).

ClaudeMarketing/
  260107_growth-audit/
    brief/
    inputs/
      ga4/
      amplitude/
      mailchimp/
      snowflake/
    working/
    outputs/

Naming rules that save your sanity

Use dates as a prefix. I like YYMM, as it sorts nicely. Add the date for even more granularity, e.g., 260107
Include folders for the source of the contents within the folder: ga4, amplitude, mailchimp, snowflake
Include what is in the file: visits_by_source, email_performance, nps_comments

Examples:

260107_ga4_visits_by_source.csv
260107_amplitude_signup_funnel.csv
260107_mailchimp_email_performance.csv
260107_snowflake_nps_comments.csv

Context: how Claude stays oriented

Most “AI frustration” comes from missing context. Here’s the simple model to ensure you have all context provided for the task:

Role: what role should Claude play?
Objective: what are you trying to achieve?
Context: which files (exports, notes, results) matter?
Constraints: guardrails that should be considered
Output format: bullets, tables, slide summary, etc.
Definition of done: what means that Claude is done with the task?

The easiest context clue: create a `CLAUDE.md` file

Inside each project folder, create a file called CLAUDE.md and write down common context items that should be consistent in every task. Claude Code reads this file every time. The format is Markdown (which comes with the .md extension), which is a convention of formatting that is simple to read without changing font formatting.

Example CLAUDE.md for marketers

# Project: 260107 Growth Audit

## Context
Our tools:
- GA4
- Amplitude
- Pendo
- Marketo
- Databricks

### Key Moment Definitions
- Activation = completed signup + created first project
- Conversion = request demo OR purchase

## Output format
- Executive summary (5 bullets)
- Findings (by funnel stage)
- Recommendations (impact vs effort)
- What to test next (A/B ideas)

Why this matters: It reduces repetition and keeps Claude aligned across the whole project. You don’t have to input this context with every prompt, like you would otherwise.

How to give Claude your files

In Claude Code, you can reference files and folders directly using @ so Claude can read them without you copy/pasting everything. Claude will also smartly understand requests to review files without a direct reference. One useful indicator is to wrap a filename or folder name in with these characters: `. For example: `260107_growth_audit`.

Analyze a single file

Prompt: “Summarize channel performance and key changes month over month. @01_inputs/ga4/ga4_2026-01_channels.csv”

Compare two exports

Prompt: “Compare these two Amplitude funnels and explain what changed. @amplitude_funnel_dec.csv @amplitude_funnel_jan.csv”

Get oriented in a folder

Prompt: “What’s in @inputs/ and what should I analyze first?”

Model Context Protocol (MCP)

If files are “bringing Claude the receipts,” then MCP is like plugging Claude into your tools directly.

MCP is a standard for connecting Claude to external tools and data sources via “MCP servers.” You can think of it as a connector layer.

Three data access levels

Level 1 (we started here, without MCP): Export → Claude (CSV files, docs, notes)
Level 2 (Read-only MCP integrations): Claude can pull info from connected tools
Level 3 (Action MCP integrations): Claude can do things inside tools (with approval)

Tip: Start with exports (Level 1). Only consider integrations once your team agrees on safety rules and access permissions.

Debugging without tears

If Claude “can’t find a file”

Usually it’s one of these:

You’re not in the right folder (use pwd)
The filename/path is slightly different (use ls)
The file is in Downloads and not in your project folder yet

Try:

pwd
ls

Then copy the file into the right place:

cp ~/Downloads/yourfile.csv 01_inputs/ga4/

Note: you can also just use your file explorer to move files, if that’s easier!

Tip: ~ means the home directory, which on a Mac is /Users/<yourusername>.

If Claude asks for permission

This is normal and good. You stay in control. If it asks to read a file you intended to share, approve it. If not, deny its access.

If the results look wrong

Don’t “argue.” Tighten the request.

Ask Claude to list assumptions
Ask what information it used from the file
Reduce scope (“only analyze these 3 columns”)

One useful way to think about Claude and its capabilities is as a highly-educated, inexperienced intern. Give it the right amount of direction, and it will do a great job.

Tip: Don’t know what to do next? Claude can help you out…ask Claude what to do or for help!

Marketing example prompts

Here are some practical examples about how to get started using Claude Code for marketing.

Example 1: Experiment result evaluation (GA4 or Amplitude)

Export: experiment/variant performance or funnel conversion by variant

Put files in: inputs/ga4/ or inputs/amplitude/

Prompt:

# ROLE
You are a senior data analyst with experience evaluating marketing test results using Google Analytics and Amplitude.

