Request available times

A LIVING PUBLIC WORKING DOCUMENT · EDITION 01

Our SEO & AIO Strategy, Built in Public

A living implementation roadmap for Markethink.ai.

This is Markethink's ongoing SEO and AIO implementation strategy. We will keep adding scoped strategies, implementation work, evidence, and lessons as the work develops.

The first 30 days establish an initial foundation; they are not the strategy's lifespan or a completion deadline. AIO builds on conventional SEO and is not a guaranteed ranking recipe.

Last substantive update: Latest update

Day ranges such as Days 1–5 and Days 6–12 are target windows in the initial foundation phase, not recorded sprint dates, a claimed start, or an endpoint for the ongoing strategy.

01 Days 1–5 Evidence 02 Days 6–12 Commercial clarity 03 Days 10–18 First-party proof 04 Days 19–24 Implementation 05 Days 25–30 Verification 06 Days 60/90 Business review
Download the reusable checklist Read the roadmap

THE OPERATING RULE

Evidence before expansion.

Publishing more pages is easy. Proving which buyer question, URL, evidence gap, and business action each page should own is the work.

01

Improve what exists first.

Map buyer needs to current URLs before proposing another page.

02

Keep proof close to the claim.

Use approved evidence, state its limits, and leave missing results unfilled.

03

Measure three different outcomes.

Organic discovery, observed AI-search representation, and qualified inquiries are separate signals.

For the repeatable method behind improving existing pages, use our SEO content feedback loop.

ONGOING REVIEW AND IMPLEMENTATION

Review, implement, verify, learn, repeat.

The roadmap stays useful by turning each evidence review into the next capacity-aware cycle. Unfinished work remains visible, and newly justified opportunities enter only after they are scoped.

  1. 01

    Review the available evidence and keep unknowns visible.

  2. 02

    Choose the next priorities within current capacity.

  3. 03

    Implement only the work that has been scoped and authorized.

  4. 04

    Verify delivery on the intended production surface and review results separately.

  5. 05

    Record the change, evidence, and lessons in the dated log.

  6. 06

    Carry unfinished work forward, then repeat from the updated baseline.

STATUS LANGUAGE

Published is not the same as proven.

Counts come directly from the structured roadmap entries. They do not move with the calendar, and this page never converts them into a completion percentage. New items enter only when they are actually scoped; dates never auto-complete them.

Planned13

Work has not started and no release is claimed.

In progress2

Work has started or is under review. It is not live unless a public evidence link says otherwise.

Live0

Published on a public URL. Publication alone does not prove that the acceptance check or business outcome passed.

Verified4

The stated acceptance check passed on the verified date. This does not mean SEO uplift or AI citations were proven.

Blocked0

A named dependency prevents progress. The blocker must be removed before work can continue.

THE MAINTAINED ROADMAP

Every scoped action has a finish line.

19 structured actions today. No automatic day-based progress. New actions appear only after they are scoped, and dates appear only where a public release or acceptance check has evidence.

Showing all 19 actions.

01

Days 1–5

Evidence

3 actions

Days 1–5

Confirm the buyer questions and the search results they produce

In progress

Why

The roadmap should start from real buyer language and current result types, not an invented keyword list.

Current progress: Buyer-query research has begun. Complete coverage, recorded scopes, intent review, and a repeatable fixed sample remain to be verified.

Target page
Research worksheet and the current Markethink.ai search footprint
Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Write a fixed sample of commercial and educational questions a buyer could ask before choosing managed marketing help.
  2. Record the result format, intent, cited sources, and pages already competing for each question.
  3. Keep the location, language, device, and collection date with the sample so it can be repeated.

Done criteria

  • Every sampled question has a recorded source, date, scope, result type, and searcher intent.
  • The sample separates commercial, comparison, and educational intent before any page changes are proposed.

Days 1–5

Map each buyer need to an existing URL before creating anything new

Planned

Why

A clear ownership map reduces duplicated intent and protects the pages that already have a defined job.

