The short answer
Marketing attribution means connecting each sale back to the marketing that produced it. For a small business the workable version is a five-link chain, lead to appointment to estimate to sale to revenue, with the lead's source recorded at intake and carried through every step, so each channel reports the revenue it produced rather than the clicks it collected.
The question usually arrives with the ad bill. You spent four figures on Google and Facebook last month, jobs came in, the month was decent, and you could not say with any confidence which dollars did the work. The ad dashboards claim credit in clicks and impressions. The bank account speaks in revenue. Nothing in between connects the two, so the budget decision gets made the way it was made last quarter: by feel. This guide is about attribution, the practice of connecting customers back to the marketing that produced them, done at a small-business scale where there is no analyst and no appetite for a fourteen-tab spreadsheet.
What is marketing attribution, in plain terms?
Attribution answers one question per customer: what did we do that caused this person to buy? A customer saw your truck, searched your name a week later, clicked an ad, filled in the form, booked an estimate, and hired you. Attribution decides which of those touches gets credit for the revenue.
Big companies argue about the crediting math, first touch versus last touch versus fractional models across every interaction. A small business can skip that debate almost entirely. Before the math matters, the data has to exist: you need to know, for each customer who paid you, which channel produced the original lead. Most small businesses cannot answer that for last month's customers, and that gap, not the choice of model, is the actual attribution problem.
Why does attribution feel impossible for a small business?
Because the trail breaks in the handoffs. The ad platform knows about the click but not the sale. The CRM, if there is one, knows about the sale but recorded the lead source as "website" or left the field blank. Phone calls arrive with no source at all unless someone asks, and "how did you hear about us?" gets answered with "Google," which could mean an ad, a map listing, or a search for your name after a neighbor's referral.
It is worth knowing that this is not a small-business deficiency. In a 2024 Gartner survey of 378 senior marketing leaders, only 52% said they could prove marketing's value to their organization, and those are teams with analysts and attribution software. The encouraging flip side: a small business has one advantage the enterprise does not, a short and visible path from lead to sale. Five links, not fifty. That path can be instrumented without a data team.
What should a small business actually track?
Track the chain, not the channels. Every paying customer passed through the same five stages, and each stage is a countable event:
- Lead. An inquiry arrived: form, call, DM, walk-in. Record the source at this moment, because it is the only moment the source is knowable.
- Appointment. The lead became a scheduled conversation, an estimate visit, a consult, a showing.
- Estimate. A number went out in writing.
- Sale. The estimate was accepted.
- Revenue. The invoice was paid, for an actual amount.
When the source sticks to the lead through all five stages, every marketing question becomes a filter. Revenue by source. Close rate by source. Average job size by source. That last one regularly rearranges budgets on its own, because channels that produce many leads and channels that produce good leads are rarely the same channels.
Why is "report revenue, not clicks" the rule?
Because clicks are the one metric your ad platform will always supply and your bank account will never confirm. A channel can win on every dashboard metric, cost per click, cost per lead, volume, and still produce customers who book small jobs, cancel appointments, or never accept an estimate. The only score that settles the argument is revenue per channel against spend per channel, and you can only compute it if the chain above is connected.
There is a second, quieter reason. Clicks measure the ad; revenue measures the whole pipeline, and the difference between the two is where the fixable problems live. If Google leads convert to appointments at half the rate of referral leads, the ad is probably fine and the follow-up is not: the form reply that goes out hours later, the voicemail that never gets returned, the estimate that goes quiet with no reminder behind it. Attribution built on revenue does not just grade your marketing. It shows you where paid-for leads stall after they arrive, which is usually a cheaper fix than buying more of them, and it is the same stall our guide to lead routing tackles from the ownership side.
How do you connect the chain without hiring an analyst?
The mechanics are less exotic than the attribution-software industry suggests:
- Capture the source at intake, automatically. Forms carry a hidden source field. Ad clicks carry tracking parameters. Calls from ads use a tracking number that logs the source before anyone answers. The rule is that no human should have to remember to ask.
- Make the CRM the single spine. Every lead, from every channel, becomes a record in one system with the source attached, and the appointment, estimate, and sale are updates to that same record rather than entries in three other tools. If your records tend to decay mid-pipeline, that is a solvable problem we cover in the CRM automation guide.
- Connect the tools you already run. The booking calendar, the invoicing app, and the ad platforms each hold one link of the chain. Data integration is the work of making them pass records along automatically, so revenue lands next to source without anyone retyping it.
- Read one report monthly. Revenue by source, spend by source, side by side. Twenty minutes, once a month, beats a real-time dashboard nobody opens.
What can you do with attribution once it works?
Three decisions get easier immediately. Budget: shift spend toward the channels that produce revenue, not volume, and cut the ones that only produce clicks. Sales process: when one channel's leads stall at the same stage repeatedly, fix that stage instead of blaming the channel. And pricing of growth: once you know a customer from a given channel costs a knowable amount and produces a knowable margin, "should we spend more on marketing?" turns from a leap of faith into arithmetic.
FAQ
What is the simplest attribution model for a small business?
Last touch with a recorded source: credit the channel that directly produced the inquiry, captured automatically at intake. It is imperfect, referrals that later search your name get filed under search, but it is consistent, cheap, and enough to compare revenue per channel, which is the decision the model exists to support.
Do I need attribution software?
Usually not as a first step. You need a source field that fills itself, a CRM that carries it to the sale, and connections between the tools you already use. Dedicated attribution platforms earn their cost at ad volumes most small businesses never reach; connected records cover the decisions that matter first.
How do I attribute phone calls?
Use tracking numbers: a distinct forwarding number per channel, on ads, the website, and listings, each logging its source when it rings. Pair that with an intake habit of confirming new callers against the CRM, and phone leads join the same chain as form leads instead of remaining a blind spot.
Want to know which of your channels actually produces revenue? The free AI audit maps your lead sources, finds where the chain breaks between inquiry and invoice, and comes back with a scoped plan to connect it.