# CONTEXT
- Data exports are in the folders `inputs/ga4/` and `inputs/amplitude`
- This experiment consisted of two groups: a control and variant
- The target KPI is overall funnel conversion, defined as [give a specific definition based on your data]
- Guardrail KPIs include cart abandonment rate ([give a specific definition based on your data]) and signup rate ([give a specific definition based on your data])

# OBJECTIVE
Evaluate the test results, declare a winner, and output executive-level summary results for this experiment.

# HARD CONSTRAINTS (NON-NEGOTIABLE)
- **Use only data in the provided export files.** Do not introduce external data not included in the provided files. Do not use placeholder data. If additional data are needed to accomplish the objective, define a specific data ask.
- Do not guess about the data. If you're unsure, **read files, inspect, and ask for clarification**.

# OUTPUT FORMAT
- Generate a powerpoint slide summarizing the outcome of the experiment. Include the primary and guardrail KPIs for each variant, an indicator of the winning variant, and commentary about learnings from the experiment and proposed next steps and next experiments to run.

# DEFINITION OF DONE
Only consider the work complete when:
- All data has been analyzed.
- All output documents are fully created and saved (all requirements accounted for).
- You provide a final summary including: what you analyzed, how you approached the analysis, and where to find the output.

Example 2: Copy pre-testing (subject lines, landing pages, ads)

Provide: copy variants in a doc or text file

Prompt examples:

Initial prompt

# ROLE
You are a senior marketer with experience in consumer behavior running a focus group with members of the target audience.

# CONTEXT
- Output copy will appear in email subject lines, on landing pages, and in online ads.
- Brand voice guidelines are provided in the file @brandvoice.txt
- The product value proposition is documented in @valueprop.docx
- Target audience personas are provided in @personas.md

# OBJECTIVE
Generate options for marketing copy to use to convince prospects to try and buy this product. Pre-test the copy options with each persona, make edits based on their feedback, and prioritize the options.

## PERSONA SUBAGENTS
Run each persona as a subagent that can respond to your questions and debate with other participants.

# HARD CONSTRAINTS (NON-NEGOTIABLE)
- Copy outputs must conform to the system constraints. Apply best practices for each surface.

# DEFINITION OF DONE
Only consider the work complete when:
- All copy options have been documented and saved to a file
- All best practice research questions have been answered
- Each persona has given feedback about each copy option
- The feedback and recommendations have been saved to a document

Follow-up prompt: Generate test variants

Prompt: “Based on the discussions with the target personas, create 12 subject line variants: 4 curiosity, 4 benefit, 4 urgency. Under 50 characters. Here’s the initial email draft: @email_draft.txt”
Follow-up prompt: Align with brand voice

Prompt: “Rewrite these headlines to match our brand voice rules. Keep meaning, improve punch.”

Example 3: Survey responses – consumer research

Data Export: CSV with open-ended responses, or a text file with one response per line

Prompt examples:

Theme clustering

Prompt: “Cluster responses into themes, estimate frequency, and give representative quotes. Then suggest messaging improvements. @survey_open_ended.csv”
Positioning angles

Prompt: “Propose 3 positioning angles and 3 message pillars based on what people value/complain about. Include proof points. @survey_open_ended.csv”
Objections

Prompt: “Extract top objections and anxieties. Draft landing page copy that addresses each. @survey_open_ended.csv”

Example 4: Campaign setup help — generate assets you paste into tools

Marketo nurture campaign kit (ready to paste)

Prompt: “Create a 4-email nurture sequence for [persona] who did [action] but didn’t [convert]. For each email: subject line, preview text, body copy, CTA, and goal metric. Keep voice [friendly/professional]. Also include send timing, segmentation rules in plain English, and what to A/B test in each email. Optimize the output for Mailchimp.”

Pendo in-app guide kit

Prompt: “Write in-app guide copy for a 3-step walkthrough introducing [feature]. Each step under 25 words, include a benefit, avoid jargon. Also write 2 tooltip variations per step.”

Cross-tool consistency

Prompt: “Ensure the Marketo nurture and the Pendo guide share the same message pillars and terminology. Create a shared mini style guide + phrase bank. @marketo_sequence.md @pendo_guide_copy.md”

The prompt formula that works

This structure provides Claude with all the information to maximize success. Treat it as guidelines because every situation is different. You can even ask Claude to improve your prompt, then use the output with minor edits to get an even better result!

Role: what role Claude should play for this working session
Context: what Claude needs to know about the situation, including any files you’re giving
Objective: what you’re trying to accomplish
Constraints: what should Claude be careful not to do?
Output format: what should Claude output? Be specific…Claude will follow instructions
Definition of done: what constitutes that Claude is done with this task in its entirety?