Target page
Homepage, commercial pages, guides, research pages, and blog URLs in the live sitemap
Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Inventory indexable Markethink pages from the live sitemap and classify each by audience, intent, and conversion job.
  2. Assign one primary page to every validated query family and flag genuine coverage gaps separately.
  3. Treat similarity as a review signal, then use intent and observed query overlap before deciding to split or combine pages.

Done criteria

  • Each validated query family has one primary URL, one page role, and one next-step destination.
  • No proposed page duplicates the homepage, commercial hub, comparison, or buyer-guide job.

Days 1–5

Record crawl, index, discovery, and conversion baselines

In progress

Why

Without a dated baseline, later movement cannot be separated from a memory of what changed.

Current progress: Baseline collection has begun. Complete source coverage, dated scopes, denominators, and the unavailable-data review remain to be verified.

Target page
Baseline ledger for Markethink.ai and the approved target URLs
Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Record crawl and index signals for the target URLs from accessible first-party sources.
  2. Record organic discovery, observed AI-search citations and accuracy, and qualified inquiry data as separate lanes.
  3. Use Not yet measured when a source is unavailable instead of substituting zero.

Done criteria

  • Every baseline observation includes source, date, scope, and denominator.
  • Unavailable measures display Not yet measured and no inaccessible vendor data is claimed.
02

Days 6–12

Commercial clarity

3 actions

Days 6–12

Improve the homepage explanation of the managed service

Planned

Why

A buyer should understand that Markethink combines an AI marketing system with accountable expert marketers, not mistake it for a self-serve tool.

Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Compare the current first viewport and service explanation against the validated buyer questions.
  2. Clarify who directs strategy, what runs day to day, what the client approves, and how work connects to leads and follow-up.
  3. Preserve the homepage as the primary commercial owner for AI marketing agency intent.

Done criteria

  • The first viewport states the offer, intended buyer, expert responsibility, and next step without generic AI claims.
  • The homepage keeps one clear commercial job and the walkthrough destination remains accurate.

Days 6–12

Strengthen the existing commercial hub and comparison coverage

Planned

Why

Commercial and comparison questions need decision support that is distinct from the homepage and educational blog content.

Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Review the existing commercial hub and comparison coverage against the query-to-URL map.
  2. Add decision criteria, fit and non-fit guidance, tradeoffs, and a restrained handoff to the walkthrough where gaps are proven.
  3. Keep comparison language factual and avoid inventing competitor features, pricing, or outcomes.

Done criteria

  • The commercial hub answers the buyer questions assigned to it without taking the homepage primary term.
  • Comparison content shows selection criteria and tradeoffs with working contextual links.

Days 6–12

Improve the AI marketing provider buyer guide

Planned

Why

Businesses evaluating an AI marketing provider need practical decision support that is separate from an optional agency-partner or white-label track.

Optional separate agency-partner track

Evaluate a distinct agency-partner or white-label guide only if query evidence proves a separate audience need. It must not replace or blur the provider guide for businesses.

Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Add a provider evaluation matrix that helps businesses compare scope, accountability, evidence, and fit.
  2. Separate work accelerated by AI from expert judgment, client approval, and follow-up responsibility.
  3. Include one documented example and link the guide back to the homepage as the primary commercial owner.

Done criteria

  • The guide includes a provider evaluation matrix, clear role separation, one documented example, and a working homepage link.
  • The guide serves businesses evaluating an AI marketing provider without becoming a white-label partner guide.
03

Days 10–18

First-party proof

1 action

Days 10–18

Prepare one documented first-party work story

Planned

Why

A specific, documented account of the work is more useful than a generic claim about capability.

Target page
One approved first-party work story and its relevant commercial page
Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Choose one workflow with documented public-use clearance and evidence that can be shown without exposing private client information.
  2. Document the starting problem, approved scope, work produced, review decisions, and observable result without inferring causation.
  3. Separate verified facts from Markethink analysis and confirm every named claim and asset is cleared for public use.

