Home Services Book
§
§ · journal

Common mistakes ecommerce SEO specialists make.

Seven operational confessions from a 2,000-store Shopify Plus partner agency. Each mistake gets a one-line cost, a one-line recovery move, and a 90-day reset playbook for the team that already shipped them.

Book my 30-min SEO audit call Jump to the seven mistakes
§ 01 · TL;DR

Seven mistakes. One 90-day reset.

The seven mistakes ecommerce SEO specialists make most often, ranked by audit frequency across roughly 220 stores we have reviewed since 2020: treating keyword research as a one-time event, confusing indexable URLs for ranking URLs, letting faceted navigation eat crawl budget, dismissing Core Web Vitals as a tiebreaker, conflating product-page intent with collection-page intent, mistaking on-page polish for off-page authority, and reporting on organic sessions instead of organic-revenue. Each one carries a measurable cost - three weeks of lost rankings, $4,000 of wasted dev hours, 22 percent of organic ceiling left on the table - and each one has a recovery move that fits inside a 90-day reset. The rest of this piece walks through every mistake with the operational confession, the cost in dollars and weeks, and the recovery move. We have shipped this reset on Shopify Plus stores like Emani, on B2B catalogs like Big Game Sports, and on multi-region retail like Noble Paris. The pattern is the same on every platform; what changes is which mistake the previous specialist concentrated on.

§ 02 · why these mistakes are systemic

The mistakes are structural, not personal.

An honest framing before the seven confessions. The mistakes below are not the result of bad ecommerce SEO specialists. They are the result of incentives, defaults, and platform constraints that push competent specialists toward predictable failure modes. We have made every one of them ourselves at least once. The recovery playbook is the same whether the work was done by an in-house team, a freelancer, a retainer agency, or our own team a few years ago when we were still learning what an ecommerce SEO program looks like at the 2,000-store mark.

Three structural reasons the mistakes recur. First, ecommerce SEO sits at the intersection of three disciplines that rarely live in one person's head: technical SEO (crawl budget, schema, indexation), content SEO (collection pages, blog posts, internal linking), and analytics (revenue attribution, conversion-rate analysis, cohort behavior). A specialist deep in one is usually shallow in the other two. The mistakes cluster around the disciplines the specialist is weaker in.

Second, ecommerce platforms ship sensible-but-wrong defaults. Shopify ships canonical tags on faceted-search URLs that point to the URL itself rather than the parent collection. BigCommerce indexes ?page=2 by default. Magento on a stock install renders pagination URLs that Google Panda would have flagged in 2011 and that the modern helpful-content systems still penalize. A specialist trusting the platform default is making a reasonable choice that produces a wrong outcome.

Third, the reporting cadence on most retainers is monthly, while the consequences of SEO mistakes are visible at 90 to 180 days. The feedback loop is too slow to course-correct in real time, so the specialist keeps shipping work in the same direction for two quarters before anyone notices the organic-revenue line is flat. The mistakes below all have the same recovery shape: identify the gap, ship the fix in two to four weeks, monitor for 90 days, document what changed.

Read the seven sections that follow as a checklist for auditing your current SEO program - whether you run it in-house, with a freelancer, with an ecommerce SEO specialist, or with a retainer agency. The companion piece on Shopify SEO services covers what good looks like in the buy-side framework; this one covers what to look out for in the work that has already shipped.

§ 03 · seven mistakes

Seven confessions. One pattern.

Every section follows the same shape: the mistake itself, the operational confession (something we shipped or audited that broke), the dollar-and-week cost, and the recovery move that fits inside a 90-day reset window.

§ 04 · mistake 01

Treating keyword research as a one-time event.

The confession. On a 2023 audit for a US-based home goods brand we now reference internally as the "stale-map" project, the in-house SEO specialist had built a 280-keyword research file at the start of the engagement, mapped it to 47 collection pages and 130 product pages, and then never re-ran the research. Eighteen months later we pulled their Search Console export and found that 31 percent of the top-100 ranking keywords were terms that did not appear anywhere in the original keyword file. The store was ranking accidentally, on queries the specialist had never targeted, while the targeted queries had drifted as the market evolved.