Copy/paste prompt template (we used it in example 1):

# ROLE


# CONTEXT
- 
- 

# OBJECTIVE


# HARD CONSTRAINTS (NON-NEGOTIABLE)
- 
- 

# DEFINITION OF DONE
Only consider the work complete when:
- 
-

Hope you found this guide helpful!

Please share your clever uses of Claude Code on social media using #MarketingWithClaude!

The KPI Prism: Decomposing Mix vs. Performance

Sun, 19 Oct 2025 00:00:00 GMT

In global SaaS, every KPI is a weighted average of the same metric across many sub-segments. If you don’t separate mix effects from performance changes, it’s easy to celebrate phantom wins or miss real degradation.

Every month, leadership gathers around dashboards of revenue, CAC, churn, and conversion. A line moves and the first question is: What drove the change? Too often, the answer reveals an analytical blind spot: improvements are merely favorable mix shifts; declines reflect changing customer composition, not operational failure. Sometimes that mix shift is even strategic.

This isn’t academic nitpicking. Misreading mix vs. performance leads to misallocated resources, poor strategic calls, and missed optimization, despite good intentions. If you want to make confident, data-driven decisions, you must decompose the change.

The Hidden Complexity Behind Every KPI

Why Simple Metrics Mask Complex Reality

Consider your monthly billing success rate, what I call the Stealthberg KPI. Suppose it drops ~40 bps month-over-month. Instinct says: inspect the processor, recent releases, or fraud rules. But the culprit may be a mix shift: more international customers (lower authorization rates) or a tilt from credit to debit. With a typical ~10-point credit-vs-debit authorization rate gap, a 4-point mix shift toward debit cards can move billing success ~40 bps without any true performance change.

This scenario illustrates the fundamental challenge: every KPI in a global business is actually a weighted average of multiple underlying segments, each with different baseline performance characteristics. When those segments shift in proportion, the “mix” effect, your overall KPI changes even when nothing about actual performance has changed (or has even moved the opposite direction).

The Business Impact of Analytical Confusion

The consequences of conflating mix and performance changes extend far beyond academic precision:

Resource Misallocation: Teams chase payment infrastructure fixes when the driver is geographic expansion, in line with the company’s growth strategy
False Confidence: Leaders applaud improvements driven by aging tenure mix, not durable gains activating customers to take the desired action
Missed Opportunities: Real improvements get masked by unfavorable mix shifts (Foreign Exchange rates can do this, too, whenever a currency-of-record is in use)
Strategic Misalignment: Business decisions are made based on partial stories, rather than the full performance picture (for retention KPIs, use the Fishbone Framework to help solve this challenge)

The Anatomy of Mix vs. Performance

Defining the Components

Before diving into methodology, it’s essential to understand what we mean by “mix” and “performance”:

Mix Effect: the change caused by shifts in the composition of segments (customer base, transaction type, region, etc.)
Performance Change: the change caused by segment performance moving up or down
Covariance: depending on the KPI, the dimensionality used for segmenting mix, and the scale of mix and performance changes period-over-period, there may be interaction when both mix and performance change at once (e.g., a growing, above-average segment also improves). It can amplify or offset the respective changes.

Identifying the Right Segmentation Dimensions

Not every dimension matters for every KPI. Start with those most predictive for the metric at hand or those that seem evident for your business. Don’t know what to start with? Ask around; you will probably get 3 or 4 common answers. Some of the common dimensions are:

Geography: Infrastructure, regulation, and behavior vary by region. For example, for billing success rate, expansion into Latin America could lower renewal authorization rates and increase involuntary churn, without operational regression.
Customer / Product Type: Enterprise customers often behave differently from small business or consumer segments. B2B customers might have higher billing success rates due to commercial credit cards and dedicated accounts payable processes, while consumer segments might show more volatility based on personal financial cycles.
Payment Method: For payments-related KPIs like the Stealthberg KPI (billing success rate), payment method differences create some of the largest mix effects. As mentioned previously, credit cards typically authorize at rates 8-12 percentage points higher than debit cards due to various factors in the payments ecosystem.
Tenure: Cohorts “shake out” over time. Older cohorts retain better simply because lower-propensity customers already churned. A couple of years of weak acquisition can raise average renewal rates (older base) and reduce total renewals—and eventually ARR—because there are fewer customers to renew.
Temporal Dimensions: Seasonality, day-of-week, or other macro cycles. Sometimes “seasonality” is a proxy for a latent customer attribute correlated with acquisition timing that is otherwise unobservable.