Done criteria

  • Every claim, name, image, and result has documented public-use clearance and a traceable evidence source.
  • The story distinguishes delivered work from measured outcome and contains no invented metric or testimonial.
04

Days 19–24

Implementation

3 actions

Days 19–24

Implement and review the approved page changes

Planned

Why

Research only creates value when it becomes a focused, reviewable change on the URL that owns the buyer need.

Target page
Homepage and approved existing commercial or guide pages
Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Turn only approved findings into page-level copy, structure, metadata, and usability changes.
  2. Use a preview and compare the candidate against the current production page before publication.
  3. Run focused tests, a full build, responsive review, accessibility checks, and link verification before release.

Done criteria

  • Every changed page has a documented reason tied to the query and URL map.
  • The approved preview passes content, responsive, schema, canonical, link, console, network, and overflow checks.

Days 19–24

Add useful proof and contextual internal links

Planned

Why

A page needs support from relevant evidence and nearby pages, not just a sitemap entry or footer link.

Target page
Approved target pages, relevant blog guides, and the AI marketing statistics research ledger
Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Place approved proof next to the claim it supports and state the evidence limit in plain language.
  2. Add crawlable contextual links from the homepage, research, and relevant guide surfaces using varied descriptive anchors.
  3. Resolve every unique destination and confirm that the target page returns the expected content.

Done criteria

  • Every material proof claim has an approved public source or is removed.
  • The strategic pages have relevant inbound and outbound contextual links with successful destinations.

Days 19–24

Research relevant external-authority opportunities

Planned

Why

Useful third-party corroboration can strengthen discoverability when the publication, evidence, and audience fit are relevant.

Target page
External-authority research ledger, not a public outreach campaign
Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Shortlist relevant publications, associations, expert-commentary formats, profiles, and communities connected to the buyer questions.
  2. Prepare the evidence each publication would need and record its audience fit, editorial standard, and contribution path.
  3. Review the evidence and audience fit before outreach so each proposed contribution is useful and relevant.

Done criteria

  • The shortlist contains relevant publications with a documented reason, source URL, evidence need, and audience fit.
  • Each proposed contribution has an evidence pack and fit review ready before outreach is considered.
05

Days 25–30

Verification

3 actions

Days 25–30

Verify production and indexing signals

Planned

Why

A merged change is not complete until the canonical page is live, crawlable, internally connected, and technically consistent.

Target page
Canonical production URLs changed during the 30-day cycle
Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Open each bare canonical URL in a fresh browser and confirm the release-specific copy is present.
  2. Check the main response, canonical, robots state, sitemap, structured data, links, console, network, and responsive layout.
  3. Record indexing observations separately from publication because discovery and indexing can lag a release.

Done criteria

  • Each released page passes its production acceptance checks on the canonical URL.
  • Publication, crawl eligibility, and observed indexing are reported as separate states.

Days 25–30

Repeat the fixed search-grounded question sample

Planned

Why

A stable sample makes directional changes reviewable without pretending that one prompt is a universal ranking report.

Target page
Fixed buyer-question observation ledger
Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Run the same approved questions with the same language, location, device, and collection method used for the baseline.
  2. Record mentions, cited pages, source diversity, answer accuracy, and material changes for each question.
  3. Keep missing citations and inaccurate representations visible rather than selecting only favorable responses.

Done criteria

  • The follow-up sample uses the same questions and scope as the baseline or clearly documents any unavoidable change.
  • Every observation includes source, date, scope, and denominator and is labeled directional.

Days 25–30

Review inquiry quality and choose the next cycle

Planned

Why

Visibility matters only when it contributes to relevant business conversations, and low volume does not justify a fabricated conclusion.

Target page
Qualified inquiry review and next-cycle decision record
Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Review qualified inquiries by landing page, stated need, source note, stage, and accepted next action.
  2. Separate inquiry count from qualification and from later sales outcomes.
  3. Choose the next cycle from the strongest verified gap in discovery, answer accuracy, page clarity, proof, or follow-up.

