Marketing Attribution for Small Business: Which Channel Is Actually Working?

If you are running SEO, Google Ads and Meta Ads at the same time, and a sale or enquiry comes in, which one gets the credit? Most small businesses cannot actually answer that question, and it means every decision about where to spend the next dollar is a guess dressed up as a strategy. Marketing attribution is simply the practice of tracking that properly, and it does not need expensive software to do it.

This guide covers what attribution actually means, the two models that matter for a small business (not the six or seven a big agency deck will show you), and exactly how to set it up using free tools, with a worked example of what the data looks like once it is running.

Why “which channel is working?” is the wrong first question

The question most business owners ask is “which channel is working?”, as if there is one right answer sitting in a dashboard somewhere. The better question is “which channel is working for which part of the job?”, because different channels do different jobs in the same customer’s path to buying.

Search Console might show someone searched your business name after seeing a Meta ad. Google Ads might show a search click converted on a landing page that SEO content first introduced the customer to weeks earlier. If you only ever look at the last click before a sale, every one of those earlier touchpoints looks like it did nothing, when it may have done most of the actual persuading. Picking a single attribution model and trusting it blindly is how perfectly good channels get cut from a budget for the wrong reason.

What attribution actually means (without the agency jargon)

Attribution is the rule you use to assign credit for a conversion across the touchpoints that led to it. Google’s own attribution documentation narrows the current GA4 reporting models down to a short list for exactly this reason: most of the elaborate multi-touch models marketers used to argue about added complexity without adding much decision-useful clarity for a typical business.

For a small business, you do not need all of them. You need two, used side by side.

The two models a small business actually needs

Last touch gives all the credit to whatever the customer did right before converting, typically the final search, ad click or form visit. It is the default in GA4, in most contact forms, and in how most people mentally think about “what worked.” It is excellent for answering: what should I spend more of my ad budget on right now?

First touch gives all the credit to how the customer first found you, days, weeks or months earlier. It is the one most small businesses never look at, and it is the one that makes SEO, content and organic social make sense, because those channels rarely close the sale directly; they open the relationship that a later ad or branded search closes.

Run both side by side on the same conversion data and a pattern usually appears quickly: paid search and paid social dominate the last-touch view because they are good at catching people at the point of decision, while organic channels dominate the first-touch view because they are good at starting the relationship. Neither view alone tells the full story. Together, they tell you whether a channel that “isn’t converting” on last touch is actually quietly filling the top of the funnel that your ads are closing.

Setting this up without buying anything

None of this requires a dedicated attribution platform. Four free pieces, set up once, do almost all of the work.

1. UTM parameters on every paid and social link. Every Google Ads, Meta Ads, email and social link should carry a UTM source, medium and campaign tag. This is what lets GA4 tell “Meta Ads: October carousel” apart from “organic Facebook post,” which a plain “Facebook” source label cannot do.
2. Google Analytics 4, read properly. GA4’s Advertising reports include both a last-click and a data-driven multi-touch view out of the box, covered in Google’s attribution documentation. Most businesses have GA4 installed and never open this section.
3. Google Search Console, cross-checked against GA4. Search Console shows which queries generate impressions and clicks before a visitor ever becomes a session in GA4. A query with strong impressions but a weak click-through rate is a trust problem at the search-results stage, worth fixing before you credit or blame the landing page it points to.
4. A hidden source field on your own enquiry form. This is the one most businesses skip, and it is the one that fixes attribution for phone and form enquiries specifically, not just web sessions. Online Road’s own site forms carry this: a hidden field, filled in automatically by a small script as the visitor browses, that records the landing page, referrer and campaign parameters, so every enquiry that lands in the inbox is already labelled with how that person actually arrived, rather than relying on a drop-down the customer fills in from memory. Our web design team builds this kind of tracking into new sites by default.

