- —
- PROSPECT
- NEW · every 48h
- 3/day · may go to 4 now
- https://valerie-mason.com
Phase 1 — Onboarding & audit
Size up what this account can carry, collect access, and audit what already exists.
- 42
- done
- 2
- skipped
- 44
- in total
Step 1.0 · Potential assessment
- What
- Score the account against three good-fit signals and two red flags, count how many URLs their sitemap yields versus how many the frequency plan needs, then record how much potential it has.
- Where
- In this dashboard. The URL count is automatic — paste the domain and the system reads their sitemap.
- Output
- HIGH POTENTIAL / AVERAGE / CHALLENGING with written reasoning, in organic.client_viability.
Good-fit check
Three signals: more than 5 products or ideas, enough URL volume, existing visual assets. All three is a strong fit, two is workable if the red-flag check is clean.
Three signals that a store will work on Pinterest. Answer each one and say what you saw — the answer is the score, the reasoning is what makes it checkable six months from now.
How it scores. All three yes and the account can carry an ambitious plan. Two is workable if the red-flag check is clean. One or none means Pinterest will be slow going here — which the client can be told now, and that is cheaper than telling them in month four. Every no raises a flag and opens a plan box: this check does not close until each one has an answer.
Are there more than five distinct products or ideas to pin?
The engine needs variety. One product becomes the same pin sixteen times, which Pinterest reads as repetition and suppresses.
Count genuinely distinct items or angles, not colourways of one thing. Collections, guides and editorial pages count.
Does the site have enough distinct URLs to keep a waterfall fed?
Each cycle needs its own URL, and a URL rests for a cooldown period after it runs. Too few pages and the same page comes back before its cooldown clears, competing with its own pins.
Target is more than twenty usable URLs. P1.0.3 counts them from the sitemap automatically — answer this on whether that count clears the bar for the planned frequency.
Do usable visual assets already exist?
If every pin needs shooting from scratch, the first cycle slips by weeks and the cost per cycle changes the commercials entirely.
Check their Instagram, existing campaign shoots, any brand asset drive. Enough for four designs per URL is the practical bar.
Red-flag check
Two signals that hold you back: a single landing page, and a restricted or sensitive niche. One is survivable with the mitigation written down, both together is a decline.
Two things that hold a store back regardless of how good the fit looked. Flag what is true and say what you saw.
How it scores. Neither flag and nothing is holding the account back. Either one is a stop-and-think: the box that opens underneath is where the case gets made for taking the account on anyway, or for telling the client no — and it is what P1.0.4 reads when it asks for the verdict.
Is the whole site a single landing page?
One page means one URL, which means the waterfall has nothing to rotate through and every cycle competes with the last.
Count real, distinct destination pages. Anchors on one page do not count.
Is the niche restricted or sensitive on Pinterest?
Restricted categories get distribution capped or the account actioned. Finding out after building the board architecture wastes the entire Strategy Core.
Check Pinterest's advertising and community guidelines for the category — supplements, CBD, weapons, adult, medical claims, financial advice.
Count URL volume
The system reads the sitemap and counts products, collections and blogs. Under 10 URLs is a red flag: too few to build a waterfall on. 20 is comfortable.
The system reads the sitemap and counts what is there. Record the breakdown so the number can be argued with later, because the frequency plan is built on it.
How it scores. Under 10 usable URLs is a red flag — too few to build a waterfall on. 20 or more is comfortable at two cycles a month.
How many usable URLs does the sitemap yield?
This number sets the maximum sustainable cycle frequency. Everything downstream — the daily pin target, the cooldown, whether the retainer is deliverable — comes off it.
The count runs automatically from the domain. Exclude cart, account, policy and tag pages: usable means a page a pin can sensibly point at.
What is the split, and which pages are the strongest candidates?
Phase 4 selects URLs from this pool every month. Knowing now which pages are seasonal, which are evergreen and which are thin saves that decision being made blind later.