A Real-World Example: Decomposing Billing Success Rate Changes

The Scenario

Let’s examine a practical example using billing success rate data from a global SaaS business over two consecutive months:

January Performance:

Overall billing success rate: 92.0%
Credit card transactions: 83% of volume at 93.7% success rate
Debit card transactions: 17% of volume at 83.9% success rate

February Performance:

Overall billing success rate: 91.6% (↓0.4 percentage points, or ~40 bps)
Credit card transactions: 78% of volume at 93.8% success rate (↑10 bps performance)
Debit card transactions: 22% of volume at 84.0% success rate (↑10 bps performance)

The Surface-Level Analysis

A simple month-over-month comparison would show declining billing success rate, potentially triggering investigation into payment processor issues, fraud detection changes, or technical problems. The leadership team might panic about revenue impacts and demand immediate operational fixes.

The Mix vs. Performance Reality

However, decomposing this change reveals a completely different story:

Mix Effect: The shift from 83%/17% to 78%/22% credit/debit composition created a -49 basis point impact on overall billing success rate
Performance Effect: Actual improvements in both segments (+10 bps credit, +10 bps debit) created a +10 basis point positive impact
Net Change: -49 bps (mix) + 10 bps (performance) = -39 bps total change

Note: As we only have one dimension to decompose this example, there is no covariance.

The Strategic Insight

Rather than indicating payment infrastructure problems, this analysis reveals that operational performance actually improved across both payment methods. The overall decline resulted entirely from a shift toward debit cards, likely driven by customer acquisition strategies, regional expansion, seasonal patterns, or even customer choice.

This insight completely changes the business response. Instead of investigating payment technical issues, leadership could:

Understand the payment method mix shift drivers
Evaluate whether this shift aligns with strategic objectives or requires other changes in other parts of the customer experience
Consider targeted investments on further debit card authorization rate improvements
Celebrate the underlying performance improvements that are actually occurring

Effectively Communicate the KPI Change

When communicating the factors like the one highlighted in this example, waterfall charts are amongst the best data storytelling visual available that easily encodes the math into a format accessible for any audience.

Tip: When multiple mix/performance factors move, stack them into one or more waterfall steps to show compounding effects.

Want to dive deeper into the math?

The Mathematical Framework

For teams ready to implement mix vs. performance analysis, here’s the mathematical framework:

Notation:

KPI₀ = Overall KPI in period 0 (before)
KPI₁ = Overall KPI in period 1 (after)
w₀ᵢ = Weight (proportion) of segment i in period 0
w₁ᵢ = Weight (proportion) of segment i in period 1
p₀ᵢ = Performance of segment i in period 0
p₁ᵢ = Performance of segment i in period 1

Overall KPI Calculation: KPI = Σᵢ (wᵢ × pᵢ)

Mix Effect Calculation: Mix Effect = Σᵢ (w₁ᵢ – w₀ᵢ) × p₀ᵢ

Performance Change Calculation:
Performance Change = Σᵢ w₁ᵢ × (p₁ᵢ – p₀ᵢ)

Covariance Calculation: Covariance = Σᵢ (w₁ᵢ – w₀ᵢ) × (p₁ᵢ – p₀ᵢ)

Verification: KPI₁ – KPI₀ = Mix Effect + Performance Change + Covariance

Applications Across SaaS KPIs

The mix vs. performance KPI decomposition methodology applies powerfully to most KPIs in SaaS businesses:

The Fishbone Framework model of retention, which is a systematic approach for measuring operational factors affecting subscription renewal business performance, lends itself to decomposing the drivers of renewals in a SaaS business.
Revenue Growth Rate: Changes might reflect shifts in customer acquisition between high-value enterprise deals and smaller SMB accounts rather than actual expansion or contraction within customer segments.
Customer Acquisition Cost (CAC): Increases could result from targeting new geographic markets with higher acquisition costs rather than declining efficiency in existing channels.
Conversion Rates: Marketing channel mix, traffic source composition, and lead quality changes all create mix effects that can obscure genuine conversion optimization results.
The Stealthberg KPI in Detail: Billing success rate exemplifies why mix vs. performance analysis is critical. Consider these common mix factors that affect authorization rates:
- Credit vs. Debit Mix: A 4 percentage point shift toward debit cards typically moves billing success rate by approximately 40 basis points due to the structural ~10 percentage point difference in authorization rates between card types
- Geographic Mix: Expansion into emerging markets can reduce overall authorization rates by 2-5 percentage points even with perfect operational execution
- Customer Tenure Mix: Newer customers often have lower billing success rates due to outdated payment information and less established banking relationships
- Transaction Size Mix: Larger transaction amounts commonly face relatively higher scrutiny from fraud detection systems, creating mix effects if average order value changes

Without decomposing these mix effects, SaaS operators might chase technical optimizations or calculation artifacts when the real opportunity lies in understanding customer acquisition strategy or implementing targeted experiential improvements for specific segments.

