What I checked, what I found, and what I'd like to fix first.
Summary. The site is in good shape strategically. We rank first for our core commercial terms, the blog is performing, and third-party coverage is strong — Reddit, Trustpilot, BuzzFeed and Pack Hacker.
The issues I found are leftovers rather than strategy problems: unfinished pages, product copy that was never updated after we discontinued two styles, and a performance fix built in July that was never published. Most of the remediation is removal, not new build.
The first batch is roughly three hours in the Shopify admin and requires no development time.
Two pieces of work this round.
1. I revisited the product consolidation project. This is the plan to merge our six near-identical market products (US, EU, UK, CA, SG, International) into one. I've recommended we keep it parked for now — it's a three-to-four week project with real risk on the day we switch over, and the benefit is mostly internal tidiness. If we do restart it, I'd do one small market first as a test rather than all six at once, which is a far safer way in.
2. I ran a full health check on the website — search visibility, page speed, product data, and how AI assistants like ChatGPT see us. I checked everything twice, by hand, before writing it down here. A few of the automated findings turned out to be wrong and I've left those out.
The six items with the clearest commercial impact, and what each one returns.
The rating data was never added to our product markup, so Google has no rating to display. Our listing renders as plain text while competitors — several with lower averages — show stars.
Benefit: higher click-through at the same ranking position. Star ratings are one of the most dependable levers on a product listing, and the reviews are already there and already verified. Best value-for-effort item on the list.
Shopify exposes a live product feed to AI shopping assistants. I queried it directly. On a general query with no location set — the most common case — it returns:
The advertising products are correctly hidden from Google. That setting does not apply to the AI feed, which is why this went unnoticed.
Benefit: correct product, correct price, correct page whenever an assistant recommends us. AI-assisted shopping grows every quarter, and we are currently feeding it inaccurate data about our own range.
Public and indexed right now:
Example heading — Give customers details about the banner image(s) or content on the template.
These sit at the URLs for our own styles — XL, Sling, Micro, Standard, Pack — which are exactly the pages that should rank for "nanobag xl" and its equivalents. A sixth is for "Air", a discontinued style. Separately there is a live page headed Master1.2, and another promoting a Winter Sale in July.
Benefit: removes a visible quality problem for both customers and Google, and frees those six URLs to be rebuilt as proper style pages targeting our own product names — which currently rank for nothing.
The product description still references "the Mini" and "the Air". It also states six weights that contradict our own comparison table on the same page — it gives the Standard as 0.85 oz where the table says 0.90 oz.
That description is shared with the products feeding our Meta advertising, so the ad copy carries the same errors.
Benefit: a single edit corrects the product page, the data Google indexes, and our ad copy at once. It also closes a credibility gap — the page contradicts itself on weight, which is the first spec a customer checks on an ultralight bag.
The fix removes unused code from every page load. It was completed and committed, then never promoted to the live site.
Separately: the homepage loads a 3 MB video and the product page nearly 5 MB before rendering completes, and one of our own scripts is doing substantially more work than required.
Benefit: load speed is a direct conversion lever, particularly on mobile where the majority of our traffic sits. The first item costs almost nothing — the work exists and needs publishing. The video change is the largest single speed gain available to us.
Page content is correctly translated across German, French, Italian, Dutch, Spanish, Danish and Swedish. The search headline and description are not — so a German shopper searching in German gets an English result.
Benefit: materially better click-through in every European market. The translation investment is already made on the pages themselves; this completes it. Handled in the Shopify translations tool, no development required.
Ordered by return on effort. The first group is largely removal, which is why it is both fast and low risk.
| Priority | What | Who | Time | What we get |
|---|---|---|---|---|
| 1 | Publish the speed fix we already built | Me, then publish | 15 min | Faster pages on every visit. Work is already done. |
| 2 | Fix the product description | Admin | 45 min | Stops advertising two bags we don't sell. Fixes the page, Google and our ads in one edit. |
| 3 | Hide the ad-only products from the AI feed | Admin | 20 min | AI assistants start recommending the right product at the right price. Needs decision D1 below. |
| 4 | Take the six placeholder pages out of Google | Admin | 20 min | We stop looking unfinished. Frees those pages up to be rebuilt properly. |
| 5 | Clean up five leftover pages | Admin | 30 min | Removes "Master1.2", the July "Winter Sale", and a duplicate homepage competing with our real one. |
| 6 | Hide two old test products | Admin | 5 min | Removes test data from the public store. |
| 7 | Stop Google crawling cart and search pages | Me | 20 min | Google spends its time on pages that can actually sell something. |
| 8 | Hide four internal working collections | Admin | 20 min | Internal lists and free-gift items stop showing publicly. |
| 9 | Fix "100,000" vs "200,000 customers" | Admin | 10 min | We currently claim both numbers on different live pages. |
That's the first batch — around three hours, almost entirely in the Shopify admin. After that:
| Next | What | What we get |
|---|---|---|
| Then | Add star ratings to our Google listings | 639 reviews at 4.84 finally working for us in search results. |
| Then | Fix the product page headings | Every product page currently announces itself as "Standard Black" rather than the product name — plus a piece of broken code is showing as a heading on every page. |
| Then | Translate search headlines for all 14 markets | European shoppers see results in their own language. Finishes a translation job that's already 90% done. |
| Then | Speed: video loading, script cleanup, image priority | The largest remaining conversion gains, especially on mobile. |
| Then | Name our press logos properly | Forbes, BuzzFeed, WIRED and The Grommet coverage is currently invisible to Google and to AI assistants. Six small edits. |
| Later | Build the missing pages: About, Shipping, Sustainability | We have founders, a 2017 start, a Kickstarter story and a tree-planting partnership — and no page telling any of it. This is what people and AI check before trusting a brand. |
| Parked | Product consolidation (six products into one) | Real benefit, but three to four weeks and switch-over risk. Recommend we leave it parked and test one small market first if we restart. |
Three of these are holding up work. None take long.
| # | Question | Who | Why it matters |
|---|---|---|---|
| D1 | Does our Meta ads feed need those six products to stay published on the website? | Ads / Marketing | This is the only thing blocking the AI-feed fix. If ads need them published, I'll find another route. |
| D2 | Is Google Merchant Center showing any "Duplicate product" warnings? | Ads / Marketing | Our six market products share the same barcodes, which can make Google suppress some free listings. If it's happening, the consolidation project becomes a revenue issue rather than a tidiness one — and moves up the list. |
| D3 | Can someone check our Cloudflare bot settings? | Whoever holds Cloudflare | Our site blocks some automated visitors. If it's also blocking ChatGPT and other AI crawlers, none of the AI-visibility work will have any effect. |
| D4 | Can I get a Shopify API key with product-edit access? | Store owner | We have around 700 product images with no descriptions attached, which costs us Google Images traffic. With a key I can fix them all in one pass; without one it's manual and realistically won't get done. |
To be clear about status: nothing on the live site has been changed yet. This round was checking and planning. Everything above is queued and ready to go once we agree the order and I have answers to the four questions.
My recommendation is that we do the first batch of nine this week. It's low risk, it's mostly deletion, it needs almost no development time, and it clears up the things that are currently costing us — the wrong AI recommendations, the ad copy for discontinued products, and the unfinished-looking pages.