Should I invest in SEO or paid ads?

Usually both, in a set order. Paid ads can bring sales this month, but they stop the day you stop paying. SEO is slower, and what it builds can keep working after the work stops. If your ads earn back their cost at your margin, run them now. If people already search for what you sell, start SEO alongside.

The usual framing is cheap-and-slow against fast-and-pricey. Both halves are only half true.

SEO is not free. Google says it doesn't accept payment to rank pages higher, so you can't buy your way up the results. You pay in people, content and time instead.

SEO is slow, though not evenly. Google's SEO Starter Guide says some changes take effect in a few hours and others take several months. It suggests waiting a few weeks before judging whether a change helped.

Paid ads won't lift your rankings either. Google's Ads help says investment in paid search has no impact on your organic search ranking. The two only meet in your bank account and in your reports.

What one store's data shows

One store's anonymised GA4 export, 1 January 2024 to 21 August 2026. It holds shares of revenue only: no ad spend, no order counts.

The usual test is a GA4 channel report: Organic Search revenue against Paid Search revenue, and the bigger number wins. One store's export shows what that duel leaves out.

Direct is the biggest row. On the Channels sheet it holds 57.7% of revenue in last click, first click and touched views alike. GA4 files a visit as Direct when someone uses a saved link or types your address. So whatever search earned in this store, it earned from the share Direct left over. Switching the credit view does not move that row an inch.

The touched view on the Channels sheet adds up to 110.4% by design, because a journey that touched two channels counts in both. Picture a buyer who clicked a search ad, then came back through an organic result. Each channel can claim that sale. Add up each side's claims and some sales get counted twice.

On the Journeys sheet, one-touch journeys hold 79.5% of revenue, with 0.5 days to buy. Those are people who arrived ready. If that single touch was a search click, paid or organic, GA4 gives it the whole sale. The export cannot say what made the buyer ready in the first place.

The AI Assistant row on the Channels sheet reads 0.0% in all three views. GA4 counts clicks from Google's AI Overviews and AI Mode as Organic Search, and keeps a separate AI Assistant channel for other assistants. If you are weighing SEO because search is changing, read both rows in your own export first.

What the export cannot show is cost. It holds no ad spend, and SEO work rarely shows up in any report as a cost at all. A channel with no cost line looks profitable by default.

Why does the usual answer mislead?

• "SEO is free" counts the clicks, not the work. Writing, site fixes and outreach cost salaries or invoices. Leave them out and SEO wins every comparison before it starts.
• Organic search harvests demand other channels made. A search for your brand name comes from someone who already knows you. Search Console's help says branded traffic often comes from users already familiar with your site, while non-branded traffic can indicate new audience growth. Only the second is SEO finding new buyers.
• Paid search can buy clicks you would get free. You may rank first for your own name and pay for it too. Then some ad clicks would have landed on your listing anyway. Test that before you cut it, with a branded search test.
• The clocks differ. Judge SEO in its first months and it looks like waste. Judge ads over the same months and they look like genius. Compare each over the time it needs to work.

What would a fair comparison look like?

One scoreboard: your store's revenue, not each platform's claims. One cost basis: ad spend for paid, and hours plus invoices for SEO. One question: what would you lose without it?

That last question splits the two. You can switch paid ads off for a random group or a few regions and measure the gap. SEO is much harder to hold out, because you can't hide your rankings from half your visitors. So paid search can be tested directly, while SEO gets judged on non-branded growth over time.

That asymmetry is a reason to test paid first. It is not a reason to trust SEO blindly.

What to do this week

1. Split brand from non-brand in Search Console. Open the Performance report for Search results, select the Query filter and choose Non-branded. Pass: non-branded clicks hold steady or grow against the same months last year. Fail: almost all organic clicks are branded, so SEO is mostly catching people who already know you.
2. Hold each search campaign against break-even. In Google Ads, select Modify columns from the Columns drop-down and add Conv. value/cost from the Conversions columns. Pass: each campaign beats 1 divided by your margin. Fail: one doesn't, so fix or trim it before you move budget.
3. Put a yearly price on SEO. Add up content, developer hours, tools and agency fees for the past year. Pass: you have one number to set next to ad spend. Fail: nobody knows, so SEO has been judged as free.

Check the homework. Your GA4 Attribution paths export already holds the evidence. Causality Engine reads that one file and shows what each channel caused next to what last-click gave it, in 1 to 2 minutes, for €99 once (excluding VAT), refundable within 30 days. Check the homework

Sources, 1 October 2026: Guide to how Google Search works (Google Search Central); SEO Starter Guide (Google Search Central); About measuring paid and organic search results (Google Ads Help); Default channel group (Analytics Help); Performance report (Search results): Common tasks and use cases (Search Console Help); About Target ROAS bidding (Google Ads Help).

Related answers

• All Budget & channel mix questions
• How to compare SEO and paid ads, step by step
• Should a sports nutrition store invest in SEO or paid ads?
• How do I measure organic search value?
• Causal attribution, explained