List the categories and call out anything obviously seasonal or obviously weak.
Sitemap yielded 224 URLs; requirement 6 (daily=1, cooldown=60d, waterfall=30d).
Assess account potential
How much room is in this account: high potential, average, or challenging. The client has already signed, so this is not go/no-go — it sets the plan and what they are told about pace. On challenging, say so in writing now.
How much room is there in this account, and what does the client need to hear about pace. Nothing else in phase 1 unblocks until this is recorded.
How much potential does this account have?
HIGH: all three good-fit signals, neither red flag — this one can carry an ambitious plan. AVERAGE: workable, with a named risk to keep an eye on. CHALLENGING: it will take longer, and the client needs to hear that in writing now rather than in month four.
If this goes wrong, what will have caused it?
One sentence — the most likely failure, not every possible one. The risk screen checks against it in month three.
Step 1.1 · Intake & access
- What
- Send the onboarding questionnaire, then collect every access the work depends on: Pinterest, GA4, Search Console, CMS, brand book, content drive, product feed.
- Where
- Questionnaire goes to the client by email. Access requests happen inside each platform. Everything is recorded on the Intake form in this dashboard.
- Output
- Filled organic.client_intake + organic.client_access rows, and every collected document link stored under Assets.
Send onboarding questionnaire
Send the questionnaire to the client. Four sections: contact, business, social presence, Pinterest goals. Maximum two priorities, which forces focus.
The completed questionnaire back from the client — attach the file or paste the link to their response.
GA4 Analyst access
Analyst level is enough. Needed to measure traffic quality, not volume: Pinterest native stays the source of truth for clicks.
Confirmation that Analyst access is granted: which GA4 property, and to which account.
Google Search Console
Look at Performance for queries and pages. Shows where the client already ranks organically. Also visible: how many pins already link to the site.
Access granted, plus the Performance export showing which queries and pages already rank.
Collect brand book
Logo as a transparent PNG, official hex codes, typography. Without this the design brief cannot be correct.
The brand book itself — logo as a transparent PNG, hex codes, typography. Attach the file or the drive link.
Connect content drive
Record the link and assess the quality; two things here decide the whole design route later. White backgrounds only makes the AI route mandatory. Social exports carrying a watermark are penalised by the algorithm, so ask for the originals rather than what is on the profile.
One link covering the client's existing visual material — the content drive plus whatever Instagram and TikTok hold, uncompressed and without watermarks — and one line on what is in it. White backgrounds only means the AI route becomes mandatory, so say so here.
Request Google keyword list
Indicative, not directly usable. Pinterest search behaviour differs from Google, but it gives direction.
The keyword list from the client's SEO or ads side. Attach the export — it is direction for phase 2, not a bank.
Audience document
Existing persona documentation. Feeds the taste graph mapping in phase 2.
The client's existing persona or audience document. Attach it — it feeds the taste graph in phase 2.
Connect product feed
XML or CSV feed URL. Needed for dynamic product pins and to know the catalogue.
The XML or CSV feed URL, and confirmation that it loads.
Step 1.2 · Account audit
- What
- Check whether the domain is blocked, whether pins are flagged, whether boards are structured correctly, whether Rich Pins fire, and capture the analytics baseline over the last three months.
- Where
- flaggedpins.com tools for the block and flag checks, Pinterest Analytics for the baseline, this dashboard for recording results.
- Output
- Every audit task closed with notes, and thirteen baseline KPIs captured for later comparison.
Domain block check
Check whether the domain is blocked. If it is: contact Pinterest Support and reply to their automated rejection to force a human review.
A blocked domain makes every pin pointing at it worthless, so this is the first thing the audit asks.
How it scores. Nothing found, or everything found now fixed — tick it and the task closes. Anything still outstanding: leave it unticked, say what is left and who has it. The task stays open until it is ticked, which is the whole point of the box.
Is the domain unblocked?
While the domain is blocked, every pin we publish is a pin that cannot be clicked. Nothing else in phase 1 is worth doing until this is resolved.