The cost. 11 weeks of unrecovered rankings on the targeted queries that drifted, roughly $84,000 of foregone organic revenue across the audit period, and the cost of the engagement itself which was running at $4,500 per month against a target the work had stopped serving. The specialist had not done bad work; they had done correct work against an outdated map.

The recovery. Refresh the keyword research on a 90-day cadence, not annually. The mechanical version: pull a fresh Search Console export at days 0, 90, 180, 270, and 360. Compare each export against the prior period to identify which queries gained or lost impressions, which queries are new (often emerging product categories or seasonal modifiers), and which queries on the original map have effectively disappeared. Re-mapping the top 50 keywords against current commercial-intent pages takes 6 to 10 hours per quarter and recovers most of the drift. We documented this process in our work with Emani, where the quarterly refresh caught a meaningful seasonal modifier (summer-edit collections) that the original keyword file had not anticipated. Search Console is the only tool you need for the data; Ahrefs or Semrush add competitor context. The hard part is the discipline of running it, not the tooling.

§ 05 · mistake 02

Misreading "indexable" versus "ranking".

The confession. Early 2022 we took over an SEO program from a previous freelancer for a US apparel brand. The freelancer's monthly report celebrated that indexed URLs had grown from 1,200 to 11,400 across the prior 14 months, framing this as evidence the SEO program was working. The reality on the ground: 9,800 of the 11,400 indexed URLs were faceted-navigation combinations (color + size + price), internal-search results, and pagination pages beyond page 2, none of which ranked for anything commercially valuable. The actual ranking-URL count was 880 - barely above the count from 14 months earlier. The freelancer had genuinely confused indexable surface area with ranking authority and was reporting against the wrong metric.

The cost. Six weeks of crawl budget wasted on URLs Google was never going to rank, $7,200 of dev time the previous engagement had spent generating and "optimizing" the parameter combinations, and a roughly 18 percent ceiling on organic revenue because the topical authority that should have concentrated on the parent collections was diluted across thousands of thin parameter URLs. Google's own crawling guidance covers the basic principle; the freelancer either had not read it or had read it and convinced themselves their case was different.

The recovery. Three weeks of work to re-establish indexation discipline. Week one: inventory every URL pattern that exists on the property using Screaming Frog, classify each by intent (commercial, transactional, navigational, parameter combination, internal search, pagination), and document which patterns should be indexed versus excluded. Week two: ship the canonical-to-parent rules on faceted-search combinations beyond the first two parameters, the noindex meta tags on internal-search results, and the robots.txt rules that prevent crawl on the lowest-value patterns. Week three: monitor Search Console for the de-indexation curve and confirm the impressions concentrate on the parent pages where they should have lived all along. The reporting cadence after the recovery: track ranking URLs (URLs with at least one impression in the last 28 days) as the headline metric, and treat the indexed-URLs total as a diagnostic only.

§ 06 · mistake 03

Faceted nav as crawl budget thief.

The confession. The mistake above lives at the indexation layer; this one lives a level deeper, at crawl budget. On a 2024 audit for a B2B-and-DTC sports retailer (similar shape to Big Game Sports), we pulled the Crawl Stats report from Search Console and found that Googlebot was spending 71 percent of its crawl budget on faceted-navigation URLs that the previous SEO specialist had already noindexed. Noindex does not stop the crawl; it stops the indexation. Every crawl on a noindexed URL is a crawl that did not happen on a fresh product page or an updated collection. Critical pages were being re-crawled every 38 days instead of every 8 days, and rankings were lagging because the search engine was working with stale snapshots of the most-important commercial pages.

The cost. Roughly 9 weeks of stale-content lag on top-revenue collection pages, $0 in direct dev cost (the noindex rules were already there) but a 12 to 15 percent organic-revenue ceiling because the rankings were always one update cycle behind reality. The compound cost was higher: every product launch and every new collection took an extra month to reach steady-state rankings.

