Independent growth advisory

I find where 8- and 9-figure e-commerce brands can grow.

My work has taken me across some of the biggest e-commerce brands. Here are three common problems that I fix.

Curtis Howland
Curtis Howland
VP, Misfit Marketing · Durham, NC
$150M+Managed ad spend · $100M+ of it on Meta
60+Brands
$500M+Revenue driven
9Clients at a time, maximum
Read weekly by operators at
Google, Procter & Gamble, Target, ByteDance, Ritual, Dr. Squatch, Lovevery, Koala, HiSmile, Tinuiti.
Subscribers to the weekly letter, by employer. Readers, not endorsements.
The three most common issues I fix
And how I fix them.

Optimizing numbers you do not trust just gets you to the wrong answer faster.

I
Measurement

You don’t know which number to trust.

Every platform claims the same sale. If no number is trustworthy you cannot set a target, and without a target you cannot say what a good ad looks like.

How I fix it
Set the tools up once. Read the metrics daily.
Attribution hierarchyOpen the framework →
Four tiers of measurement, ordered by how much to trust each.
MER / nCAC
Is the business working?
Channel allocation
Where should we spend?
Ad-level optimization
Which ads should we scale?
Most brands start here
Tracking & targets
Is the data clean?
Source: Misfit Marketing attribution framework
Attribution works top down. Most brands start in the middle.
II
Optimization

You’re not properly using the ads you already made.

All digital channels have strong algorithms, but they do not know your business context, and they optimize to make those companies money, not you. Left alone they crown a winner early, starve everything else, and hand you a story about what worked that is really a story about what they funded.

How I fix it
Careful optimization. Detailed tracking. Consistency.
Scale and killPlot your data →
At low spend a CPA can land almost anywhere by chance. The bands converge as spend accumulates.
Kill
Scale
Target CPA
$100$1K$10K$100K
CPA per ad ↑Spend per ad, log scale →
Source: Ad Scale Scatter. 75 ads; Poisson confidence bands at z = 1.645; illustrative account.
We have cut CPA by over 50% on this optimization alone for a nine-figure business.
III
Production

What to make, how many, and how to test them.

Once allocation is right, production becomes the constraint, and it needs volume and diversity together. Roughly 85% of what you make will fail. That is fine, as long as you are testing it correctly.

How I fix it
Set production targets. Track carefully. Push for diversity.
What a programme produces with one, both, or neither.
Volume →
Variations
High output, one idea.
Compounding
Enough shots, spread wide enough to find the outlier.
Guessing
Too few tests to learn anything.
Scattered
Nothing reaches a readable spend.
Diversity →
Source: Misfit Marketing creative testing programme. Axes are dimension-agnostic.
Testing compounds only when volume and diversity rise together.
About

A data analyst who had to go make marketing performance happen.

First · the analyst

I started in the data. Once you have built the query yourself, you cannot unsee the assumptions inside the number.

Then · Carpe

I owned marketing at Carpe and learned the P&L implications. Being accountable for marketing performance, not reporting on it, is what taught me how to move it.

Now · Misfit

Now I run Misfit Marketing. Nine clients maximum, because most agencies optimize for retention and reporting instead of the founder’s P&L.

Curtis Howland
“Most paid media problems are not paid media problems. They are creative problems, pricing problems, margin problems, or organizational problems wearing a paid media mask.”
Working together

A first call: we evaluate each other.

I am working out whether I can move your number, and you are working out whether that is worth what it costs. There are two ways I work.

As a consultant

A handful of brands at a time, privately, alongside your team.

Hands-on, through Misfit Marketing

In the accounts as your media buyer. Nine clients, maximum; no long-term contract.

This is for

8-9 figure ecom brands with a team already running paid, where growth has flattened or CAC is climbing.

Not for

Launching paid from scratch, agency of record across every channel, or a second opinion you already agree with. If that is you, the tools above run on your own numbers, free.

Request a conversation

You give me read access, I go through the account, and I tell you where it is actually stuck and what I would do first. If it is not a fit, I will say so.

I read every request and reply within 24 hours. How this works

Not the right fit? Get the weekly email instead.

One tactic a week, to 3,475 operators.