Done criteria

  • The review records the source, date range, audience scope, and denominator for every inquiry observation.
  • The next cycle names one priority, one target URL, one acceptance check, and one business guardrail.
06

Days 60/90

Business review

2 actions

Day 60 review

Review the first business-outcome window

Planned

Why

Search discovery, citations, and inquiry quality need a longer observation window than a production release check.

Target page
Day-60 evidence review
Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Compare the day-60 evidence with the original baseline and the exact release dates.
  2. Review organic discovery, AI-search observations, and qualified inquiries separately before looking for a relationship.
  3. Record confounding changes such as campaigns, site releases, seasonality, or offer changes.

Done criteria

  • The day-60 review uses dated sources, scope, and denominators and does not infer causation from coincidence.
  • The decision record states continue, revise, or stop for each active hypothesis.

Day 90 review

Review the second business-outcome window

Planned

Why

A second fixed review tests whether a directional signal persists, disappears, or needs a different explanation.

Target page
Day-90 evidence review and next-quarter roadmap
Verified date
Not yet verified
Public evidence
No public evidence link available

Repeatable how

  1. Repeat the day-60 evidence collection at the same scope where possible.
  2. Compare page-level discovery, answer accuracy, cited sources, qualified inquiries, and downstream stages.
  3. Publish only conclusions supported by the available evidence and preserve Not yet measured where gaps remain.

Done criteria

  • The day-90 review documents comparable inputs and every scope change.
  • The next roadmap decision names the evidence used and the remaining uncertainty.
07

Pre-existing progress

Already shipped

2 actions

Released September 12, 2026

Publish the AI marketing statistics page and source ledger

Verified

Why

A durable research asset gives readers attributable evidence with definitions and limits instead of untraceable summary statistics.

Verified date
September 12, 2026

Repeatable how

  1. Maintain one public ledger with stable statistic IDs, original-source URLs, study scope, sample notes, evidence type, and limitations.
  2. Render the research page and downloadable CSV and JSON from the reviewed ledger.
  3. Keep Markethink analysis separate from third-party findings and date substantive reviews.

Done criteria

  • The public page identifies 60 statistics from 9 original sources and preserves source context.
  • The downloadable CSV and JSON ledgers are linked from the page and publicly available.

Public evidence reviewed September 12, 2026

Publish source-backed search and marketing operations guides

Verified

Why

Educational guides provide useful context around search visibility, evidence, approvals, and operating discipline without turning every query into a new commercial page.

Verified date
September 12, 2026

Repeatable how

  1. Use distinct informational intent for each guide and link it to the nearest relevant commercial or research destination.
  2. Support methodology claims with current primary sources and state platform and measurement limits plainly.
  3. Update visible dates only when the content changes substantively.

Done criteria

  • The blog index and the AI-search guide are publicly reachable with server-rendered content.
  • The guide explains that AI-search visibility is not guaranteed and links to relevant Markethink context.
08

Concurrent corrections

Under review

2 actions

Published and independently verified September 13, 2026

Publish and verify the V2 B2B editorial image system

Verified

Why

The statistics resource and editorial library need a coherent B2B visual system across responsive pages, cards, and sharing surfaces without presenting conceptual scenes as customer proof.

Current progress: The exact approved V2 page, card, responsive, and sharing image system is published and independently verified in production.

Verified date
September 13, 2026

Repeatable how

  1. Use the exact approved V2 source selections and responsive derivatives across the statistics hero, nine article heroes, blog cards, and related cards.
  2. Preserve the full-scene P04 mobile treatment without CSS recropping and keep rejected D and the first P08 absent.
  3. Provide a dedicated 1200×630 sharing derivative for every placement and use it consistently in Open Graph, Twitter, and dimensioned ImageObject metadata.
  4. Verify public routes, image decoding, responsive selection, card order, metadata, useful conceptual alt text, and protected statistics and palette invariants.