Test the bare domain, not a product URL. The flagged-pin report from P1.2.2 is a per-URL list and does not answer this — a domain can be blocked while individual URLs look fine, and the other way round. A support ticket that is still open is not fixed; the reference goes in the box below.
Whether the domain is blocked. If it is, the support ticket reference.
Flagged pin check
Upload the report. The system translates the status codes into priority. Blacklisted is high, idea pin is low.
The flagged-pin report says which existing pins Pinterest is already suppressing, and how badly.
How it scores. Nothing found, or everything found now fixed — tick it and the task closes. Anything still outstanding: leave it unticked, say what is left and who has it. The task stays open until it is ticked, which is the whole point of the box.
Has every flagged pin been dealt with?
Blacklisted pins drag on the account they sit in. Left in place they suppress the boards we are about to build on.
Work down the report by priority: blacklisted first, idea pins last. Deleted or corrected both count as dealt with.
The flagged-pin report, uploaded. The system turns the status codes into priorities.
Redirect and canonical check
Broken redirects and canonical errors slow distribution down. Fix them in the CMS and always use the final URL.
Broken redirects and wrong canonicals slow distribution on every pin pointing through them.
How it scores. Nothing found, or everything found now fixed — tick it and the task closes. Anything still outstanding: leave it unticked, say what is left and who has it. The task stays open until it is ticked, which is the whole point of the box.
Are the broken redirects and canonical errors fixed?
A pin pointing at a redirect chain loses the signal at every hop, and a wrong canonical hands the credit to another page.
Fixed in the CMS, and the pins repointed at the final URL rather than the one that redirects.
The broken redirects and canonical errors found, and which of them have been fixed.
Find homepage pins
Never pin to the homepage. Pinterest cannot distinguish that URL from other pages. Mind the limit of 150 pin edits per day.
Pinterest cannot tell the homepage apart from any other page, so a homepage pin measures nothing.
How it scores. Nothing found, or everything found now fixed — tick it and the task closes. Anything still outstanding: leave it unticked, say what is left and who has it. The task stays open until it is ticked, which is the whole point of the box.
Do no pins point at the homepage any more?
Every homepage pin is a pin whose performance cannot be attributed to anything, and they are usually the account's oldest and strongest.
Repoint each one at the page it is actually about. Mind the 150 pin edits per day limit — spread over days is fine, half-done is not.
How many pins point at the homepage, and where they now point instead.
Find pins without a URL
The biggest missed opportunity: a pin that goes viral without a link returns nothing. Fix these first.
A pin with no destination is the one loss that cannot be recovered later — the reach happens and returns nothing.
How it scores. Nothing found, or everything found now fixed — tick it and the task closes. Anything still outstanding: leave it unticked, say what is left and who has it. The task stays open until it is ticked, which is the whole point of the box.
Does every pin now carry a destination URL?
This is the biggest single missed opportunity in the audit. A pin that goes viral without a link earns the account nothing at all.
Every pin found gets the URL it should have had. Fix these before anything else in this step.
The pins with no destination, and the URL each one now carries.
Canonical pin ID check
Open one of our pins, take the number out of its URL (pinterest.com/pin/<id>/) and open that URL in a private window. The account it lands on holds the canonical. Repeat for five to ten of the best performers. If another brand holds it, the clicks earned by our image land on their site — report it through Pinterest's copyright form and record the reference.
The canonical pin is the one Pinterest treats as the original. If another brand holds it, our traffic goes to them.
How it scores. Nothing found, or everything found now fixed — tick it and the task closes. Anything still outstanding: leave it unticked, say what is left and who has it. The task stays open until it is ticked, which is the whole point of the box.
Do we hold the canonical pin on the pins that were checked?
Someone else holding the canonical means the clicks earned by our image land on their site, and nothing in our reporting shows it.