The recovery. The fix is the robots.txt block, not the noindex meta tag. Two weeks of work. Week one: identify every URL pattern that should not be crawled at all (faceted combinations beyond the first two parameters, internal-search results, sort-order parameters, tracking parameters that survive into Google's index), document them in a robots.txt update, and verify in Google's robots.txt tester. Week two: deploy the robots rules, monitor the Crawl Stats report for the redistribution of crawl budget, confirm that critical pages are being re-crawled at the expected cadence (every 5 to 12 days for revenue pages on most stores). On Shopify, the platform-specific reality is that robots.txt customization shipped in 2021; before then SEO specialists had no choice but to rely on noindex tags. After 2021, there is no excuse.

§ 07 · mistake 04

Skipping site speed because "Core Web Vitals are a tiebreaker".

The confession. Mid-2023 we inherited an SEO program where the previous specialist had documented Core Web Vitals issues across 38 of the top-50 templates and then deprioritized the fix because, in their report, "Core Web Vitals are a tiebreaker, not a major ranking factor." The store sat at 3.1-second LCP on mobile and 312-millisecond INP at the 75th percentile. Both metrics missed the Google thresholds by 25 to 60 percent. On commercial-intent queries, where the search engine is choosing among hundreds of relevance-equal candidates, the tiebreaker compounds. The store was losing ranking tiebreakers on roughly 22 percent of available commercial-keyword positions to faster competitors.

The cost. 14 weeks of underperformance against direct competitors who shipped Core Web Vitals improvements in the same period, an estimated $140,000 of foregone organic revenue across the audit window, and a higher paid-acquisition cost-per-acquisition because the store was leaning harder on paid traffic to compensate for the organic shortfall.

The recovery. Image weight first, font-loading second, render-blocking script audit third, in that order. Week one: audit image weight on the top 50 templates using the Shopify image_url filter or the equivalent in your platform; the typical reduction is 60 percent of asset weight in 4 to 8 hours of theme work. Week two: font-loading strategy - switch to system fonts where the brand allows, use font-display: swap, subset the custom faces. Week three: render-blocking script audit - defer non-critical JavaScript, lazy-load third-party tags below the fold, audit and remove apps that ship blocking script tags. The combined LCP improvement on most stores after a three-week sprint is 40 to 55 percent, taking a 3.1-second LCP to 1.4 to 1.7 seconds. INP follows a similar curve once the third-party tag audit ships. web.dev/vitals documents the field-data thresholds; web.dev/inp covers the INP-specific guidance that replaced FID in 2024.

§ 08 · mistake 05

Conflating product-page intent with collection-page intent.

The confession. A 2022 engagement with a multi-region cosmetics brand (similar to the work we documented on Noble Paris) had shipped 1,800-word product-page descriptions on every PDP, including 90-second video and a comparison table to three competitor products. The collection pages, meanwhile, were rendered as bare product grids with a 60-word category description above the fold. The keyword strategy was entirely backwards. PDP queries - branded product names with size, color, or variant modifiers - convert because the searcher already wants the product; they do not need 1,800 words to make the decision. Collection queries - buyer-investigation phrases like best red lipstick for olive skin or matte foundation for combination skin - convert when the page reads like a topical hub with buying-guide content, comparison content, and editorial framing. The previous specialist had over-invested on the wrong template and under-invested on the right one.

The cost. $62,000 of writer fees on PDP descriptions that did not need to exist at that length, 22 weeks of the engagement spent on the wrong template, and a roughly 28 percent revenue ceiling because the collection pages - which capture commercial-investigation intent - were ranking on page 3 against competitors whose collection pages read like authoritative buying guides.

The recovery. Six weeks of content work. Weeks one to two: identify the top 20 collection pages by impressions and traffic potential using Search Console, build a topical-hub brief for each (1,200 to 1,800 words covering buyer-investigation queries, 3 to 5 outbound authority links, internal links to the most-relevant products, and a clear progression from informational to transactional intent). Weeks three to five: ship the rewrites against the brief. Week six: trim the over-long PDP descriptions to 400 to 700 words covering differentiated specs, real customer review excerpts, and clear shipping and return information. The asymmetry matters because Product schema on PDPs already carries most of the structured-data weight; the on-page text only needs to support the schema, not duplicate it.