Done criteria

  • The approved P01–P10 responsive images are live in the statistics hero, nine article heroes, blog cards, and related cards; P04 preserves the full logistics scene at 320px, and rejected D and the first P08 remain absent.
  • All ten routes use distinct versioned 1200×630 sharing images consistently across Open Graph, Twitter, and dimensioned ImageObject metadata with useful conceptual-editorial alt text; the rejected legacy food and pottery sharing images are absent.
  • Production verification confirms ten canonical routes, exact P02–P10 card order, decoded assets, contained narrow layouts, unchanged statistics ledgers and charts, and the established charcoal, warm-white, soft-wash, and green website palette.

Current review workstream

Review the isolated website palette correction

Verified

Why

The blog and research surfaces should use the canonical charcoal, warm white, soft green, vivid green, and established gray palette without bundling unrelated redesign work.

Current progress: The canonical palette release and final base cleanup are independently verified in production.

Target page
Published blog index, representative article, and statistics chart
Verified date
September 12, 2026

Repeatable how

  1. Keep the correction isolated to the blog index, one representative article, and one statistics SVG chart.
  2. Review each changed surface at desktop and mobile sizes and confirm that text contrast remains accessible.
  3. Keep the final canonical-base cleanup open until an independent production check closes it.

Done criteria

  • The published correction preserves the green blog and article treatment, contained desktop and 320px views, CAPTCHA behavior, IAB SVG colors, the 82/45/37 values, byte-identical CSV and JSON hashes, and all 60 statistic anchors.
  • The final canonical-base cleanup is independently verified and closed without changing the protected research values or links.

MEASUREMENT NOTES

Three lanes, three honest baselines.

Future observations must include a source, date, scope, and denominator. A missing baseline is Not yet measured, never zero. The current baseline is evidence for the next decision, not a permanent cap on future actions.

01

Organic discovery

Not yet measured

What we will observe

Crawl and index state, search impressions, clicks, landing pages, and qualified organic visits for the approved URL set.

Important limit

Publication and crawl eligibility are not proof of impressions, clicks, rankings, or business impact.

Required on every observation: source · date · scope · denominator
02

Observed AI-search citations and accuracy

Not yet measured

What we will observe

Mentions, cited pages, source diversity, and factual representation across a fixed sample of buyer questions.

Important limit

A sampled answer is directional. It is not a universal AI-search ranking or a guarantee of future citation.

Required on every observation: source · date · scope · denominator
03

Qualified inquiries

Not yet measured

What we will observe

Inquiries connected to the reviewed landing pages, including source note, stated need, qualification, stage, and accepted next action.

Important limit

Inquiry count, inquiry quality, meetings, opportunities, and revenue are separate stages and must not be collapsed into one result.

Required on every observation: source · date · scope · denominator

METHOD, NOT MAGIC

What AIO changes, and what it does not.

Conventional SEO, useful original or first-hand content, accurate facts, crawlable pages, internal links, and clear site structure remain central.

Google's current official guidance says its generative AI search features use core Search ranking and quality systems. It recommends valuable, non-commodity content and a clear technical structure. It also says there is no special schema.org type, AI text file, or additional technical requirement that guarantees inclusion.

Structured data still helps search systems understand eligible page types when it matches visible content. It does not guarantee a rich result, ranking, mention, or citation.

To adapt this method to another business and turn its real inputs into an initial cycle plus a capacity-aware next backlog, use the ongoing SEO and AI-search strategy playbook. This roadmap is Markethink's public implementation; the library playbook is a reusable visitor method, not independent proof.

Google Search Central · Updated July 10, 2026 Optimizing your website for generative AI features on Google Search Google Search Central · Site-owner guidance AI features and your website Google Search Central · Content guidance Using generative AI content on your website

Experiment boundary

We are documenting what Markethink changes, when it changes, what acceptance check passed, and what we observe afterward. We are not claiming that every older strategy or vendor measurement has been independently verified, and we are not promising rankings, citations, traffic, leads, or revenue.

