Siloed marketing data: why it hides growth and how to fix it
Short answer Siloed marketing data means each platform grades itself: Google Ads credits its ads, Meta counts people who only saw an ad, Search Console counts clicks, not customers. Added up, they often claim more leads than you got and hide SEO's early role. Fix it with one lead list recording where each person first heard of you, judged by the total.
Want help applying this to your business? Talk it through on a free 30-minute call.
Book a callWhat is siloed marketing data?
It's when each marketing tool reports only on itself and nobody lines the numbers up. Every platform counts the customers it touched, by its own rules, so each one looks like the reason you got them.
| Report | What it counts | What it can't see |
|---|---|---|
| Google Ads | Leads after a click on your Google ad. Its reports credit Google paid clicks by default (Analytics Help: Select attribution settings) | The free listing or page the person found first |
| Meta Ads Manager | Actions within 1 or 7 days after a link click and, if the setting includes views, 1 day after someone only saw the ad (Meta Business Help Center) | Whether that person would have found you on Google anyway |
| Google Analytics 4 (GA4) | Website visits and key events (the actions you mark as leads), by channel such as Paid Search or Organic Search | Calls straight from your Google listing, referrals, anything that never touches your site |
| Search Console | Impressions, clicks and searches on Google, activity "before a person arrived at your website" (Google Search Central) | Whether the click became a lead |
| Google Business Profile | "Calls", meaning taps on the call button (Business Profile Help) | Whether anyone answered or booked |
| Your own lead list | Every call, form and booking, once | Where each lead came from, unless you write it down |
Only the last row counts each customer once. It's the one to trust, as long as it records the source.
Is my marketing data siloed? A quick check
Add up the leads each platform claims for last month. Then count the new leads in your own records for the same month.
- Platforms' total is higher: some people are counted in two or more reports, or some "leads" weren't real (spam, wrong numbers). The gap is the least that's overcounted.
- Platforms' total is lower: leads arrive through doors no report watches, such as phone calls, referrals and walk-ins.
- You can't do the check: you have no lead list of your own. Start there (below).
Even Google's own tools won't agree. Google says Search Console and Analytics "use different metrics and systems, which means the data won't match completely" (Google Search Central).
Why siloed data hides growth
SEO and paid ads compete for the same searches and the same people, so each one's report absorbs part of the other's result. It happens three ways.
SEO starts the sale, the ad gets the credit
A homeowner reads your page on what a new furnace costs, leaves, and a week later searches your company name and clicks your ad. Google Ads logs a lead from the ad. Google's own settings page explains why the free visit got nothing there:
"For Google Ads reports: Google Paid attribution is the default setting, and can be updated to Paid and Organic." (Analytics Help: Select attribution settings)
GA4's own reports, by contrast, let paid and organic channels share credit by default (Analytics Help: Select attribution settings). Read only the ad report and the page that started the sale looks worthless. Cut SEO on that basis and you may be cutting where those name searches come from.
Ads take clicks you'd get for free
When you already rank first for a search, part of what your ad gets would have gone to your free listing anyway. Google's researchers measured this across 390 studies of advertisers pausing their search ads:
"we found that on average, 50% of the ad clicks that occur with a top rank organic result are incremental." (Google Research, 2012)
Incremental means the click wouldn't have happened without the ad. The same study put it at 82% when the free result ranked 2nd to 4th, and 96% below that. It measured clicks, not sales, and Google adds that individual advertisers vary. A field experiment at eBay found its ads on its own brand name had "no measurable short-term benefits" (NBER Working Paper 20171, 2014). eBay is a household name. A small business whose rivals bid on its name may differ, which is why you test rather than assume.
One customer, counted twice
Someone sees your Instagram ad without clicking, then searches your name on Google, clicks your free result and books on your website the same day. If your Meta setting counts views, Meta claims the booking, because it came within a day of the view. GA4 files the visit under Organic Search. Each is right by its own rules, and your check now shows two leads for one customer.
What one list shows
At an accounting firm we run marketing for, its 51 leads up to November 11, sorted by source, showed who reached "ready to start": 6 of 16 from organic search, 2 of 8 from Google ads, 1 of 5 from Meta ads, and 9 of 9 sent by ChatGPT answers. No single platform's report could show that ranking, because each counts only its own leads.
How to see SEO and paid ads together
You don't need a new tool. You need one list, two free Google reports, and one rule for decisions.
1. Keep one lead list. One row per new lead, entered once, whatever the source.
| Column | How to fill it |
|---|---|
| First heard | Ask "How did you first hear about us?" on the form and on the first call |
| Came back through | The source of the visit or call that became the lead: GA4 or tagged links for forms, a tracking number or the caller's answer for calls |
| Booked, paid | Your calendar and invoices |
| Revenue | The invoice |
The two source columns split the work: "first heard" shows what SEO, social posts and word of mouth start; "came back through" shows what ads and name searches finish. To make GA4 record forms and calls: GA4 key events.