Open one of our pins and take the number out of its URL (pinterest.com/pin/<id>/). Open that same URL in a private window: the account it lands on is the canonical holder. Repeat for five to ten of the best-performing pins. Ours: fine. Another brand: report it through Pinterest's copyright form and put the reference below. A report sent and unanswered is not fixed.
Whether another brand holds the canonical pin, with the pin URL you checked.
Board architecture audit
Rename cryptic board names to exact parent interests. ''Dream Vibe'' becomes ''Modern Luxury Living Room''. The algorithm does not read poetry.
Board names are read by the algorithm as interests, not as brand voice. A cryptic name is a board with no context.
How it scores. Nothing found, or everything found now fixed — tick it and the task closes. Anything still outstanding: leave it unticked, say what is left and who has it. The task stays open until it is ticked, which is the whole point of the box.
Has every cryptic board name been renamed?
"Dream Vibe" tells Pinterest nothing about who should see it. The exact parent interest is what gets the board distributed at all.
Renamed live on Pinterest, not just written down. The rename list belongs in the work panel; this box is whether it has been applied.
The renames: each cryptic board name against the exact parent interest it becomes.
Inventory group boards
Outdated for brands. They dilute context with pins from other accounts. Leave or archive them.
Group boards pull in pins from other accounts and dilute the context of everything around them.
How it scores. Nothing found, or everything found now fixed — tick it and the task closes. Anything still outstanding: leave it unticked, say what is left and who has it. The task stays open until it is ticked, which is the whole point of the box.
Has every group board been left in place or archived deliberately?
The dangerous state is neither: a group board nobody has looked at, quietly mixing someone else's pins into our topic signal.
A decision per board, not a blanket one. Archived, or left with a reason recorded.
The group boards found, and whether each was left in place or archived.
Boards under 10 pins
A public board with too few pins gives no context. Set it back to secret until it has 10 to 15.
A public board with a handful of pins gives Pinterest too little to work out what it is about.
How it scores. Nothing found, or everything found now fixed — tick it and the task closes. Anything still outstanding: leave it unticked, say what is left and who has it. The task stays open until it is ticked, which is the whole point of the box.
Stays open. Until “Is every board under ten pins set back to secret?” is ticked, this task does not close itself. Set the status by hand if the remaining work is not ours to do.
Is every board under ten pins set back to secret?
A thin public board dilutes the profile's topic signal, and it stays thin until we fill it — which is phase 3, not now.
Secret until it holds ten to fifteen pins. Boards we are about to fill this week can stay public if you say so below.
What is still open, and who is fixing it?
Most of what this step finds is not ours to fix — it is the client's developer, their content person, or Pinterest support. Named and dated, it is a thing somebody chases; unnamed, it is a task sitting at IN_PROGRESS that nobody can act on.
Name what is left, who owns it, and when it was last chased. If the answer is that it will not be fixed, say that and say what we do instead — that is a decision and it closes the loop.
The boards under ten pins, and which were set back to secret.
Rich Pins and Open Graph
Paste a product URL into the Pin Builder. If title and price appear automatically, it works. If not: follow the per-CMS fix from the SOP.
Rich Pins pull title, price and availability from the page automatically. Without them every pin carries whatever was typed by hand.
How it scores. Nothing found, or everything found now fixed — tick it and the task closes. Anything still outstanding: leave it unticked, say what is left and who has it. The task stays open until it is ticked, which is the whole point of the box.
Do Rich Pins pull title and price through in the Pin Builder?
Rich pins update themselves when the page changes. Hand-typed pins go stale the moment a price moves, and a wrong price is a lost click.
Paste a product URL into the Pin Builder and look. If nothing appears, the Open Graph tags are missing — the per-CMS fix is in the SOP.
Whether title and price appear automatically in the Pin Builder. Attach the screenshot.
Determine inactivity period
Determines the account class. Longer than 6 months silent, or never posted, means NEW with 48-hour spacing, even on an old account. Record the number of months and the class here, and set the class on the client settings — the answers on this task do not write it there.