Implementation Best Practices

Build the plumbing

Prioritize Dimensions: Not every dimension matters equally for every KPI. Identify the ones with the largest variance when decomposing the your specific KPIs. Add any additional ones you expect will cause variance as you execute your growth strategy.
Establish Baseline Reporting: Create dashboards that automatically decompose your critical KPIs into mix and performance components. This prevents the need for ad-hoc analysis every time metrics move unexpectedly.
Set Materiality Thresholds: Define minimum thresholds for mix effects worth investigating. For example, if geographic shifts create less than 10 basis points of impact, you might choose to focus analytical resources elsewhere, rather than digging in deep into why this mix shift occurred.
Create Decision Frameworks: Establish clear protocols for how different types of mix vs. performance insights should influence business decisions. Performance improvements might trigger scaling efforts, while unfavorable mix shifts might prompt strategic discussions about customer acquisition.

Communicate for action

Waterfall Charts: use this visuals first, keeping the math visible and intuitive, like shown above
Add Narrative Context: always accompany the decomposition with business context about what is driving observed changes. What changes did you make that affected performance changes? Bringing receipts helps spur the necessary conversations and investments.
Enable Action: present results at the level of organizational control. If your marketing team cannot influence geographic mix but can optimize channel performance, segment results accordingly. At minimum, emphasize what can and cannot be controlled.
Trend Analysis: Show historical trends of mix & performance so leaders can see whether changes are a signal or just noise.

Advanced Considerations & Common Pitfalls

Latent dimensions: If residual “performance” swings with patterns, you may be missing a segmentation variable (e.g., device, browser, payment flow). Add it to the mix effect.
Over-segmentation: Tiny segments create noise. Enforce significance thresholds and remove dimensions where needed.
Temporal smoothing: Rolling windows or continuous monitoring reduce period boundary artifacts.
Covariance: Usually small, but during big changes (e.g., rapid geo expansion and other business changes) it can be material—measure it explicitly.

Building Organizational Capability

Cross-Functional Collaboration

Mix vs. performance analysis requires collaboration across traditionally siloed functions:

Marketing and Growth Teams need to understand how customer acquisition strategy creates mix effects in downstream metrics like retention and billing success.
Operations Teams focus on performance effects but need visibility into mix changes that might mask or amplify their optimization efforts.
Finance and Analytics Teams typically own the analytical methodology but require domain expertise from operational teams to interpret results correctly.
Executive Leadership must understand both mix and performance implications when making strategic decisions about market expansion, customer targeting, and resource allocation.

Training and Change Management

Implementing mix vs. performance analysis often requires shifting organizational mindset from simple metric monitoring to sophisticated analytical thinking. As with most mindset changes, you will be most successful by changing actions and letting the mindset follow. Implement the analytical framework, explain business changes more clearly than ever before, and then watch how the rest of the organization lines up behind this methodology.

Integration with Broader Analytics

Mix vs. performance decomposition works most effectively when integrated with other analytical frameworks:

Cohort Analysis: Understanding how mix vs. performance effects vary across customer cohorts provides additional insight into long-term business health.
Attribution Modeling: Marketing attribution becomes more accurate when you can separate performance improvements from favorable audience mix shifts.
Forecasting Models: Incorporating expected mix changes into forecasting models improves prediction accuracy and scenario planning capability. It also helps with root cause analysis when the actual mix varies from expectation and to prevent panic when known mix changes are ahead.

The goal isn’t analytical perfection. It’s analytical sufficiency to make better business decisions with higher confidence and fewer surprises.

Conclusion: Clarity Through Decomposition

Every number tells a story, but without decomposition, it may be the wrong one. Mix vs. performance analysis turns reactive KPI watching into proactive understanding. You will know whether a dip in billing success demands a payment fix or a strategy review; whether a conversion bump merits budget or a closer look at audience mix; whether churn shifts signal satisfaction issues or shifting cohorts.

The investment is modest compared to the payoff: better decisions, smarter allocation, and fewer surprises. The clarity creates a competitive advantage. Whether you’re optimizing the Stealthberg KPI, applying the Fishbone Framework for retention, or simply trying to understand what’s really happening in your business, the principle holds: separate signal from noise, control from context, and never mistake correlation for causation.

Your KPIs are trying to tell you a story. Make sure you’re listening to the right narrative.

This post is part of a series on subscription business optimization. Start applying this framework to understand your mix vs. performance today!

For more insights on data-driven subscription business optimization, connect with me on LinkedIn.