§ 09 · mistake 06

Confusing on-page polish for off-page authority.

The confession. Late 2023 we inherited an SEO retainer where the previous specialist had spent 11 months perfecting on-page implementation. Title tags rewritten on every page, meta descriptions hand-written on every page, schema validated on every template, internal-linking structure documented in a 47-page audit deliverable. The work was good. The problem was that the store was sitting on a Domain Authority of 24 (per Ahrefs) against direct competitors at 38 to 52, and no amount of on-page polish closes that gap. The retainer was producing a beautifully on-page-optimized store that could not rank for any keyword above category-tail volume because it lacked the link-equity to compete.

The cost. 11 months of retainer fees, roughly $33,000, against an unmovable ceiling. The on-page work itself had real value but it was being asked to do a job that on-page work cannot do. The opportunity cost was higher than the direct cost: the link-building program that should have run in parallel did not exist.

The recovery. A 90-day off-site work program that runs in parallel with whatever on-page maintenance the team is already shipping. Weeks one to four: identify 10 to 20 high-quality outbound link opportunities - credible publications in the category, partner directory listings, podcast appearances, guest content on aligned domains, digital PR moments tied to product launches or category insight reports. Weeks five to eight: pitch the relationships, ship the supporting assets, place the first 3 to 5 links. Weeks nine to twelve: track the link velocity in Ahrefs or Semrush, monitor for unnatural-link patterns that Google's spam policies would penalize, document the placement quality. The link program is slow and editorial; it is not a sprint. But on a store with a Domain Authority gap to competitors, it is the only thing that moves the rankings ceiling. On-page work shipped without an off-site program is cosmetics on a problem the cosmetics cannot solve.

§ 10 · mistake 07

Reporting on organic sessions, not organic-revenue.

The confession. The most expensive mistake on the list, and the one we have made ourselves. On a 2021 engagement we ran for a US lifestyle brand, our monthly retainer report celebrated 35 percent growth in organic sessions across the year. The client signed off, the engagement renewed, and the work continued. Eighteen months in, the head of growth ran a contribution-margin analysis and found that organic revenue had grown only 4 percent across the same period. We had been driving traffic to informational queries - gift-guide content, how-to articles, category explainers - that rarely converted, while ignoring the commercial-intent collection pages where 80 percent of revenue lived. The session metric was real; the revenue impact was not.

The cost. Twelve months of an engagement that was reporting against the wrong metric, roughly $54,000 of retainer fees, and a 12-month gap in the commercial-page work that should have been the priority. The brand had paid for 12 months of session growth that did not pay for itself in revenue. We refunded a portion of the engagement and rebuilt the program around a revenue-first dashboard.

The recovery. A single-page monthly dashboard that tracks four numbers. One: organic-revenue (from analytics, attributed to the organic channel, compared against the prior 90-day baseline). Two: organic-conversion-rate (how organic sessions convert versus paid and direct, on the same period). Three: top-20 commercial-keyword position (the keywords that, by impressions and revenue potential, should be carrying the program). Four: Core Web Vitals at the 75th percentile (the diagnostic that explains a meaningful share of ranking variance). Sessions are a useful diagnostic when revenue moves; sessions are vanity when revenue does not. The dashboard fits on one slide, it takes 90 minutes to build the first time and 20 minutes per month to refresh, and it is the difference between an engagement that is doing the right work and an engagement that looks like it is.

§ 11 · the 90-day reset

Ninety days. Four phases. One revenue dashboard.

The recovery playbook we run on every audit-and-reset engagement, regardless of which of the seven mistakes the previous specialist concentrated on. The phases are sequential; skipping a phase compounds the debt rather than reducing it.

01

Days 1 to 14 - diagnostic and baseline

Pull the Search Console export, pull the analytics export, run a Screaming Frog crawl of the property, audit the indexed-URLs versus ranking-URLs gap, document the Core Web Vitals at the 75th percentile by template. The deliverable is a single spreadsheet with the gap inventory, the technical-debt list, and the baseline metrics that the rest of the program is measured against. Two weeks because a thorough crawl on a 5,000-product catalog takes longer than most agencies budget; the rushed version misses the long-tail patterns that drive 15 to 30 percent of organic revenue.

