How long does Shopify SEO take to work?

The '6 to 12 months' timeline people quote is mostly a symptom of slow execution, not an inherent property of SEO. At New Seas, technical SEO is done in the first 7 days, existing pages are optimized in the first 30 days, and first backlinks go live within 10 days of signing. When you execute fast, you get data fast — and results follow much sooner than the conventional timeline suggests.

Should a small e-commerce brand focus on technical SEO?

Rarely. For a Shopify brand with fewer than 100 SKUs and decent site hygiene, technical SEO won't unlock meaningful growth. In years of agency work, Kai has seen exactly one client where a technical fix directly caused significant growth. For most brands, on-page content and backlinks are where the results come from. Technical SEO matters for sites with thousands of SKUs, major keyword cannibalization, severe page speed issues, or custom CMS builds with structural problems.

How much of the SEO workflow should be automated?

About 80%. Automation is excellent for grunt work and getting you most of the way there. The last 20% — quality review, judgment calls, formatting edge cases — needs a human. New Seas spent over $50,000 this year on automation tooling and estimates $10,000 of it was wasted. Internal linking done by an automated SaaS tool produced worse results than manual internal linking. AI content drafts always go through a human content editor before client review or publishing.

When should an e-commerce brand start investing in SEO?

When you can comfortably spend $3,000/month and your category has enough search demand to generate a meaningful return. For most e-commerce brands, that comfort level arrives around $100K/month in revenue. Before that, do SEO in-house. Also check the math: take your main keyword's monthly search volume, multiply by a 25% CTR at position one, apply a 2% conversion rate, and multiply by your AOV. If the projected revenue is less than what you'd pay an agency, it doesn't make financial sense yet.

What's the minimum AOV for Shopify SEO to make sense?

There's no single number, but the math has to work. A $75 AOV product with a 2,000 search/month keyword will likely generate less revenue from SEO than a quality agency costs. Higher AOVs ($1,000+) can compensate for lower search volume — Kai points to a $50,000 hot tub brand and a $100,000 cryogenic chamber brand as examples where low traffic volume still justifies SEO investment. Ideally a couple hundred dollars minimum AOV; a couple thousand dollars is better for low-volume categories.

Should e-commerce brands write blog content for SEO?

Not unless it serves a specific, measurable purpose. Writing blogs just to 'keep the site fresh' is busy work. The only blog formats worth investing in are: competitor comparison articles, competitor alternative articles, solution comparisons, ultimate FAQs, buying guides, and product roundups targeting specific keywords. Everything else is time that could be spent on backlinks, collection pages, or internal links — things that directly drive revenue.

8 SEO Lessons I Wish I Knew Before Starting My Agency

Kai started in SEO completely by accident — dropped out of a marine biology graduate program in 2020, heard about SEO during a 2019 summer internship at an e-commerce brand, and taught himself through Backlinko and Brian Dean's YouTube channel. His first two years were freelance: personal fitness brands, personal injury lawyers, B2B companies, and a few e-commerce stores. In 2022 he went all in on e-commerce, and by 2025 he was running a seven-figure Shopify SEO agency working with 25+ brands generating $7M–$9M in revenue.

These are the eight lessons that took him three years — and a lot of expensive mistakes — to learn.

Lesson 1

Move with speed

The biggest competitive SEO advantage available to any brand, agency, or freelancer is the ability to execute faster than everyone else. This is especially true for smaller, more agile operations competing against larger brands with longer approval cycles.

In his first six months at a local SEO agency, Kai spent most of his time running Semrush crawls and reading about backlinks without being allowed to actually implement anything. He started updating location pages without telling his agency, cold emailing local businesses for link exchanges, and building a network of client link swaps — all without approval. He was getting results. He quit two months in to freelance full time.

At New Seas now: technical SEO is done in the first 7 days of any engagement. All existing pages are optimized within the first 30 days. First backlinks and Reddit comments go live within 10 days of contract signing. The reason people say SEO takes 6–12 months is largely because most organizations take 3 months just to get a change approved. If your competitor builds 10 new pages in 30 days, build 20 in 15 days. More speed means more data, and more data means better decisions faster.

Lesson 2

Break the rules

Google says buying links is against their spam guidelines. It is. Buying bad links — PBNs, low-traffic sites, irrelevant domains — will waste money and can get you penalized. That part is true. What's also true: buying quality backlinks with a proper filtering system delivers top rankings consistently, and it's more predictable, more scalable, and more profitable than trying to earn links through content or digital PR.

Earned links require too much from other people and deliver too little predictability. At an average link cost of $200–$250, you know exactly how many links you can build per month at a given budget, and you can scale that budget as you grow. Trying to earn links at scale is harder to systematize, harder to forecast, and slower to compound. If you don't buy links, your competitors who do will outrank you. That's not a threat — it's just what happens.

