The 5-step Amazon title optimisation
Amazon's new 75-character title cut meant every listing got rewritten either Amazon did it for you, or you did it yourself before they could.
Either way, the same risk applies: clean, tidy copy that's completely blind to which of your words were actually bringing in sales.
I've watched brands lose a third of their organic traffic to a rewrite that looked better than what it replaced. Nothing broke. Nothing flagged. Sessions just quietly fell off, and by the time it showed up in the monthly numbers, three months had gone.
Here's the 5-step audit to catch it. You can run four of them yourself this afternoon.
Price every word
Open your title and write the monthly search volume next to each word. Any keyword tool will do it: Helium 10, Jungle Scout, even Amazon's own Brand Analytics if you're brand registered.
You'll usually find two or three words pulling almost all the weight, and a handful doing nothing at all. Filler like "premium", "high quality" and your own brand name (already indexed elsewhere) is eating characters that could be earning.
Protect your head term
Your head term is the one phrase doing the heavy lifting, the one that, if it disappeared, you'd feel it in a fortnight.
A rewrite can drop a term doing 13,000 searches a month without anything looking wrong. Check yours survived, and check it's near the front. Position within the title still carries weight.
Mine your reviews
This is the free one, and it's the one most people skip.
Pull your last 100 reviews and look for the phrases buyers repeat back at you. "True to size." "Dishwasher safe." "Fits a standard pram." These are the words real customers use when they go looking, and they're almost never the words a copywriter would have chosen.
The same trick works on your Q&A section and your returns reasons. Your customers are handing you keyword research for free.
Check adjacency
This one costs people the most money and gets noticed the least.
Amazon only matches a phrase when the words actually sit together. Split "leave in conditioner" from "spray" by fifty characters, and the phrase simply doesn't match even though every word is technically in your title.
One brand I looked at had all five words present and was matching on none of them. That's 179,000 searches a month, gone, on a listing that read perfectly well.
Measure what actually changed
This is the step you can't really do by hand, and it's why most title audits never get a second one.
After an edit, your sales move for a dozen reasons at once. A deal ran. Ad spend went up. A competitor went out of stock. Your rating shifted by a tenth of a point. Isolating the title's own effect means stripping out twenty-odd variables and then doing it again for every listing, every time you touch one.
Most founders make the edit, see sales go up, and credit the title. Sometimes that's right. Often it was the deal.
Here's the bit that matters:
Steps 1 to 4 are a listing job. Step 5 is a numbers job.
If you're running Shopify and Amazon together and can't see which changes are genuinely moving margin after fees, refunds and real COGS, then the problem isn't your copy. It's that your reporting can't answer the question you're asking it.
That's fixable, and it's usually faster than people expect.
We run a free 30-minute numbers sanity check for exactly this: where your reporting is hiding the answer, whether your margins are real, and three clear next steps. No pitch, and useful even if we never work together.
Drop me a message if that'd be handy!
