Mr Robert Prime

If you're already with an Amazon agency, here are a few questions worth asking on your next call:

— What's our TACoS trend over the last 90 days, and why?
— Which ASINs are leaking margin to fees we haven't claimed back?
— If this were your own money, what would you do differently this quarter?

A good agency answers all three without flinching. Watch what the silence tells you.

More below👇

B2B Amazon Secrets: The Questions Your Agency Should Answer If you're already with an Amazon agency, here are the questions a g...

If I started an Amazon business from scratch today, I wouldn't begin with the product.

I'd begin with the moat — the reason a competitor can't just copy your listing and undercut you by 15%.

Then timing. Then margin, protected before the first ad pound goes out. Then making sure I'm never at the mercy of one platform.

22 years in, that's the order. Most people do it backwards.

What would you add?

More below⬇️

If I Started on Amazon Today: 5 Things I'd Get Right 22 years and 12 exits in — if I started on Amazon from scratch toda...

Today only: grab my new book for free on Ai and Ecommerce.

I've spent 25 years watching e-commerce windows open and close.

Number one in the world for my market before Google existed. Built and sold a seven-figure Amazon brand. Watched a wave of factory-direct competitors flood the same niche until the easy money was gone.

The pattern repeated enough times to harden into a conviction: every easy road closes, and the only edge that lasts is the one you build before it does.

That conviction is what I wrote into my new book — The Art of War for Selling in the Age of AI.

It's a field manual for the moment the buyer became a machine. AI agents now search, compare, and check out on behalf of shoppers. If your product isn't machine-legible, it doesn't rank lower — it gets omitted entirely. And the free organic traffic road closed at the same time. Most sellers will meet this moment by doing more of what they already do. That's the wrong move.

The book maps 16 laws — Sun Tzu re-cut for the way commerce actually works now — plus three Grounds he never mapped: Price, Product, and Cash. It ends with a 90-day plan and an operator's checklist.

Today only, the Kindle edition is free on Amazon.

Link here to get it in the UK: https://www.amazon.co.uk/dp/B0H6G1V57D

Link here to get it in the USA: https://www.amazon.com/dp/B0H6G1V57D

I put in everything I could to make it insightful and a gift to my subscribers today only.

Share it with anyone who is interested in multichannel and the future of ai there.

Does Ai Ever Recommend Your Product?

I built a tool because I wanted to know exactly where brands need to get listed to be recommended by AI.

Most sellers haven't noticed what's happening.

Customers used to search Amazon, scroll through results, compare products, and make a choice.

Now they're opening ChatGPT, Gemini, Claude, Perplexity, or Amazon's Alexa for Shopping and asking:

"What's the best air fryer under £100?"

The AI gives them 3-4 recommendations.

If your product isn't one of them, you don't exist.

No impression.
No click.
No second chance.

So I built a tool to measure exactly where brands stand in this new world of AI-driven discovery.

At MrPrime.ai, the tool asks five major AI engines the questions real shoppers ask:

• Best products in a category
• Alternatives to competitors
• Budget recommendations
• Gift ideas
• Product comparisons
• "Just tell me the best one"

It then scores brands out of 100 in two areas:

1. Does AI know your product?
2. Does AI recommend your product?

Those are very different things.

Last week I ran an audit for a US fitness brand.

When I asked AI about their product by name, it knew everything about it.

Perfect recall.

But when I asked "What are the best jump ropes?" it never recommended them once.

Instead, it consistently suggested two competitors.

That's the hidden problem most brands don't see.

Being known isn't the same as being recommended.

The useful part is that the audit doesn't just show the problem.

It shows exactly how to fix it.

The tool identifies the sources AI relies on when making recommendations:

• Review websites
• "Best of" articles
• Reddit discussions
• Industry publications
• Comparison lists

Then it ranks them based on how valuable it would be for your brand to get featured there.

In other words, it turns AI visibility into an actionable outreach plan.

It also generates a paste-ready Amazon listing rewrite based on how AI systems actually interpret product pages.

Why did I build this?

Because AI search is becoming a major discovery channel, and almost nobody is measuring it properly.

Most tools in this space are built for enterprises and start at hundreds of dollars per month.

I think brands doing £200k, £500k, or £1m deserve access to the same insights.

And the opportunity is real.

In the audits I've run so far, attribute completion averages around 30% across UK sellers.

The brands that consistently convert Alexa for Shopping traffic are often above 90%.

The gap is huge—and it's fixable.

This isn't theory.

I used a similar approach with my previous business, UpDownDesk, where getting featured in the right lists and recommendation sources played a huge role in scaling the company.

If AI is becoming the new shelf space, the question isn't whether your product is good.

It's whether AI knows you're worth recommending.

Try the free visibility check at mrprime.ai and see where your brand stands.

She Asked For a Rufus Audit I said To Bin It 🤔

Was just in a team meeting and there was about to be a sales call. The potential client asked for a Rufus audit (we offer that)

I looked at notes and it was clear a Rufus audit was the last thing she needed.

Her situation:

4,700 SKUs
11% net
£120k a month sales - busy, stressed, lost, sales declining.