Billing Success Rate—The Stealthberg KPI

Fri, 25 Jul 2025 00:00:00 GMT

The payments ecosystem is highly complex due to the linkage of many different companies that each independently have control mechanisms that affect authorization rates. In a global SaaS subscription business, the potential silent killer of ARR and retention rates is the billing success rate: how many customers are successfully auto-renewed. I call it the Stealthberg KPI, as it is typically among the top five most influential factors affecting revenue, and often the single most sensitive operational factor in the business as a whole, but far too few leadership teams pay close attention to measuring and investing in the control of billing success.

Like the massive ice formation that destroyed the “unsinkable” Titanic, your billing success rate lurks beneath the surface of your metrics dashboard, invisible to most operators, yet capable of sinking even the most promising SaaS business. While you’re focused on optimizing conversion funnels, reducing customer acquisition costs, and tracking monthly recurring revenue, the Stealthberg KPI is quietly causing your most loyal customers into involuntarily churn.

Why Billing Success Rate Is Your Most Critical Hidden Metric

The Foundation of Recurring Revenue

Subscription businesses know all too well the importance of retention to boost expected customer lifetime value and therefore have invested heavily in frictionless signup experiences that put customers’ subscriptions on auto-renewal. That means most recurring revenue and nearly the entire customer base falls into this journey that requires automatic payment to function correctly to smoothly ensure revenue collection.

That’s brilliant when it works, but when it doesn’t work right, what is the first place you look when you see a drop in revenue? Is it the trend of your auto-renewal payment success?

Many SaaS operators can tell you their monthly churn rate to two decimal places, their customer acquisition cost down to the dollar, and their conversion rates across every funnel stage. But ask them their current authorization rate for subscription renewals, and you’ll likely get a blank stare followed by, “I think that’s a payments team thing.”

The Invisible Revenue Leak

Here’s what makes billing failures so insidious: unlike voluntary churn, these customers want to stay subscribed. They have integrated your product into their workflow or lifestyle, they see value in your service, and they have no intention of canceling. Yet a declined payment transforms them from happy, renewing customers into involuntary churn statistics.

Consider the math: if you’re operating at a 90% renewal authorization rate—which many businesses would consider healthy—you’re losing 10% of your renewal revenue not to dissatisfaction or competitive pressure, but to payment processing failures. For a $10M ARR business, that’s $1M in annual revenue walking out the door through no fault of the customer experience, product quality, or market positioning.

The Ecosystem Complexity That Makes Control So Difficult

Payments Isn’t Just a System, it’s an Ecosystem

Like rainforests, there are complex interactions between all the different elements at play. Plants, animals, and insects interact in simple and complex ways, and when one element changes, it can have small or large ripple effects throughout the rest of the ecosystem. Understanding when these changes are affecting everyone or just you is incredibly difficult, as authorization rates are often held quite closely, as if they were a guarded secret.

In my experience leading payments strategy across multiple subscription businesses, I’ve learned that authorization rates fluctuate constantly due to factors completely outside your control:

Issuing bank policy changes: a bank updates their fraud detection algorithms
Network-level adjustments: Visa or Mastercard modifies merchant and transaction risk scoring models
Economic conditions: recession fears trigger more conservative approval patterns or smaller bank account balances
Seasonal patterns: Holiday spending affects risk thresholds (and also bank account balances)
Geographic events: Regional regulations or economic instability impact specific markets (it wasn’t long ago that US and EU merchants had to halt sales in Russia essentially overnight due to sanctions)

These complexities also make the overarching strategy for business improvements much more complicated. Not only do you have to account for changes within your own business and customer base, you also have to account for changes in the ecosystem.

During a recent industry discussion on payments experimentation, I noted:

Auth rates are continuously changing due to a variety of reasons. Simply put, the world changes quickly, and not all changes are within your control. It could be the network, the issuers, global economic movements…

At every point in time, you also have a different customer base renewing, so you never face the exact same situation twice.

How the Stealthberg KPI Can Sink Your Ship

The Vicious Downward Cycle

Having such a strong influence on renewal revenue and, by connection, customer lifetime value, billing success has the potential to start a vicious downward cycle that can spiral out of control if not monitored and managed carefully. The adage says that it’s far cheaper to keep a customer than to acquire a new one. Billing success is the path to passively renewing subscribed customers.

Unless you have a product with very high stickiness (think streaming services where non-payment blocks your ability to watch the next episode of your favorite show), overcoming the “activation energy” to motivate a customer to take action presents a significant challenge.