The inactivity period decides the account class, which decides the pin spacing for every cycle after this.
How it scores. Nothing found, or everything found now fixed — tick it and the task closes. Anything still outstanding: leave it unticked, say what is left and who has it. The task stays open until it is ticked, which is the whole point of the box.
How many months has the profile been silent?
This is the number the account class is derived from. Six months is the line.
The date of the most recent pin on the profile, counted back from today. A profile that has never posted counts as new, not silent.
Which account class does that make it?
NEW is 48-hour spacing; ESTABLISHED is the normal rhythm. Setting this wrong sets the posting pace wrong for the whole engagement.
Silent longer than six months, or never posted: NEW. Otherwise ESTABLISHED. Record it on the client settings as well — this answer does not write it there.
Is the inactivity period recorded and the account class set from it?
Longer than six months silent puts an old account back to NEW with 48-hour spacing. Getting this wrong sets the posting rhythm wrong for the whole engagement.
Recorded on the client settings, not only written in a note. Check the class on the settings screen actually changed.
The inactivity period, and the account class it sets.
Shadowban signals
Single-digit impressions despite history. Also: an account moved back to a less valuable audience after long inactivity.
A shadowbanned account publishes into nothing, and every number after it is a number about nothing.
How it scores. Nothing found, or everything found now fixed — tick it and the task closes. Anything still outstanding: leave it unticked, say what is left and who has it. The task stays open until it is ticked, which is the whole point of the box.
Are the shadowban signals cleared, or confirmed absent?
Building a Strategy Core on a suppressed account wastes the whole of phases 2 and 3, and the symptom — single-digit impressions — looks like a slow start.
Absent is a real answer and the common one. If signals are there, the recovery route is what goes in the box, not the observation.
The shadowban signals found — or a clear statement that there are none.
Analytics baseline
Thirteen KPIs over three months. Filters: Organic, Claimed Domain, Your Pins. Untick realtime data, it is unreliable.
Baseline KPIs (3mo):
Top performing pins
Top 5 to 10 on outbound clicks and saves. Not on impressions: those say nothing about intent. These URLs are your quick wins in month 1.
The top five to ten pins on outbound clicks and on saves, and the URLs behind them. A fresh account that has no pins with clicks or saves yet is a finding in itself — record that and skip the task with it as the reason, rather than leaving it open on numbers that cannot exist.
Annotations of top pins
Paste the pin URL into PinClicks pin stats and read the annotated interests. Candidates for the keyword bank, provided they have volume.
The annotated interests PinClicks returned per pin — candidates for the keyword bank.
Step 1.3 · Technical setup
- What
- Business account, claimed domain, Pinterest tag, catalogue connection, shopping recommendations off, site speed, image filenames, meta descriptions, save buttons, mobile experience, and the content bank audit.
- Where
- Pinterest settings, the client's CMS, and this dashboard for recording.
- Output
- A technically clean account with the content bank counted per funnel stage.
Business account
No business account means no analytics. You can convert in the settings.
Without a business account there is no analytics, no claimed domain and no catalogue. Everything else in this step assumes it.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Is this a Pinterest business account?
A personal account has no analytics and cannot claim a domain, so the whole of phase 1 step 3 is unavailable on one.
A business account shows an Analytics tab and a Business hub. Convert from Settings → Account management → Convert to business.
Social claimed check
Confirm Shopify and Instagram are connected under Settings → Claimed accounts. Instagram auto-publish MUST be OFF: Pinterest re-posts otherwise get flagged as spam. Shopify claim unlocks catalog + rich pins; without it product pins never carry price. Needs the account owner's login — with partner access only, this is a request to the client. Record that in the box and skip; it is not something the organic operator can do.
Claimed social accounts unlock the catalogue and rich pins. One of these settings actively causes harm if left wrong.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Is the shop connected under claimed accounts?
The Shopify claim is what unlocks the catalogue and rich pins. Without it product pins never carry a price.
Settings → Claimed accounts.