RELATED FIELD NOTES

Read the evidence behind the operating plan.

Practical guide How to get your business to show up in AI search

Read the source-backed foundation, readiness checks, evidence model, and measurement limits.

 Research ledger AI marketing statistics, with the source behind every number

Review the public statistics, study scopes, original-source links, and downloadable source ledgers.

APPEND-ONLY CHANGE LOG

What changed.

  1. Homepage and 404 accessibility corrections published and verified

    Published and verified the English and Spanish homepage CTA corrections with 32px clearance, successful activation to the correct localized walkthrough routes, and normal reachable content flow in short landscape viewports. The branded unknown-route experience retains genuine HTTP 404 status and now gives every action, including the green primary button, a visible offset keyboard-focus outline. A semantic review also closed 52 contextual or decorative image warnings across both language variants: all reviewed images already receive appropriate context from their surrounding controls or labels, so no redundant alt-text edits were required.

  2. SEO and AI-search playbook 1.2.0 published and verified

    Published the reviewed editorial playbook with buyer-benefit introduction, Markdown-first download, matching standalone copy and packaged skill, preserved strategy-library and checklist entries, and production checks across desktop, mobile, links, downloads, imagery, accessibility, and all 28 sitemap URLs. Measurement access remains limited: first-party analytics returned an access denial, the requested Semrush route was unavailable, and the available Ahrefs evidence was a five-row sample rather than a full-domain backlink conclusion. These gaps remain Not available, not zero, and do not establish traffic, ranking, citation, lead, or revenue outcomes.

  3. Ongoing strategy and repeat-cycle method added

    Reframed this page as Markethink's living SEO and AIO implementation strategy. Kept the first 30 days as an initial foundation phase, preserved all 19 actions, IDs, evidence, history, day-60/90 checkpoints, and status counts, and added a repeat process for evidence review, capacity-aware priorities, implementation, verification, lessons, carry-forward work, and newly scoped opportunities. Updated the reusable visitor playbook to version 1.1.0 with a next-cycle backlog and separate delivery and results review.

  4. AI Marketing Strategy Library and first playbook launched

    Published the AI Marketing Strategy Library with a complete 30-day SEO and AI-search playbook, a standalone agent brief, a self-contained Markdown download, and a portable Agent Skills package. Added navigation, sitemap, AI-discovery, and bounded contextual links while preserving all 19 roadmap actions, their acceptance criteria, and status counts.

  5. V2 B2B editorial image system published and verified

    Published the exact approved P01–P10 editorial system across the statistics resource, blog cards, related cards, and nine article heroes with responsive sources and protected P04 mobile framing. Added ten distinct 1200×630 sharing derivatives across Open Graph, Twitter, and dimensioned ImageObject metadata. Production review confirmed exact card order, decoded assets, useful conceptual alt text, ten canonical routes, absent rejected legacy imagery, contained narrow layouts, and unchanged statistics ledgers, charts, and website palette.

  6. Canonical blog base independently closed

    Independent production review closed the final canonical-base cleanup at 19:05 on September 12, 2026. The blog base uses #F5F5F1 to #EDF4EA with the intended green treatment, while the protected research values, source links, and earlier verification evidence remain unchanged.

  7. Palette correction published and independently checked

    Published the canonical palette correction. Independent verification confirmed the green blog and article treatment, contained desktop and 320px views, CAPTCHA behavior, IAB SVG canonical colors with the 82/45/37 values preserved, byte-identical CSV and JSON hashes, and all 60 statistic anchors. Palette work remains In progress until the final canonical-base cleanup is independently closed.

  8. Initial public roadmap published

    Published the public 30-day plan, recorded the public statistics and blog evidence as pre-existing progress, and kept the B2B imagery and palette correction workstreams marked In progress rather than live.

USE THE PLAYBOOK

Bring us the search and pipeline gap you need to understand.

We will map the buyer question, page ownership, evidence, measurement, and next decision before recommending more content.

Request available times

The existing form requests available times. It does not book a meeting.