Set those four up once and every new enquiry from that point on arrives pre-labelled. No guessing, no asking the customer “how did you hear about us?” and hoping they remember correctly.

A worked example (illustrative, not a real client)

To make this concrete: imagine a Sydney-based trade business spending on both Google Ads and SEO, with a hidden source field now running on its enquiry form.

Looking at last touch only, Google Ads appears to produce most enquiries, since paid search is good at capturing people the moment they decide to call a tradesperson. On that view alone, the obvious move is to cut the SEO budget and push everything into ads.

But cross-checking the same enquiries by first touch tells a different story: a large share of the people who eventually enquired through a Google Ads click had first landed on the site weeks earlier through an organic search result, looked around, left, and only converted once a remarketing-style ad later caught them at the point of deciding. Cutting SEO in that scenario would not fix anything, it would just mean fewer people ever entering that path in the first place, and the Google Ads cost per lead would climb as the warm, already-aware audience it had been closing dried up.

This is exactly the trap last-touch-only reporting sets, and it is why running both models matters more than picking “the right one.”

Three mistakes that quietly waste attribution data

• Judging SEO on a 30-day window. Paid channels can be assessed quickly because they capture existing demand. SEO builds demand over months. A channel that takes 90 days to show its real shape will always look weak on a 30-day scoreboard, and plenty of businesses cut it right before it was about to pay off.
• Leaving phone enquiries unlabelled. Many small businesses track form fills carefully and then take every phone call as an unattributed mystery. A hidden source field or a simple “which page were you looking at when you called?” note in the CRM closes most of that gap for free.
• Comparing cost per lead across channels without comparing lead quality. A channel that produces cheap leads that rarely convert into paying customers is not actually cheaper than one that costs more per lead but closes at a much higher rate. Track cost per sale, not just cost per lead, once you have enough volume to do it.

How this connects to a bigger diagnosis

Attribution tells you where your customers are coming from. It does not, by itself, tell you why a channel that is bringing plenty of the right people still is not converting them, which is a different problem entirely. If your attribution data shows healthy traffic from a channel but weak conversion once people land, that is a sign to look at the page itself rather than the channel. Our framework for diagnosing traffic that isn’t turning into sales covers exactly that next step, and our guide to setting up Google Ads conversion tracking covers the paid-channel half of the same tracking problem in more technical detail.

A five-minute attribution audit you can run today

Before setting up anything new, check what you already have. Most businesses are closer to a working system than they think.

1. Open GA4’s Advertising reports. Switch between the last-click and data-driven views for the same date range. If the two tell noticeably different stories about which channel matters most, that gap is exactly what first touch versus last touch is meant to surface.
2. Check your paid and social links for UTM tags. Click through a recent Google Ads, Meta Ads and email link and look at the URL. If there is no utm_source or utm_campaign parameter, that channel’s traffic is landing in GA4 as generic “paid” or “social,” with no way to tell one campaign from another.
3. Pull up your last 10 enquiries. For each one, can you say with confidence which channel actually brought that person in, or are you guessing from the subject line and hoping? If it is mostly guesswork, the hidden source field above is the single highest-leverage fix available.
4. Compare Search Console impressions against GA4 sessions for your top three commercial pages. A big gap between people seeing you in search results and people actually landing on the site points to a click-through problem worth fixing before you judge that page’s conversion rate at all.

If two or more of those four come back weak, tracking is the gap, not your marketing. Fix the tracking first; the channel decisions get a lot easier once the data underneath them is trustworthy.

Where to go from here

Getting attribution right does not require a new platform or a bigger budget, just four pieces set up properly and read side by side instead of trusting whichever number GA4 shows you by default. You can see what properly attributed, multi-channel results actually look like across different Australian businesses on our case studies page, where every figure is tied back to a real, tracked campaign rather than a projection.

If you are running multiple channels and genuinely cannot tell which one is doing the work, a free strategy session includes a look at your own tracking setup, not just a pitch for more spend.