Is Instagram connected?
It attributes saves of their Instagram content back to the Pinterest profile.
Settings → Claimed accounts.
Is Instagram auto-publish switched OFF?
This one matters more than the other two. Auto-reposted Instagram content gets flagged as spam and the whole account carries the penalty.
On the Instagram claim, the auto-publish toggle must be off. Check it even if someone says it already is.
What is not right, and what has to happen?
A failed check on its own is a dead end — the next person sees a red mark and has to redo the whole investigation to find out what it meant.
Name the specific thing that is wrong, who has to fix it (us, the client, their developer), and whether anything downstream is blocked until it is.
Re-optimise existing top pins
Only applies to an account that already published organically — a fresh account has nothing to re-optimise, and skipping it with that reason is the right answer. For pins with impressions but no destination URL, add the correct link: those are the highest ROI edits. Then fix any pins with generic titles/descriptions on high-impression board pages. Cap yourself at 10 to 20 edits per day. Hard platform limit: 150 pin edits per day per account. Above that Pinterest rate-limits the whole account.
Pins that already have impressions are the highest-return edits available — the distribution exists, only the destination is missing. This is a takeover task: an account that has never published organically has nothing to re-optimise, and skipping it with that reason is the right answer rather than three empty ticks.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Do the pins with impressions but no destination URL now have the correct link?
These are impressions already being earned and thrown away. Nothing else in phase 1 pays back this fast.
Filter existing pins by impressions with an empty link, and add the right URL to each.
Has generic copy on the high-impression pins been rewritten?
A pin titled with the file name ranks for nothing, however many impressions it happens to have.
Work down the boards by impressions and rewrite titles and descriptions with real keywords.
Did you stay under the daily edit cap?
Above 150 pin edits in a day Pinterest rate-limits the whole account, which costs more than the edits gained.
Ten to twenty a day is the working pace. 150 is the hard platform limit — do not go near it.
What is not right, and what has to happen?
A failed check on its own is a dead end — the next person sees a red mark and has to redo the whole investigation to find out what it meant.
Name the specific thing that is wrong, who has to fix it (us, the client, their developer), and whether anything downstream is blocked until it is.
Verified Merchant status
Verified Merchant is a Pinterest badge that unlocks product tagging, catalog boosts and shop tab prominence. Requires: business account, claimed domain, active shopping catalog, no policy violations, and passing a Pinterest review. Apply from Settings → Verified Merchant. Track status per client; some verticals require additional review.
Verified Merchant unlocks product tagging, catalogue boosts and the shop tab. It is a review, so it is applied for once the prerequisites are actually met.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Are all the prerequisites met — business account, claimed domain, active catalogue, no policy violations?
Applying before these are in place gets a rejection, and a rejected account waits before it can reapply.
Walk back through P1.3.1, P1.3.2 and P1.3.4. All three have to be green first.
Where does the Verified Merchant application stand?
This is the only item in the step whose outcome we do not control, so what gets tracked is the state and the date — not a tick that could mean either.
Settings → Verified Merchant. Some verticals get an extra review round, so 'submitted' can sit for weeks — put the submission or rejection date in the box at the bottom, since that is the date the wait is counted from.
Claim domain
Place the meta tag in the CMS until the checkmark appears. A condition for reliable analytics.
The domain claim is what ties pins back to the site. Without it analytics are partial and rich pins never work.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Does the domain show as claimed, with the checkmark visible?
Unclaimed means no attribution on pins others save from the site, and no rich pin data.
Settings → Claimed accounts. It has to be the shop's own domain (valerie-mason.com), never the myshopify.com address — claiming the wrong one leaves the live site unclaimed and looks identical on this screen. Check the profile username is right while you are there. A checkmark, not 'pending'.
Is the verification tag in the live theme, where a theme update cannot wipe it?
A tag pasted into a duplicated or preview theme disappears the moment the live theme is republished, and the claim lapses without anything saying so.
