Judging an SEO Campaign When Word of Mouth Already Drives the Business

Some businesses have a lovely problem when it comes to measuring marketing: word of mouth already works. A referral-heavy firm - think an interiors or specialist health group that gets a steady flow of work from architects, builders, GPs or existing clients - starts an SEO or ads campaign from a high baseline. Enquiries were already coming in before you spent a cent. So when a new enquiry lands, the honest question is uncomfortable: did the campaign bring that in, or would they have found you through a referral anyway? Judging the campaign in that environment is genuinely harder, and I would rather say so than pretend a dashboard makes it simple.

Why a strong referral base muddies the water

The problem is that referral and search are not cleanly separate journeys. Someone hears your name from a colleague, then searches for it to find your site - that shows up in your analytics as a search visit, but the referral did the real work. Or the reverse: they find you in search, sit on it, and only enquire weeks later after a friend mentions you too. Two channels touched the same person, and the last one to get clicked takes all the credit in a naive report. When referrals are already strong, this overlap is not an edge case - it is most of your traffic. That is exactly why B2B buyers researching for months is such a recurring theme: the buying journey is long and multi-touch, and the enquiry is the visible end of a process you mostly cannot see.

The short version: you cannot cleanly attribute every lead to a single channel, and any tool that claims you can is overselling. What you can do is measure the parts that are genuinely measurable, be transparent about the parts that are not, and avoid inventing precision that is not there.

Get the tracking right before you argue about anything

Before any of the harder attribution thinking, the basic measurement has to be sound, because if it is wrong every conclusion downstream is wrong. That means the analytics events are set up properly - a form submission is recorded as a form submission, a phone-number click is separated from an actual call, and obvious noise like bot traffic is filtered out. I have written separately on getting conversion tracking right before you claim any ROI, and it matters double here, because when the signal is already muddy you cannot afford to also be measuring the wrong thing.

Separate the channels as far as you honestly can

You will never get perfect separation, but you can get useful separation. A few practical moves:

• Split brand from non-brand searches. Someone searching your company name almost certainly already knew you - likely a referral. Someone searching a generic category or product term found you cold. Watching those two pools separately tells you far more than one blended traffic number. This is the same reason brand and non-brand terms need handling separately.
• Ask new enquirers how they heard about you. A single dropdown or a one-line question on the enquiry form is imperfect and people forget, but across enough leads it is a real signal - and it is the only one that captures the offline referral your analytics can never see.
• Watch the non-brand search trend over time, not any single lead. If cold, generic-term traffic is climbing month on month, the campaign is bringing in people who did not previously know you existed. That is the part referrals were not already delivering.
• Keep the channels in separate columns in reporting. Do not let SEO quietly take credit for referral-driven brand searches, and do not let referrals hide the genuine new demand that search opened up.

Think in terms of incrementality

The idea that cuts through the fog is incrementality - not "how many leads touched this channel" but "how many leads exist now that would not exist without it". A brand-name search from a warm referral is largely not incremental; that person was going to find you regardless. A non-brand enquiry from someone who searched a generic term, had never heard of you, and would otherwise have gone to a competitor - that is incremental, and it is the honest measure of what the campaign added on top of your existing word of mouth.

You will rarely get an exact incremental number in a business like this, and I would be suspicious of anyone who hands you one to two decimal places. But you can get the direction, and the direction is what matters: is the pool of people arriving cold, through generic search, growing? If yes, SEO is doing a job referrals were not. If the only movement is in brand searches, you are largely paying to capture demand your referral network already created - which is worth knowing, because it changes whether the spend is worth continuing.

Measure what is cleanly measurable, be transparent about what is not, and do not invent precision that was never there.

What I would actually do

If referrals already drive most of your work, set the expectation up front that attribution will be honest rather than tidy. Fix the tracking, split brand from non-brand, add a "how did you hear about us" question, and then judge the campaign mainly on whether cold, non-brand enquiries are growing over time - because that is the ground referrals were never covering. Resist the pull of a single dashboard number that claims to settle it; in a referral-heavy business that number is usually more confident than it has any right to be. If you want to sanity-check whether a provider will be straight with you about this rather than claim clean credit for everything, it is one of the sharper things to probe when choosing an SEO provider. For the wider context of measuring search in a long-cycle B2B market, this sits under SEO for B2B and technical companies.