Here’s how the downard spiral unfolds:

Payment fails → Customer receives dunning notification
Customer ignores or misses notification → Service gets suspended after grace period
Customer experiences service disruption → Frustration builds with your brand
Manual intervention required → Customer service costs increase
Customer updates payment → But maybe downgrades or considers alternatives
Or customer churns entirely → Zero recovery of the residual lifetime value

That’s also why this business model favors the auto-renewal choice architecture from the beginning: it’s better for the business to auto-renew customers and allow them to take action to undo the renewal than to force them to manually renew their subscription every period. It’s also better for the consumer in many cases, as it ensures that there is no interruption of their service.

The Measurement Challenge: Mix vs. Performance

Why Your Auth Rate Moves (Even When Performance Doesn’t Actually Change)

One of the biggest obstacles to managing billing success is understanding whether changes in your authorization rate reflect actual performance improvements or simply shifts in your customer mix.

Mix factors are a major component: your customer base is constantly changing and using different payment methods and card types. Factors like customer tenure, geography, credit cards vs. debit cards vs. other payment methods, and more can cause your overall billing success to move simply because the weighted average changed, not because the performance of any single segment changed.

For example, if your business experiences rapid growth in international markets where authorization rates tend to be lower, your overall billing success rate might decline even if your performance in every individual market improved. Conversely, if you acquire more enterprise customers who predominantly use commercial credit cards (which often, but not always, have higher authorization rates), your metrics might improve without any operational changes.

I learned this lesson firsthand when analyzing what appeared to be declining performance, only to discover through deeper segmentation that our core performance was actually improving; we had simply acquired a larger proportion of customers using payment methods with historically lower authorization rates.

The key is implementing the KPI Prism, a measurement framework that can decompose your KPIs to separate signal from noise, allowing you to identify when changes reflect genuine performance shifts versus natural customer segment evolution. Such a framework also helps measure controlled changes through experimentation to improve billing success.

Building an Experimentation Framework for Payments

The Foundation: Measure and Control

Fundamentally, you can run an experiment anywhere in your business, as long as you can do two things: measure and control. You need to be able to make a change that does something differently, then measure the outcome.

The experimentation principles that drive performance marketing translate extremely well to payments. Both disciplines require understanding how different entities behave and react—whether it’s customers responding to ad creative or banks responding to technical payment parameters.

Establishing Your Baseline

The first step of experimentation is understanding your current state—the baseline against which you’ll measure improvement. When I joined my current role, I honestly didn’t know our baseline, and that’s half the battle. I started by gaining an understanding of what we could measure and control in each tool and system.

Something that is particularly challenging in payments is that some leaders think that payment KPIs will stay flat forever. As described above, billing success rates continuously change due to a variety of factors. When it comes to conducting experiments, you must reassess and re-evaluate your baseline each time. Even better, use A/B testing to have a control baseline to measure against.

Hypothesis Framework: Control Lever + Segment + Outcome

I like to think about experiments in terms of three components:

A control lever (what you’re changing)
A customer segment (who you’re testing with)
An outcome (what you’re measuring)

My mantra is: how can we maximize learning for minimum effort? That’s how you get the most out of experimentation.

For prioritizing experiments, I find the ICE framework most useful:

Impact: How much could this move the needle?
Confidence: How sure are we this will work?
Ease: How difficult is this to implement?

On my team, we crowdsource ratings from 1 to 5 for each factor, multiply them together, and sort our experiment backlog accordingly. Sometimes the simplest approaches are the most effective.

The Control Levers: Your Payment Optimization Toolkit

Internal vs. External Levers

There are both internal and external control levers, and they differ significantly between customer-initiated transactions (CITs) and merchant-initiated transactions (MITs).

For CITs, the customer is involved and engaged, so there’s more the customer can do to make the transaction go through.

For MITs, which include all your subscription renewals, it’s all about getting the bank to feel comfortable enough, statistically speaking, to authorize the transaction.

Multi-Acquirer Routing

One of the most powerful levers is the ability to route failed transactions to different processors. Not all acquirers have the same relationships with issuing banks, and what fails on one processor might succeed on another. This approach requires technical infrastructure to support real-time routing decisions, but the impact can be substantial.

Transaction Timing Optimization

The timing of both initial and retry billing attempts can dramatically affect authorization rates. Time of day, day of week, and day of month all play roles in boosting auth rates among certain segments.

During our experimentation, we discovered fascinating patterns. For instance, we found that certain customer segments in specific geographic regions had significantly higher authorization rates during seemingly random time windows—patterns that only emerged through systematic testing.

One experiment I was initially afraid to try was changing the time of day for our renewal batch job. Our hypothesis was that fraud rules differ throughout the day. After all, is someone really buying something in the middle of the night? While possible, transactions during unusual hours face more scrutiny.

We analyzed authorization rates for customer-initiated transactions by hour to identify peak performance windows, then tested those times for our automated renewals. The results validated our hypothesis and delivered measurable improvement.