Just realised her net profit was 11% (before she'd properly factored returns, staffing and freight, so it's probably worse). I took the call and told her not to bother with the Rufus audit.

This was not at all a passive £12k a month this was a manufacturing, multi person full time job.

As a manufacturer, a baseline net profit should be 20% IMO.

Why the listing audit was the wrong tool

A Rufus rewrite is the cherry on top. You put it on once the cake is baked. At a sub-10% real margin, prettier listings just send more traffic into a catalogue that loses money in places she can't yet see.

That's a pill for a problem that needs surgery.

Where the money actually is, in order

✅ Profit by SKU first: cost of goods
✅ returns and FBA fees against every one of the 4,700, not a blended average. You can't fix what you haven't segmented.
✅ Follow the ad spend: £10k a month, and she couldn't tell me which SKUs it fed. Money pointed at the wrong shelf is money gone.
✅ Cut ruthlessly: the SKUs that only make you busy come off the catalogue.
✅ Then the real levers: B2B, bundles, pack sizes, landed cost. All before you touch a single image.

The number that should sting: at 11% net, £100k a month is about £11k take-home. Get the same revenue to 20% (not a big leap on your own products) and that's £20k: £9k more a month, £108k a year, without selling one extra unit.

In fact I would hope she will sell LESS units.

And there's a clock on the cosmetic work:

Amazon announced in June it's cutting title length to 75 characters from 27 July 2026, every category bar media. That hits all 4,700 of her listings. She didn't know.

Do this if you have a lot of skus

✅ Pull a profit-by-SKU report with COGS, returns and FBA fees against each ASIN, not a blended margin.
✅ Map every advertised SKU to its contribution, not its sales; pause spend on anything you can't trace to profit.
✅ List your bottom 20% of SKUs by contribution and decide what gets cut this month or what can be price increased.

THEN look at the cool stuff.

The pattern I see most weeks: a busy, stressed seller reaching for a shiny add-on while a sub-10% margin quietly bleeds them.

The fix is never another tool.

It's knowing your numbers before you spend the next pound.

Clarity cost her one call; the alternative is another year at 11%, and 11% compounds the wrong way.

I am going to offer an audit sprint over two months.

It won't be cheap but will be guaranteed value add.

If she wants to then handle the work internally, she should do so.

But I cannot offer a Rufus audit to her and feel ok about it.

Why your old playbook now works against you on amazon

🤕 The keyword-stuffed title was the cheat code for a decade.

As of 27 July it's a liability, and against the rules.

75 characters, spaces included, for everything except media. That's room for maybe a dozen words to carry brand, product and the one feature that matters.

The long, comma-spammed string is finished.

Amazon's giving you a new field to land the keywords somewhere legal:

Item Highlights, 125 characters for materials and recommended use cases.

It's searchable and shows next to titles in search results and on the detail page. So the terms don't vanish; they move to a structured field the assistant can actually parse.

Let the deadline slide and Amazon generates the title for you. Brand owners get 14 days inside Review Listings Changes to review, modify and approve the AI's version before it goes live. Ignore that window and you inherit whatever the machine writes about your own product.

Industry coverage of the 10 June notice says promotional language, repeated keywords, special characters and unapproved symbols are now flagged for auto-edits or truncation. So the grey-hat title tricks don't just underperform anymore; they get stripped.

The other grey-hat moves (competitor brand names in your copy, fake " #1 best seller" claims, review nudges in inserts) carry the same shift: they no longer just cost you rank, they put the listing, and sometimes the account, at risk.

Half-empty attribute fields make you invisible to AI shopping. The assistant builds its recommendation from structured data.

Blank field, no place on the shelf.

The gap that decides it

In the audits I run, attribute completion averages about 30% across UK sellers. The brands actually converting AI-shopping traffic sit at 90%+. That 60-point gap is the structured data the assistant needs before it will even consider recommending you.

Picture £1M in sales. If AI-assisted search routes even a tenth of your buyers (and one analysis already has Rufus at 14% of Amazon searches), that's £100k of sales exposed to a shelf you may not even be on. You're not ranked low on it. You're left off it.

Do this before 27 July

Rewrite your top 10 ASIN titles to 75 characters or less now, on your terms, so Amazon never generates one for you. Lead with brand, product, key feature, size.

Move the keywords you're cutting into the new 125-character Item Highlights field, where they stay searchable.

Pull those same ASINs and fill every relevant attribute in Product Details, not just the required ones.

Ask Alexa+ "what's the best [your product]" and note who it names. If it's a competitor, that's your roadmap.

The sellers who win in 2026 aren't spending more on ads; they're the ones whose listings a machine can actually read and won't get auto-rewritten under the new rule.

Make Amazon Boring Again (MABA)🧢

The More “Boring” The Lever Is, The More Money It Makes On Amazon.

After 25 years, the pattern is undeniable:

The money on Amazon is usually hiding in the stuff nobody wants to put in a keynote.

