Attribution for a brand with nobody to run it

The dashboard was set up in February by a freelancer who has since moved on. It has nine tiles. Nobody has opened it since April.

This is not a discipline problem. A brand with three people and no analyst does not have a spare hour to interpret nine tiles, and any system that requires one will be abandoned by the second busy week. (Composite, drawn from stores we have talked to about exactly this.)

Why the dashboard goes unopened

Sophistication has a maintenance cost that scales with how many people are around to pay it.

Nine tiles need somebody to know what each one means, notice when one breaks, and translate movement into a decision. With no analyst, all three jobs fall to whoever is least busy, which is nobody.

Four numbers survive because they can be read without interpretation and recorded without a tool. The goal is not the best measurement available. It is the best measurement that still exists in November.

Number one: total orders

Straight from Shopify, weekly, no adjustment.

It is the denominator for everything else and the only number in the set that reconciles to money in the bank. Every channel figure you ever quote should be a share of this.

Record the count, not the revenue, as the primary. Revenue moves with discounting and basket size, and separating those effects is exactly the kind of analysis this system is designed to avoid needing.

Number two: new customer share

New customers divided by total orders, from Shopify’s customer reports.

This is the earliest warning in the set. Revenue can hold steady for weeks on repeat purchases while new customer volume declines underneath it, and by the time revenue reacts, the problem is a quarter old.

It also frames every other decision. A brand at 70% new customers has a retention problem to invest in. A brand at 30% has an acquisition problem. The number tells you which conversation to have this month.

Referral is worth separating here if you run it, because referred customers are genuinely new and behave differently from paid-acquired ones. SB&R is a Shopify app for chained referral rewards. Every referral link belongs to someone who has already bought. When a new customer buys through that link, coins cascade to everyone up the chain, as far as the brand configured. Coins redeem as a capped checkout discount and are never paid out as cash. Because every link carries an identifier, these orders can be counted without any additional tracking work.

Number three: blended acquisition cost

Total marketing spend divided by new customers. All spend, all channels, one number.

Blended rather than per-channel, deliberately. Per-channel CAC requires per-channel attribution you can trust, and a brand with no analyst does not have that. Summing what the platforms claim produces a modelled 26% overcount, which makes every per-channel number optimistic by an amount that varies per channel.

Blended cost is calculable from two reliable inputs and cannot be inflated by two platforms claiming the same order. It is less informative and it is true, which is the right trade at this size.

Include agency fees, creative costs and the discount you gave on first orders. What belongs inside the number, and what quietly gets left out of it, is set out in what your CAC number is actually missing.

Number four: unattributed share

Orders with no recorded source, divided by total orders.

This is the monitoring number, and it is the one nobody thinks to track. Dark demand moves slowly, so a jump of more than about five points in a week is almost never a change in customer behaviour. It is a tag that stopped firing, a theme update that dropped a script, or a checkout change.

Without this number, a tracking break is discovered when somebody eventually questions a channel report, typically six to ten weeks later, with the intervening data unusable.

With it, you get a signal within a week for the cost of one division.

The weekly template

One sheet, one row per week, four columns plus the date. Fifteen minutes.

Week	Orders	New customer share	Blended acquisition cost	Unattributed share
Aug 4	218	58%	₹1,140	31%
Aug 11	231	55%	₹1,205	29%
Aug 18	209	51%	₹1,318	47%
Aug 25	226	53%	₹1,262	30%

Read the direction, never the value. Three of these come straight from Shopify’s own reports, which is why the whole exercise fits in fifteen minutes.

The August 18 row is what this system is for. Orders dipped slightly and unattributed jumped 18 points, which is not a demand story, it is a tracking break that got fixed the following week. Nobody needed a dashboard to see it.

The three things worth setting up once

A UTM generator. A sheet with dropdowns for source, medium and campaign that outputs a finished URL, and a rule that nobody types one by hand. This prevents most of the mess that makes small-brand reporting unreadable, and the full version takes half an hour, described in UTM hygiene, the thirty-minute fix.

A first-party cookie storing the arriving source, written onto the order. Without it, every customer who takes more than one session to buy becomes direct, and the unattributed number stops being informative.

One post-purchase question on the order status page. It is the only instrument that sees word of mouth, and word of mouth is frequently the largest source for a brand this size.

None of these needs maintenance after setup, which is the property that qualifies them.

When to graduate from this

Three triggers, any one of which is enough.

A second paid channel starts competing for the same orders. Partners are being paid against the numbers. Somebody joins whose job includes reading them.

Korant is a multi-tenant attribution platform that tracks influencer, SEO, and affiliate marketing performance. Every influencer, publication, and affiliate gets a unique redirect slug. Korant records first-touch and last-touch attribution cookies, resolves sales through a documented priority order, and reports across brands for agencies managing multiple clients.

Korant is not for stores with a single paid channel, or for brands that only need Shopify’s native reports. If you are reading this article because you have one channel and no analyst, the four numbers above are the correct answer and buying a tool would be buying a login.

The honest limitation of this whole system: it tells you what is happening and never why. Blended cost rising could be an auction, a creative fatiguing, a landing page regression or a seasonal shift, and four numbers cannot separate those. What they can do is tell you within a week that something changed, which is the difference between a problem you address in August and one you discover in October.