Expiry Date Optimization

Account updater services are imperfect. Banks don’t always pass back new card information when cards are replaced, leaving you with outdated payment methods. However, trying different expiry dates in some cases can make a huge difference.

This technique requires careful implementation. Systematic testing of expiry date variations can recover revenue from cards that appear expired but may still be valid.

Network Tokenization

Network tokens issued by the card networks are typically seen as less risky by issuing banks because they represent validated, network-approved payment credentials. However, they’re not fully deployed across every card yet, as many banks have yet to fully enable their entire card portfolio.

When available, network tokens can provide meaningful authorization rate improvements, but adoption requires understanding which customer segments have access to tokenization and building systems to request and utilize these tokens effectively.

Debit Card Network Routing

Similar to multiple acquirers, if you can route debit card transactions across different networks (like PIN vs. signature debit), it can make a significant difference in authorization rates. Different networks have varying relationships with issuing banks and different risk assessment models.

Segmentation Strategy

For all these techniques, testing them with different customer segments can reveal powerful insights. Whether segmenting by business vs. consumer customers, various issuing banks, geographic regions, card types, or other dimensions, you must experiment across different aspects to determine what matters for your specific business.

The Dark Horse: Issuer Outreach

Beyond Technical Solutions

The technique I consider the dark horse in the world of payments optimization requires no technical or logic changes at all. Find the right person at the issuing bank with the ability and authority to adjust fraud rules. You might find authorization rate improvements just by having the bank make a change.

This approach requires relationship building and deep understanding of your transaction patterns, but the potential impact is enormous. When you can demonstrate to an issuing bank that your business represents legitimate recurring revenue with low dispute rates, many banks are willing to adjust their risk parameters for your merchant ID.

The key is presenting compelling data that shows:

Your historical dispute rates
Customer tenure and satisfaction metrics
Transaction pattern consistency
Geographic and demographic insights about your customer base
The aligned interests to support your mutual customer

Most payment optimization focuses on what you can control through technology. Issuer outreach recognizes that authorization decisions ultimately happen at the bank level, and sometimes the most effective path to improvement is direct communication rather than technical workarounds.

Creating a Culture of Continuous Improvement

Celebrating Learning, Not Just Wins

One critical aspect of building an effective experimentation program is managing expectations around success rates. If you’re just starting to experiment, you might see a high win rate, but after you’ve implemented several experiments, your success rate will probably approach 10-33% winners. That’s the right balance of risk and reward.

Nobody likes to lose, so as someone leading an experimentation effort, it’s critical to create a culture that celebrates the learning that comes from all experiments. You cannot celebrate only the winners.

As Thomas Edison reportedly said:

I didn’t fail; I found 2,000 ways not to make a light bulb…but I only needed to find one way to make it work.

The more experiments I conduct, the more surprising the results become, both in marketing and payments. It’s often the tests you least suspect that yield the greatest benefit.

Taking Action: Your Stealthberg Assessment

The first step in addressing any hidden threat is bringing it into the light. Most SaaS businesses have sophisticated dashboards tracking dozens of growth and retention metrics, but billing success rate rarely makes the cut.

Take the Stealthberg assessment survey to see where your business stands.

Conclusion: Navigating Safer Waters

The Stealthberg KPI represents more than just another metric to track. It’s a fundamental shift in how subscription businesses should think about retention and revenue predictability. While your competitors focus on visible threats like pricing pressure and feature differentiation, mastering billing success rate creates an invisible competitive advantage.

Every percentage point improvement in authorization rate flows directly to your bottom line without requiring customer acquisition spend or product development resources. More importantly, it preserves the customer relationships you’ve worked so hard to build, ensuring that payment infrastructure never becomes the reason a satisfied customer churns.

The complexity of the payments ecosystem means this challenge won’t solve itself. But with systematic measurement, thoughtful experimentation, and strategic optimization of the control levers available to you, you can transform the Stealthberg KPI from a hidden threat into a sustainable competitive advantage.

Your customers trust you with their payment information and their continued business. Make sure your payment infrastructure is worthy of that trust.

This post is part of a series on growing global subscription SaaS businesses. Keep in touch for posts that dive deeper into the Fishbone Framework for isolating billing success performance and techniques for measuring mix vs. performance in metrics.

Sea Level Rise Research

Sun, 19 Mar 2023 00:00:00 GMT

Could we adapt?

I walked into the wrong room freshman year at Stanford and got involved with SUPERSLR. The questions: 1) could seaports around the world protect themselves against 2 meters of sea level rise, and 2) does enough construction capacity exist to meet the additional demand from such a massive project?