Not:

❌ “We upgraded to a premium A+ template”
❌ “We made image 3 look like a Nike billboard”
❌ “We doubled spend on the hero SKU because it felt brave”
❌ “We plugged in DSP and waited for the rocket ship”

But:

✅ Realigning ad budget to where we actually make money after fees, returns and all the annoying little costs nobody wants to talk about
✅ Growing Amazon B2B by 9% and AOV by 20% by targeting business buyers properly
✅ Shaving 3cm off packaging height to drop into a lower fulfilment tier and save £0.90 per unit
✅ Setting up buy box and suppression alerts and moving buy box % from 89% to 98%
✅ Renegotiating FBA prep so inbound costs dropped 14%

You get the idea.

The most profitable Amazon work is rarely shiny.

It is not always the new creative.
It is not always the bigger budget.
It is not always the latest ad product with a confusing acronym.
Sometimes it is packaging dimensions.
Sometimes it is fee leakage.
Sometimes it is buy box monitoring.
Sometimes it is business pricing.
Sometimes it is discovering that your “best” SKU is only best if you ignore half the costs.

For brands trying to grow on Amazon:

Become an accountant with a Seller Central login.

Boring to do.

Fun to spend the profit.

Make Amazon Boring Again.

📢 We'd like to invite you to our First MrPrime’s webinar of 2026 on Wednesday May 20th at 6:30 PM CET / 12:30 PM EST “MrPrime: Unlocking Amazon B2B, the Hidden Goldmine for Many Businesses”

Grab Your Spot Here 👉
https://lnkd.in/dRDXHq7S

😎 Host: Robert Prime - Co-Founder @ Robert Prime (MrPrime.com)

🎙️Speakers:

Robert Prime - Co-Founder @ Robert Prime (MrPrime.com)
Chelsea Cohen - Co-Founder @ SoStocked.com
Shane Barker - Owner @ TraceFuse
Jonathan Tilley - CEO & Co-founder @ ZonGuru
Andrew Smith - CEO & Founder @ Swapt
Benjamin (Ben) Kotch - President @ AccrueMe, LLC

🎙️Topics:

▶️ Robert Prime - Unlocking Amazon B2B, the Hidden Goldmine for Many Businesses!

▶️ Chelsea Cohen - Leveraging B2B to Increase Profitability!

▶️ Shane Barker - Why Your Amazon Reviews Matter More Than Your Ad Spend!

▶️ Jon Tilley - Structured Listing Engineering. How Amazon’s AI Is Changing Which Products Get Recommended!

▶️ Andrew Smith - From One-Time Orders to Long-Term Accounts: Retention Strategies for Amazon B2B!

▶️ Ben Kotch - Different Structures to Fund B2B Growth!

▶️ Q&A

👉 Grab Your Spot Here:
https://lnkd.in/dRDXHq7S

LinkedIn Login, Sign in | LinkedIn Login to LinkedIn to keep in touch with people you know, share ideas, and build your career.

Does AI recommend your book?

When someone asks ChatGPT, Claude, or Perplexity for "the best book on [your topic]," one of two things happens.

Your book gets named. Or a competitor's does.

There's no middle ground. And most authors have no idea which side they're on.

I built a free tool that runs the test in 30 seconds. It works for:

– Published Amazon books
– Live listings you want to improve
– Pre-launch manuscripts (before you've spent a penny on marketing)

It tells you what AI currently knows about your book, which books are being recommended instead, and the three fastest things to fix.

Free, see the result:
https://lnkd.in/gy7mRav6

Comment "deeper" if you want the full 18-module version for your specific book.

If someone gives you the “ideal” TACOS or ACOS…

They’re guessing

Because on Amazon—

There is no universal playbook.
It always depends on the account.

Here are two recent examples:

1) Supplement brand (stuck for 6 months)

Sales flat. Ads steady. No growth.

The obvious answer? “Optimise ads.”

The real answer? Fix the funnel.

We audited:
• New-to-brand sales
• Repeat purchase rates
• Subscribe & Save trends
• TACOS + profit by SKU
• CTR → Click → Add to Cart → Purchase

What we found:
→ Low CTR on key SKUs (main image, price, reviews issue)
→ Weak repeat purchase behaviour
→ Some SKUs losing money
→ Low new-to-brand contribution

Changes:
• Improved creatives + listing conversion drivers
• Pushed Subscribe & Save with stronger entry incentives
• Rebalanced SKU focus

Result:
→ 3x revenue in 90 days

2) Office products brand (7,000 SKUs)

Problem wasn’t growth.

It was clarity.

We audited profit by SKU and:
→ Cut 3,000 underperforming SKUs
→ Focused on top 100 revenue drivers
→ Set strict TACOS target (5%) aligned to margins

Then:
• Increased AOV with bundles (customers were buying in bulk anyway)
• Optimised for Amazon Business (B2B tiers, case packs)

Result:
→ Revenue doubled to £175k
→ Profit more than doubled
→ Fewer units sold, higher efficiency

Same platform.

Completely different levers.

That’s why “best practices” often fail.

Most brands aren’t underperforming because of ads.

They’re pulling the wrong lever.

If your growth has plateaued, it’s rarely random.

It’s structural.