Shopify: Online Store → Themes → the theme marked Live → Edit code → theme.liquid, inside <head>. Not in a duplicate, not in a preview, and not inside an app block that disappears when the app is uninstalled. Claiming by DNS record counts too, and is the more permanent option.
Pinterest tag
Check that the base code is present and that PageVisit, AddToCart and Checkout fire. Needed for the organic-to-paid flywheel.
The tag is what makes organic Pinterest measurable and what feeds the retargeting audiences later.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Is the base code present on every page?
A tag on the homepage only measures nothing that matters — the conversions all happen deeper in the funnel.
Pinterest Tag Helper, or view source on a product and a checkout page.
Do PageVisit, AddToCart and Checkout all fire?
Checkout is the event every downstream number is built on. A tag that only fires PageVisit reports traffic and no revenue.
Run a test purchase with Tag Helper open and watch all three fire in order.
Connect catalogue
Connect the feed via Ads > Catalogs or the Shopify app. Enables dynamic product pins.
The catalogue is what turns URLs into product pins that carry price and stock automatically.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Is the product feed connected and ingesting?
Without it every product pin has to be built by hand and none of them carry price.
Ads → Catalogues, or the Shopify app. Status should be 'active', not 'processing' or 'failed'.
Does the ingested product count match the shop, with no feed errors?
A feed that ingests 40 of 300 products looks connected and quietly excludes most of the catalogue.
Compare the ingested count against the shop's product count and read the error report.
Profile public
Check that the profile is not set to private.
A private profile is invisible to Pinterest search, which is the entire channel.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Is the profile public and visible to search engines?
'Hide from search engines' is on by default on some accounts and silently caps distribution to nothing.
Settings → Privacy and data → the search-engine visibility toggle must be off.
Turn off shopping recommendations
Settings > Social Permissions. If this is on, Pinterest shows competing products next to your pin and takes your clicks. Needs the account owner's login — with partner access only, this is a request to the client. Record that in the box and skip; it is not something the organic operator can do.
Shopping recommendations put competitors next to your pin and take the click you paid to earn.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Are shopping recommendations turned off?
Left on, Pinterest shows similar products from other shops directly under the pin.
Settings → Social permissions → shopping recommendations. Needs the account owner's login — if we only have partner access, this is a request to the client rather than something we can do, and that is a legitimate reason to skip it with a note.
What is not right, and what has to happen?
A failed check on its own is a dead end — the next person sees a red mark and has to redo the whole investigation to find out what it meant.
Name the specific thing that is wrong, who has to fix it (us, the client, their developer), and whether anything downstream is blocked until it is.
Limit messages
Turn messages off if the client does not monitor the inbox. Unanswered messages create a bad impression. Needs the account owner's login — with partner access only, this is a request to the client. Record that in the box and skip; it is not something the organic operator can do.
An inbox nobody reads is worse than no inbox: it is a visible unanswered message on a brand profile.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Does the messages setting match whether the client actually monitors it?
Unanswered messages sit publicly on the profile and read as an abandoned account.
Ask who reads the Pinterest inbox. If the honest answer is nobody, turn messages off: Settings → the messages section on the business account. It needs the owner's login, so without it this is a request to the client — say so in the box rather than leaving it unanswered.
Measure page speed
Slow pages suppress outbound clicks. The algorithm judges the destination too.
Pinterest judges the destination, not just the pin. A slow page suppresses outbound clicks on everything pointing at it.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Is mobile load time acceptable on the pages we will be pinning?
Outbound clicks are the ranking signal, and they are lost while a page is still loading.
PageSpeed Insights on two or three of the URLs from the pool, mobile tab. Under three seconds to largest paint.
Is there no single obvious blocker?
Most slow shops are slow for one fixable reason, and naming it is worth more than the score.
Look for uncompressed hero images, render-blocking app scripts, a bloated theme.
Review URL slugs
Pinterest reads the URL. Do not change existing URLs, that breaks Google SEO, but advise better slugs for new pages.
