Skip to content
Dispensary Rank
How the work runs

Our Cannabis SEO Process, In The Order It Actually Happens

Six phases. About six months for a mid sized dispensary chain, faster for a single store. The order matters more than any one tactic. Every phase below tells you what you get, what is needed from you, and what would slow it down.

Start With Phase One, Free See The Six Phases
  • 6 phases, run in this order every time
  • ~6 months for a mid sized chain, faster for one store
  • 1st phase is the free audit, before anything is agreed
Why This Order

Most Cannabis SEO Fails On Order, Not On Skill

It goes wrong the same way almost every time, and it is easy to spot once you know what to look for.

Content gets bought first, because content is easy to buy and easy to show a board. Then it lands on a site where Google cannot see the menu and the product filters are eating the crawl budget. A quarter of the year goes by, nothing moves, and everyone decides SEO does not work in cannabis. It does. It was just run in the wrong order.

  1. 01
    Delivery before content A page Google cannot read is worth nothing regardless of how well it is written. Every technical fault is closed before a word is commissioned, which is why phase three comes before phase four.
  2. 02
    One term per store Two stores targeting the same phrase split their links, their internal links, and their engagement between two pages. Nothing warns you this is happening, and both pages sit lower than one would have.
  3. 03
    Nothing bought No links, no placements, no networks. Half the recoveries in the case studies exist because somebody bought links, and the cleanup costs more than the links ever returned.
  4. 04
    Named, not averaged Reporting per location rather than as a chain average, because an average across fifty stores hides the five going backwards. If a location has not moved, it is named.
Phase By Phase

The phases

Every phase lists the same three things at the end: what lands on your desk, what is needed from your side, and the thing most likely to slow it down. The short version of all six sits on the front page.

01 Week 1 to 2

Discovery and audit

Before anything is proposed, the site is read the way a crawler reads it.

What Google actually renders on your pages, index coverage against what you think is indexed, crawl behaviour in the logs where they exist, manual action status, and Business Profile health across every location. Nothing is recommended at this stage, because a recommendation made before the diagnosis is a guess with an invoice attached.

What happens
  • Fetch every template as a crawler receives it, with no JavaScript executed
  • Compare index coverage against the sitemap and against what you assumed was live
  • Check every Business Profile for eligibility and suspension risk
  • Measure how alike your location pages have become
You get

A prioritised audit document, ordered by what each finding costs rather than by how easy it is to fix

Needed from you

Search Console access, analytics access, and a named person who can answer questions about the site history

What makes this take longer

Waiting on access. This is the single most common delay and it is entirely avoidable.

02 Week 2 to 4

Keyword and competitor mapping

One distinct primary term per store, so your own pages stop competing with each other.

Every location gets a term it can own, mapped across the chain so no two stores are chasing the same phrase. Alongside that, the product, brand, and education searches competitors are winning quietly, which is usually where the volume actually sits.

What happens
  • Map one primary term per store, checked for overlap across the whole estate
  • Identify the product and brand searches your competitors rank for and you do not
  • Separate terms worth pursuing from terms no single store can win
You get

A keyword map, store by store, with the reasoning for each assignment

Needed from you

A list of every location with its address, and which stores you consider priorities

What makes this take longer

Stores opening or closing mid engagement, which changes the map.

03 Month 1 to 2

Technical work

Delivery is fixed before a single word is written.

Menu rendering, age gate behaviour, filter crawl traps, canonicals, redirects, structured data, and speed. This phase is first for a reason: writing content onto a foundation Google can only half read is how agencies burn a quarter of budget with nothing to show for it.

What happens
  • Write a developer specification, in their language rather than in SEO terms
  • Get the menu into the HTML, or set out what your provider would need to change
  • Close the crawl traps generating thousands of filter URLs
  • Fix the age gate so a crawler is not treated as a repeat first visit forever
You get

A developer specification with acceptance criteria, then verification once it ships

Needed from you

Developer time, or permission to work with your platform provider directly

What makes this take longer

A menu platform that will not render server side. Sometimes the honest answer is that the platform has to change, and that is a business decision rather than a technical one.

