Marketing Attribution: How to Know What's Actually Working

Marketing attribution is the practice of working out which of your marketing efforts actually drove a sale, so you can put more money into what works and less into what doesn't. It exists to answer a deceptively hard question: when a customer buys, what made them buy? For any business spending money across more than one channel, getting a sensible answer to that question is the difference between investing and guessing. Here is how attribution works, why it is harder than it sounds, and a realistic approach for small businesses.

What it is

Attribution assigns credit for a conversion — a sale, a lead, a sign-up — to the marketing touchpoints that contributed to it. A touchpoint is any interaction a customer has with your marketing: seeing a social ad, reading a blog post, clicking a search result, opening an email, visiting your site.

The reason this matters is simple. If you knew exactly which efforts produced customers, you would shift every pound toward those and stop wasting the rest. Attribution is the attempt to get close to that knowledge. It is the engine behind honest measurement of marketing ROI — without it, ROI figures are built on guesswork about which channel gets the credit.

The problem it solves

Here is the core difficulty: customers almost never buy after a single interaction. A typical purchase might look like this:

1. The customer sees a short video ad on social media and notices your brand.
2. A week later they read one of your articles after a search.
3. They sign up to your email list.
4. An email offer brings them back.
5. They search your brand name directly and buy.

Now ask the obvious question: which of those five steps deserves the credit for the sale? The social ad created awareness. The article built trust. The email prompted action. The brand search was just the final click. Give all the credit to that last search and you will conclude your ads and content do nothing — and cut the very things that started the journey.

The central trap of measurement is rewarding the last click and starving everything that made the last click possible. Attribution exists to avoid exactly that mistake.

This is also why naive measurement so often overvalues some channels and undervalues others, distorting where the budget goes.

The common attribution models

Different models split the credit differently. None is perfect; each emphasises something.

First-touch and last-touch are the simplest and the most misleading, because each ignores most of the journey. Linear, time-decay and position-based models try to share credit more fairly across multiple touchpoints — collectively called multi-touch attribution. They give a fuller picture but require more data and more effort to maintain. A deeper side-by-side is in our piece comparing attribution models.

Why perfect attribution is impossible

It is worth saying plainly: no attribution model is fully accurate, and chasing perfection wastes time. Several forces guarantee blind spots.

• Offline and word of mouth. A recommendation from a friend or a billboard glimpsed in passing leaves no digital trail.
• Privacy and tracking limits. Cookie restrictions and privacy rules mean you simply cannot see every step a customer takes — and rightly so. Marketers should understand the basics of UK GDPR as it affects marketing data.
• Cross-device journeys. People research on a phone and buy on a laptop, breaking the chain.
• The dark funnel. Conversations, group chats and content consumed where you cannot measure it.

Accepting that attribution is directional, not precise is the beginning of using it well. The goal is to be roughly right about what works, not exactly right about everything.

A simple approach for small businesses

Small businesses do not need an elaborate attribution stack. In fact, an overcomplicated model you cannot maintain is worse than a simple one you trust. Start here:

1. Just ask. Add a "How did you hear about us?" question at sign-up or checkout. It is unfashionable and imperfect, but it captures word of mouth and offline sources that no tracking tool can see — and it is often the single most valuable input.
2. Track the basics well. Set up your analytics properly, use tagged links on campaigns, and make sure conversions are recorded. Clean basic data beats sophisticated data full of gaps.
3. Look at the whole picture. Combine what customers tell you with what analytics shows. Where they agree, act with confidence. Where they disagree, investigate.
4. Watch trends, not single numbers. Is a channel's contribution rising or falling over time? Direction is more useful than a falsely precise percentage.

There is sensible industry guidance on this for smaller firms; CM Beyer, for example, has examined what marketing attribution actually means for small businesses, and the practical message is consistent: start simple and improve as you grow.

How to act on what you learn

Attribution is only useful if it changes decisions. Once you have a directional read on what works, shift budget toward the channels that consistently appear in the journeys of your best customers, and reduce spend on those that never do. Pay particular attention to the channels that start journeys, not just those that finish them — starving awareness to feed the last click is the most common and most costly attribution error.

The bottom line

Marketing attribution is how you work out which efforts actually earn their keep, in a world where customers touch several channels before they buy. The models — first-touch, last-touch, linear, time-decay and position-based — each tell part of the story, and none tells all of it. For small businesses, the winning move is not a complex system but a simple, honest one: ask customers how they found you, track the basics cleanly, watch the trends, and accept that being roughly right is enough to spend your money far more wisely than guessing ever could.