Output: gap inventory + technical-debt list + 90-day baseline

02

Days 15 to 45 - technical fixes

Ship the high-impact technical-SEO fixes in the order that compounds fastest: image weight first (highest LCP impact), canonical and noindex rules on faceted nav (recovers crawl budget), schema parity across product and collection pages, robots.txt updates for the lowest-value patterns, redirect map for any orphaned URLs from prior site changes. Most fixes are 4 to 12 hours of focused work; the discipline is shipping them rather than logging them in a backlog deliverable. Rich Results Test validates the schema work; Search Console validates the indexation work.

Output: technical-debt list closed to under 10 percent of intake count

03

Days 46 to 75 - content and intent fixes

Rewrite the top 20 collection pages with topical-hub treatment (1,200 to 1,800 words covering commercial-investigation intent), refresh the keyword map against current Search Console data, fix the internal-linking gaps from homepage and top collections to revenue pages, trim over-long product descriptions to 400 to 700 words, audit the schema depth on PDPs against schema.org/Product, validate hreflang on multi-region stores. The collection-page rewrites are the highest-impact single deliverable in this phase; the rest are supporting work that compounds slower.

Output: rewritten collection hubs + refreshed keyword map + internal-linking improvements

04

Days 76 to 90 - dashboard and handover

Stand up the organic-revenue dashboard (organic-revenue, organic-conversion-rate, top-20 commercial-keyword position, 75th-percentile Core Web Vitals), document what shipped and what is still in the backlog, train the in-house team or transition to a maintenance retainer. Day 90 is the formal handover. The recovery is not the same as a maintenance program; the work continues monthly after the reset, but the bulk of the gap-closing has already shipped. Track organic-revenue weekly for the first 90 days post-reset; most stores see a 1.4x to 2.1x recovery against the pre-reset baseline by month six.

Output: live dashboard + backlog documentation + handover

§ 12 · summary table

Seven mistakes. Honest mid-points on cost and recovery.

Mistake Typical cost Recovery time
Keyword research as one-time event15-30% revenue ceiling6-10 hrs / quarter
Indexable confused with ranking18% revenue ceiling3 weeks
Faceted nav as crawl thief12-15% revenue ceiling2 weeks
Skipping site speed work22% commercial-position loss3 weeks
Product-vs-collection intent confusion28% revenue ceiling6 weeks
On-page polish, no off-page programDA gap unmovable12 weeks (slow)
Sessions reporting, not revenue$54K of misdirected work90 minutes (build dashboard)

Mid-points across roughly 220 audits since 2020. Tails exist on either side; the median is what matters for budgeting the recovery.

§ 13 · what good looks like

The opposite of these seven mistakes.

Read this section as the inverse checklist. If your current ecommerce SEO program is producing the seven outputs below, the seven mistakes are not your problem. If it is not, you have at least one of them.

One. A keyword map that is dated within the last 90 days and has been refreshed at least three times across the last 12 months. The map is a living document, not a deliverable.

Two. Search Console data shows ranking URLs as the headline metric, indexed URLs as a diagnostic only. The ratio is in the 75 to 90 percent range - most indexed URLs are also ranking for something.

Three. Crawl Stats report shows Googlebot spending more than 70 percent of its crawl budget on the templates that drive 80 percent of organic revenue. The faceted-navigation, internal-search, and parameter URLs are blocked at the robots.txt layer, not just at the noindex layer.

Four. Core Web Vitals at the 75th percentile pass on at least 80 percent of templates. LCP under 2.5 seconds, INP under 200ms, CLS under 0.1, all measured on field data not lab data.

Five. The top 20 collection pages read like authoritative buying guides covering commercial-investigation intent. Product pages emphasize differentiated specs, real customer review excerpts, and structured-data depth. The two templates do different work.

Six. The off-site work is shipped on a documented cadence -1 to 3 high-quality editorial placements per month, partner directory hygiene maintained, digital-PR moments tied to product launches and category insight reports. Domain Authority moves up by 2 to 5 points per year as the link program compounds.