04 Month 2 to 4

Local search and content

Profiles, citations, and location pages built from facts rather than from a template.

Business Profiles brought into line, citations corrected, and location pages rebuilt around real store detail: parking, landmarks, staff, hours that are actually true. The standard is that no paragraph should still make sense if the city name were swapped for another.

What happens
  • Rebuild location pages from per store facts, prototyped first and then rolled out
  • Bring every Business Profile into line with the rules that actually cause suspensions
  • Correct citations where the name, address, or phone disagree across sources
  • Tie it together with internal links whose anchors describe the destination
You get

Rebuilt location pages, corrected profiles, and a content calendar for what follows

Needed from you

Someone at each store who can confirm the local detail, and profile access

What makes this take longer

Collecting real detail from fifty stores. This is genuinely slow and there is no way around it, because invented detail is worse than none.

05 Month 3 onward

Authority and links

Earned rather than bought, because bought links are what caused half the recoveries in the case studies.

Digital PR, local partnerships, and content the industry has a reason to reference. Slower than buying links and the only version that survives a manual review. If your previous agency bought links, this phase starts with cleaning up rather than building.

What happens
  • Build content the industry has an actual reason to cite
  • Pursue local partnerships and press that would exist without the link
  • Clean up an inherited spam profile where one exists, with a documented log
You get

A referring domain profile that would survive a manual review, and the log to prove it

Needed from you

Patience. This phase shows almost nothing for the first two months.

What makes this take longer

An inherited link profile worse than it first appeared, which is common and only becomes visible once the audit is deep.

06 Continuous

Measurement and reporting

Reported against the window, not against the best month.

Monthly reporting on what moved, what did not, and what is planned next. Locations that have not improved are named rather than averaged away, because an average across fifty stores can hide five that are going backwards.

What happens
  • Report per location rather than as a single chain average
  • Name what did not work and say why
  • Track the geo grid per store, since a proximity search has no single position
You get

A monthly report you could hand to a board without translating it first

Needed from you

Thirty minutes a month to review it, otherwise the reporting is theatre

What makes this take longer

Nothing. This phase runs alongside everything else.

Read this honestly

These timings assume a mid sized chain and a site in ordinary condition. A single store moves faster. A chain carrying a manual action moves slower, and the first month goes on recovery rather than on growth. Anyone quoting you a fixed timeline before looking at your site is quoting a template.

See what this sequence produced on real chains
Questions

Questions About How This Runs

Technical fixes can show within weeks, because they remove a blockage rather than build something new. Local and content work usually shows between eight and sixteen weeks. Link and authority work is slower again and shows almost nothing for the first two months.
Because content written onto a site Google can only half read returns nothing. If your menu sits inside an iframe and your filters create thousands of near identical URLs, new pages land in a site the crawler is already struggling with. Fixing the site first is what makes the content spend pay off.
Search Console and analytics access, Business Profile access, developer time or permission to speak to your platform provider, and one person who can answer questions about the site history. Waiting on access is the most common delay and the easiest to avoid.
Sometimes. If you have a recent audit and the technical work is genuinely finished, phases one and three get much shorter. More often, a site that believes its technical work is done has never had it checked, and skipping the check means building on an assumption.
You hear about it straight away, not at the next monthly report. An inherited spam link profile, or a penalty nobody told you about, changes both the plan and the timeline. Pretending otherwise only moves the bad news further down the road.
Yes, and it usually works better that way. The developer brief gets written in their language rather than in SEO terms, with pass or fail criteria they can test against, so nobody has to translate anything.

What phase one would find on your site

The discovery phase runs these checks first. All three are free and none need an email address.

What a crawler receives from your pages Ten checks against the HTML, ordered by what each finding costs. Whether your products reach the index The fault that decides whether phase three is a fix or a rebuild. What this sequence has produced Six engagements, reported per location with their windows.

Start With The Diagnosis

Phase one costs nothing to preview. Send your domain and you will get a written read on what is costing you visibility and what to fix first, whether or not the rest follows.

Get Your Free Audit Or see what the work covers