2. Link Search Console to Google Ads. This turns on the paid and organic report, which puts your ad clicks and free clicks for each search side by side (Google Ads Help: About measuring paid and organic search results). Link it now, because the report starts on that day:
"Your organic data is only reported back to the date you started importing it from webmaster tools, so you won't find historical organic data from before you established that link." (Google Ads Help)
How to use it to test cutting ads where you rank: SEO vs Google Ads.
3. Read GA4's attribution paths. In GA4, Advertising, then Attribution paths lists the channels people touched before a lead, paid and organic together (Analytics Help: Get started with attribution). If GA4 is linked to Google Ads, leave the "Channels that can receive credit" setting alone unless you mean to change Google Ads too: Google says a change applies to all linked Google Ads accounts and can affect its bidding and reporting (Analytics Help: Select attribution settings).
4. Judge every change by the total. When you add, cut or move money, compare total new customers in your list for comparable weeks, not one platform's count. Google's researchers advise the same for search ads:
"We recommend that advertisers employ randomized experiments (e.g., geo-based experiments) to better quantify the incremental traffic and lift in conversions from the search ad campaigns" (Google Research, 2012)
For a small business, that can mean pausing ads in one town but not another, or for four weeks against the four weeks before. Note anything else that changed, such as season, prices or staff.
Which numbers to put side by side
| Question | Where to look |
|---|---|
| Are my ads buying clicks I'd get free? | Paid and organic report: searches with both ad and free clicks, and combined clicks before and after a pause |
| Is SEO starting sales that ads finish? | Your list: "first heard" says Google search or your website, "came back through" says an ad |
| Do ads raise searches for my name? | Search Console clicks on searches containing your business name, in weeks with ads on and off |
| What does a customer cost overall? | Everything you spend on marketing (ads, SEO fees, tools) ÷ new paying customers in your list |
| Which channel should get more money? | Paying customers per channel, counted by "first heard" and by "came back through" |
A channel that wins on both source columns is the safest place for the next dollar. One that wins only on "came back through" may be collecting sales something else started.
Example
Illustrative cases with round numbers.
A plumber whose reports add up to more than reality (US dollars)
Last month Google Ads reported 30 leads, Meta 12 and GA4 18 from organic search: 60 in all. The call and form log shows 42 new leads, so at least 18 were counted twice or weren't real. The "first heard" column shows 15 of the 42 first read a page on the plumber's website from a Google search, and 9 of those later called through an ad on the company name. The Google Ads report credited those 9 to the ad.
A dental clinic paying for its own name (Canadian dollars)
The clinic spends CAD 600 a month on ads for searches of its own name, where it already ranks first for free. It pauses them for four weeks: 52 new-patient bookings, against 54 in the four weeks before. It keeps them paused, moves the CAD 600 to emergency and implant searches where it doesn't rank, and checks monthly whether a competitor starts bidding on its name.
A landscaper on Instagram (US dollars)
Meta reports 20 leads in May from USD 1,000 of ads. The lead list shows 11 new customers who first heard of the business on Instagram; 6 of them later searched the name and came in through the free Google listing, so GA4 filed them under Organic Search. The owner judges Instagram on those 11 customers, about USD 91 each, not on Meta's 20.
FAQ
Should I cut SEO if Google Ads gets credit for most sales?
Not on that report alone. Google Ads reports credit Google paid clicks by default, so they can't show the pages people read first. Check the "first heard" column in your own list before cutting.
Do Google Ads help my SEO rankings?
No. Google says: "Investment in paid search has no impact on your organic search ranking." (Google Ads Help). Ads still teach SEO something: the searches behind your paying ad clicks show which pages are worth writing.
Does good SEO content lower my Google Ads costs?
Not through rankings. Quality Score has three parts: expected clickthrough rate, ad relevance and landing page experience (Google Ads Help: About Quality Score). A useful page behind the ad helps the third; your free ranking isn't one of them. And Google says the score "is not an input in the ad auction"; it's a diagnostic.
Which attribution model should a small business use?
GA4 offers three: data-driven, paid and organic last click, and Google paid channels last click (Analytics Help: Get started with attribution). First click, linear, time decay and position-based were removed in November 2023 (Analytics Help: Select attribution settings). With a few dozen leads a month, asking each customer where they first heard of you tells you more than switching models.
Do I need a paid dashboard tool?
Not to start. A shared spreadsheet with one row per lead does the core job. A dashboard can put each platform's numbers on one page, but it still shows each platform's own count; it can't tell which leads are the same person.
Different for you?
- You want the monthly report laid out: the report to ask for.
- GA4 doesn't count your calls and forms yet: GA4 key events.
- You want Google Ads to learn from real jobs, not clicks: tracking a lead to a booked job.
- You're deciding between SEO and ads, or testing ad cuts: SEO vs Google Ads.
- You're checking SEO on its own: is my SEO working.
- You're splitting the whole budget: what $500, $1,500 or $3,000 buys.
- You want every lead logged, answered and followed to a sale: Conversion Agent.