Seven. The monthly report leads with organic-revenue, not organic sessions. The dashboard tracks the four numbers that matter and treats sessions as a diagnostic when revenue moves. The team can name which keywords moved on which dates, and which on-page or off-page work drove the move.

The companion top skills every ecommerce SEO specialist needs covers the buyer's framework for hiring the work in the first place - what to look for when evaluating candidates or agencies. Read the two together: this piece is the audit lens, the sibling piece is the procurement lens. Both link out to our broader Shopify SEO services overview and the migration playbook for context on the engagements where these mistakes most often show up.

§ 14 · about the author

Honest because we have shipped them.

Written by Prasun Anand, founder and lead engineer at Digital Heroes. Eight years on Shopify, twelve years across ecommerce platforms (Magento, BigCommerce, WooCommerce, custom builds before that), 2,000-plus stores shipped since 2017, Trustpilot 4.9 across 70-plus reviews, Premier Shopify Plus partner agency. NY and Delhi headquartered. The engagements referenced anonymously in this piece are real - names withheld to protect the brands, with explicit references to Emani, Big Game Sports, and Noble Paris drawn from work where the brands have given written permission to share specifics.

We have made every one of the seven mistakes ourselves at least once across the last decade. The recovery playbook is what we wish someone had handed us in 2018. If you are reading this and recognizing your current program in any of the seven sections, the right next step is the diagnostic phase - pull the Search Console export, pull the analytics export, run the audit. Whether you do that yourself, with your in-house team, or with us, the diagnosis is the entry point and the rest of the recovery follows from there.

§ 15 · questions auditors ask

Six honest answers.

What is the single most common mistake an ecommerce SEO specialist makes?

Treating keyword research as a one-time event. The pattern we see on roughly seven of every ten audits is a 12-month-old keyword map written during the original engagement, never refreshed against current Search Console data, and gradually drifting out of alignment with what the store actually ranks for and what users actually search. Search behavior on ecommerce queries shifts inside 90 days as new product categories emerge, brand names get added to the long tail, and seasonal modifiers come and go. A keyword map that has not been refreshed in six months is a map of a market that no longer exists. The right cadence is a 30-day Search Console crawl, a 90-day full keyword refresh against the current commercial-intent inventory, and a hard rule that any new collection page or product launch generates its own keyword research at brief stage rather than inheriting whatever the existing taxonomy assumed. The fix is a $0 line item that sits inside the in-house SEO team's monthly cadence; the cost of skipping it is roughly 15 to 30 percent of organic-revenue ceiling on a well-trafficked store.

How can I tell if my ecommerce SEO specialist confused indexable URLs with ranking URLs?

Three quick checks. First, pull the Indexed URLs total from Google Search Console and compare against the Ranking URLs total (URLs that have at least one impression in the last 28 days). On a healthy ecommerce store the gap is roughly 10 to 25 percent. On stores where the SEO specialist confused the two metrics, the indexed-URLs total is two to five times the ranking-URLs total because thousands of faceted-navigation URLs, internal-search results, and pagination pages are indexed but rank for nothing. Second, run a site:yourdomain.com query and look for URL parameters in the results - ?sort=, ?filter=, ?page=2, ?q=. Each one is wasted crawl budget and dilution of topical authority. Third, check robots.txt and the meta robots tags on faceted-nav templates. If the canonical points to itself rather than the parent category, and there is no noindex on the parameter combinations, the indexable surface area is uncontrolled. The recovery is roughly two weeks of work to add canonical-to-parent rules, noindex on parameter combos beyond the first two, and a robots.txt block on internal-search results URLs.

Are Core Web Vitals actually a ranking factor for ecommerce SEO?