Lesson 3

Not everything applies to you

Enterprise SEO best practices were built for enterprise-scale websites. They don't apply to a 20-SKU Shopify brand. Local SEO tactics — Google Business Profile, local citations, review acquisition — are built for businesses serving a geographic area. They're irrelevant for a national e-commerce store. SaaS SEO playbooks were built for subscription products with long buying cycles. Applying them to a DTC brand will send you in the wrong direction.

Take in as much information as you can, but run everything through a filter: does this apply to my business type, my scale, and my situation? Things that sound outlandish usually are. Things that might be relevant — test them. Don't blindly accept any claim as universal fact, including advice from this channel. Validate in your own niche with your own data before committing to a tactic.

Lesson 4

Automate 80%, not 100%

New Seas spent over $50,000 this year on external agencies, in-house hires, and software to automate as much of their workflow as possible. About $10,000 of that was completely wasted. Automating 100% of any SEO workflow doesn't work — and in some cases actively makes the output worse.

Things they tried to fully automate that didn't work: internal linking (the SaaS tools produced worse results than manual linking), digital PR outreach at scale (AI pitches don't get responses consistently), and blog publishing from Google Docs to Shopify across 25–30 different client themes (formatting broke constantly due to theme variability). All three ended up being partially or fully reverted to manual handling.

The right framework: use AI and automation to handle grunt work and get 80% of the way there. A human takes the last 20%. AI content drafts always go through a human content editor before client review or publishing. Internal links are placed manually. That last 20% of human judgment is what separates output that's good enough from output that actually ranks and converts.

Lesson 5

Volume wins

New Seas used to build 5 collection pages per month per client. Then 10. Now 15–20. They've done 110 in a single month for one client. The output quality is better now than when they were doing fewer pages — because they got better by doing more, and because AI tooling let them produce more in less time without sacrificing quality.

More pages means more ranking opportunities means more client revenue. If you take 10 shots at goal per month and convert 30% of them, you get 3 ranking pages. If you take 20 shots and convert the same 30%, you get 6. The math is simple. Volume combined with speed is the compounding advantage that most brands never unlock because they spend too long perfecting each individual page before shipping it.

Lesson 6

Be quick to pivot

SEO is in one of the biggest periods of change in decades. It's not just Google anymore — it's ChatGPT, Claude, Perplexity, Gemini, and every other LLM that people are using to find and buy products. Tactics that worked in 2020 are losing ground. New tactics — Reddit brand seeding, LLM visibility strategies, multi-platform social for AI citation — are being tested right now by the people who will make the most money from them.

Think of it like investing: you run 100 experiments, 98 of them teach you nothing or cost you something, and 1 or 2 of them deliver outsized returns that more than make up for everything else. The SEOs at the cutting edge of new tactics are going to burn some money testing — and then find things that work at scale that most agencies won't figure out for another year or two. The cost of not testing is falling behind.

Lesson 7

Don't invest in SEO too early

New Seas won't work with any brand doing less than $100,000 per month in revenue or that can't generate a meaningful return on their services. There are two separate tests to run before investing in agency SEO.

Test one — can you afford it? For most e-commerce brands, $3,000 per month becomes a comfortable SEO budget around the $100K/month revenue mark. Below that, do SEO in-house. Above it, agency leverage starts to make financial sense.

Test two — does the math work? Use this formula before committing:

Monthly search volume for main keyword × 25% (estimated CTR at position 1) = monthly visits

Monthly visits × 2% (generous conversion rate on cold organic traffic) = conversions per month

Conversions × AOV = projected monthly revenue from SEO

If projected revenue < what you'd pay an agency, SEO doesn't make financial sense yet.

Kai turned down a brand this week with a $75–$80 AOV and a main keyword at 2,000 searches per month. Even at an aggressive conversion rate, the projected SEO revenue was less than the agency fee. The previous agency had taken them on anyway and gotten fired six months later. High AOVs ($1,000+) can compensate for lower search volume — a $50,000 hot tub brand or a $100,000 cryogenic chamber brand can justify SEO on low traffic because each conversion is worth so much. But the math has to work before you start.

If your category doesn't have enough search volume yet, your money is better spent on demand generation — Meta ads — rather than demand capture channels like Google SEO.

Lesson 8

Don't do anything just because you think you should

Two examples of busy work that masquerades as SEO: writing one blog post per week "to keep the site fresh," and spending time improving page speed by 1%. Neither of these is going to move revenue. One is a myth about how Google freshness works; the other is a distraction from tactics that deliver 10–20% gains.

The rule: only do SEO activities you genuinely believe will deliver an outsized return on time and money. Building quality backlinks — yes. Optimizing existing collections and product pages to rank higher — yes. Creating competitor comparison content to capture branded search — yes. Building branded Reddit comments for LLM visibility — yes. Improving Core Web Vitals by two points — probably not. Writing a blog no one will read — definitely not.

Busy work feels like progress. It isn't. You can't get stuck in a cycle of it.

I spent 4 years and $1m learning these 8 SEO lessons