Pinterest reads the URL itself. Existing slugs stay as they are — changing them breaks Google — but new pages can be better.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Are the existing slugs readable and keyword-bearing?
A slug like /products/12345 tells Pinterest nothing about what the page is.
Look at the URL pool from P1.0.3. Words, not IDs.
Has the convention for new pages been agreed with the client?
This is advice, not a change we make. Unrecorded advice is advice nobody follows.
Only worth agreeing where new pages are actually coming — because the URL pool is thin (see P1.0.3), or because the client publishes regularly. A shop with enough URLs and no plans for more is a yes with 'no new pages planned' in the box; it is not an open item.
Image file names
The OCR reads file names. IMG_8492.jpg says nothing, modern-vanity-lighting.jpg does.
Pinterest runs OCR and reads file names. IMG_8492.jpg says nothing; modern-vanity-lighting.jpg says what the picture is.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Are the image file names descriptive rather than camera defaults?
File names are one of the few text signals on an image-first platform.
Inspect a handful of product images on the live site and read the src.
Is a naming convention agreed for everything uploaded from now on?
Renaming the back catalogue is rarely worth it; getting new uploads right always is.
Agree the pattern with whoever uploads images and record it here.
Meta descriptions
Pinterest scans the summary text of the landing page. Keywords there help.
Pinterest scans the landing page's summary text. Keywords there help the pin that points at it.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Does every URL in the pool have a meta description?
A missing description leaves Pinterest to guess the page's subject from the body copy.
Crawl the URL pool, or spot-check the collections and top products.
Do those descriptions carry the keywords we will be targeting?
A generic brand boilerplate on every page gives Pinterest the same signal for every URL.
Compare the descriptions against the keyword direction from phase 2.
Save buttons on the site
Pins that visitors save themselves rank faster. If the button is missing, advise a plugin.
A pin a visitor saves themselves ranks faster than one we publish. The save button is what makes that possible.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
What is not right, and what has to happen?
A failed check on its own is a dead end — the next person sees a red mark and has to redo the whole investigation to find out what it meant.
Name the specific thing that is wrong, who has to fix it (us, the client, their developer), and whether anything downstream is blocked until it is.
Mobile experience
More than 80 percent of traffic is mobile. Judge loading, scrolling and buying on your phone.
More than eight in ten Pinterest sessions are mobile. The desktop site is not the one that matters.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Does the site scroll smoothly on a real phone, with any popup dismissable?
Emulators hide the things that actually lose the click — janky scroll, popups that cannot be dismissed.
Open the site on your own phone on mobile data, not office wifi.
Can you complete a purchase end to end on mobile?
If checkout breaks on a phone, every click we send is wasted and nothing in the reporting will show why.
Go all the way to the payment step on your phone.
Content bank audit
Count per funnel stage: top for inspiration, middle for consideration, bottom for conversion. At least two URLs per stage. Determines the achievable frequency.
The content bank sets the achievable frequency. Two URLs per funnel stage is the floor a waterfall can run on.
How it scores. Every yes and this is settled — nothing to write. One no and the box at the bottom opens: say what is wrong and what has to happen, because that is what the next person needs.
Are there at least two URLs for inspiration (top of funnel)?
Top-of-funnel pages are what earn reach. Without them the account only ever talks to people already looking to buy.
Count guides, editorial, lookbooks and inspiration collections in the URL pool.
At least two for consideration (middle)?
The middle is where saves turn into visits. Skipping it makes the funnel a cliff.
Count comparison pages, collections, category pages with real copy.
At least two for conversion (bottom)?
Without bottom-of-funnel URLs the cycles produce traffic that never has anywhere to convert.
Count product pages strong enough to be a pin destination on their own.
What is not right, and what has to happen?
A failed check on its own is a dead end — the next person sees a red mark and has to redo the whole investigation to find out what it meant.
Name the specific thing that is wrong, who has to fix it (us, the client, their developer), and whether anything downstream is blocked until it is.