Yes, but the framing most SEO specialists use is wrong. Google's documentation describes Core Web Vitals as part of the page-experience signal, and the public guidance has been that page experience is a tiebreaker between equally-relevant results. The honest reality on ecommerce is that on commercial-intent queries - buyer keywords like best running shoes for flat feet or queen size mattress under 1000 - the search engine has hundreds of equally-relevant ranking candidates. Tiebreakers compound. A store with LCP at 1.8 seconds beats a store with LCP at 3.2 seconds on most of those queries even when the slower store has marginally better content. The 75th-percentile LCP target is 2.5 seconds, INP target is 200 milliseconds, CLS target is 0.1. Stores that miss those thresholds on mobile against a competitor that hits them lose roughly 8 to 22 percent of available rankings on commercial keywords. The fix is image weight (the highest-impact lever, usually a 60 percent reduction in afternoon's work), font-loading strategy, and render-blocking script audit. None of it requires custom engineering on Shopify, BigCommerce, or Magento; it requires the SEO specialist actually shipping the fixes rather than logging them as recommendations.

What is the most common reporting mistake ecommerce SEO specialists make?

Reporting on organic sessions instead of organic-revenue. Sessions are vanity metric one and ecommerce SEO has at least three more important numbers: organic-revenue (the actual money attributed to organic search), organic-conversion-rate (how well organic traffic converts versus paid and direct), and revenue-per-organic-session (which collapses the two into a single comparable number across periods). A retainer that sends a monthly slide saying organic sessions are up 22 percent without naming the revenue impact is hiding the answer to the question that matters. We have audited stores where organic sessions grew 35 percent across a 12-month engagement and organic revenue grew 4 percent because the SEO specialist drove traffic to informational queries that rarely converted while ignoring the commercial-intent collection pages where 80 percent of revenue lived. The fix is a single-page monthly dashboard that tracks organic-revenue, organic-conversion-rate, top-20 commercial-keyword position, and Core Web Vitals against the prior 90-day baseline. If your SEO specialist cannot produce that dashboard, the engagement is reporting against the wrong metric.

Should ecommerce SEO specialists treat collection pages and product pages the same way?

No, and the assumption that they should is the fifth-most-common mistake we audit. Collection pages serve commercial-investigation queries - running shoes, queen size mattress, blue area rug - where the searcher is comparing options. Product pages serve transactional queries - Brooks Ghost 16 size 10 mens - where the searcher knows what they want and is checking price, stock, and shipping. The keyword targets are different, the on-page templates should be different, and the internal-linking strategy should reflect both. Most SEO specialists write the same generic 200-word block for every collection page and the same template description for every product page, then wonder why the store ranks below competitors who treat the two intent layers separately. The fix is a 1,200 to 1,800-word topical-hub treatment on the top 20 collection pages by traffic potential - covering buying-guide content, comparison content, and outbound links to authority sources - paired with product-page descriptions that emphasize the differentiated specs, real customer review excerpts, and structured-data depth that Google rewards on transactional queries.

How do I recover from an ecommerce SEO specialist's mistakes without rebuilding the site?

A 90-day reset is enough to fix most of what a previous SEO specialist got wrong, without a full replatform or theme rebuild. Days 1 to 14: a Search Console audit that captures the gap between indexed URLs and ranking URLs, a Screaming Frog crawl to inventory the technical-debt surface, a baseline of organic-revenue against the prior 90 days. Days 15 to 45: ship the technical-SEO fixes - canonical and noindex rules on faceted nav, image-weight optimization on the top 100 templates, schema parity across product and collection pages, redirect map for any orphaned URLs from prior site changes. Days 46 to 75: rewrite the top 20 collection pages with topical-hub treatment, refresh the keyword map against current Search Console data, fix the internal-linking gaps from homepage and top collections to revenue pages. Days 76 to 90: stand up the organic-revenue dashboard, document what shipped and what is still in the backlog, hand the engagement back to the in-house team or transition to a maintenance retainer. We have run this 90-day reset across more than 40 stores in the last three years; the typical recovery is 1.4x to 2.1x organic revenue against the pre-reset 90-day baseline by month six.

§ 16 · the next step

Bring the Search Console export. We'll bring the gap inventory.

A 30-minute SEO audit call. Named lead engineer plus SEO lead on the call, not a sales rep. Written gap inventory and 90-day reset scope returned within two business days.

Book my 30-min SEO audit call Shopify development service
related reading