20 Ecommerce Skills for LinkedIn: The B2B Half of a B2C Business

LinkedIn does not sell shampoo to consumers. Nobody is buying a candle from a connection request.

But every ecommerce business has a B2B half that runs entirely on LinkedIn, and almost nobody works it. That is what these 20 skills are for.

The five B2B surfaces inside ecommerce

Surface	Who you are messaging	Skills
Selling services TO ecom brands	Founders, heads of ecom, marketing leads	10
Wholesale and retail buyers	Category buyers at retailers and distributors	4
Supplier and 3PL partnerships	Ops leads, BD at logistics and manufacturing	2
Brand and affiliate partnerships	Marketing leads at complementary brands	2
Hiring	Operators, media buyers, retention leads	2

The first row is the biggest opportunity and the one this guide weights toward. Agencies, apps, 3PLs and freelancers all sell to ecom brands, and almost all of them do it badly, with a generic "I help Shopify brands scale" opener that gets ignored a hundred times a week.

The second row is the one ecom brands themselves underuse. Retail category buyers are on LinkedIn, they are reachable, and nobody messages them properly.

Why ecom outreach converts better than most

One property makes this vertical unusually good for outbound: the problems are visible without any access.

You can see their ad creative in the public ad library. You can see their PDP, their review coverage, their mobile speed, their schema, their product feed errors. You can see whether they have retargeting running.

In most B2B outreach you are guessing at the problem. Here you can screenshot it.

That is why the audit works as an opener in this vertical when it would be presumptuous in others, and why nine of the twenty skills below are built around observable evidence rather than inference.

The intel layer

These connectors are what turn a guess into a finding. Connect one or two, not all of them.

# Shopify AI Toolkit. Open-sourced April 2026, MIT.
# Official MCP server stack, agent skills and a Claude Code plugin
# in one namespace.
# github.com/Shopify/Shopify-AI-Toolkit
shopify dev mcp install

# Klaviyo. Community skill covering flows, segmentation,
# predictive analytics, catalogs.
npx skills add sales-skills/sales --skills sales-klaviyo

# Full DTC stack: Klaviyo audit, Shopify store audit, GA4,
# Looker Studio, paid media across Google, Meta and Microsoft.
# github.com/thatrebeccarae/claude-marketing

# Web extraction, for everything you do not have access to.
claude mcp add firecrawl --transport http https://mcp.firecrawl.dev

TWO NOTES on scopes. Run any Shopify Orders query in read-only first and confirm the result set before granting write_orders. And if you are a vendor rather than the brand, you have none of this access, which is exactly why the observable-evidence skills matter.

The context file

# Ecom Outreach Context

## Which surface am I working
[selling to brands / wholesale / supplier / partnership / hiring]

## If selling TO brands
What I sell, one line:
The revenue band I can serve:
Categories I know well enough to audit credibly:
What I refuse to do:

## If I AM the brand
Category and sub-category:
AOV band and revenue band:
Retail channels currently live:
What a buyer would need to believe to stock us:

## Proof, with permission
Numbers I can stand behind publicly:
Named brands I may reference:

## Send budget
Connected accounts:
Daily connection requests available, account-wide:

## Banned
<never_use_these>
Any ROAS figure without the spend level attached:
Any percentage without the absolute beside it:
Claims about their revenue I cannot observe:
</never_use_these>

The banned block is stricter here than in other verticals for a reason. Ecom operators read ROAS claims all day and discount them instantly. A figure without spend attached is the tell that marks you as one of the hundred.

Install all 20

mkdir -p .agents ecom/{research,copy,report}

# Selling to ecom brands
mkdir -p .claude/skills/{ecom-icp,ad-library-prospector,store-audit,feed-health-check,pdp-teardown,lifecycle-gap-finder,ecom-opener,ecom-sequence,ecom-objection,ecom-diagnostic}

# Wholesale and retail
mkdir -p .claude/skills/{retail-buyer-finder,category-fit-brief,line-sheet-opener,buyer-sequence}

# Supplier and partnerships
mkdir -p .claude/skills/{supplier-sourcing,threepl-shortlist,brand-partnership-finder,partnership-pitch}

# Hiring
mkdir -p .claude/skills/{operator-sourcing,hiring-outreach}

10 skills. The agency, app, freelancer and 3PL motion.

The whole approach rests on one thing: you can see their problems without any access, so the audit replaces the pitch.

01. ecom-icp

---
name: ecom-icp
description: Define which ecom brands to target. Use before any list
  building.
---
OUTPUT: platform, estimated revenue band, AOV band, category,
whether they run ads currently, and who owns marketing.

THE SWEET SPOT: running ads, spending enough to matter, without a
specialist agency. All three visible from the public ad library plus
the site footer.

AOV BAND matters more than revenue band. A brand at 25 pound AOV
has different economics and different problems from one at 250, and
the same pitch does not work on both.

EXCLUDE: no ads running, agency credited in the footer, revenue band
below your price threshold, category you cannot audit credibly.

The last one is the discipline. If you cannot write a specific
finding about supplements, do not target supplement brands.

02. ad-library-prospector

The list source that makes this vertical work.

---
name: ad-library-prospector
description: Build a list from observable ad activity.
---
SOURCES, best first:
1. Brands with ads live in the public ad library in your category
2. Brands whose ads have run UNCHANGED for 60+ days
3. Recently launched stores now spending
4. Brands hiring an in-house media buyer

SOURCE 2 is the wedge. An unchanged creative running two months is
creative fatigue you can screenshot, date and name.

CAPTURE per brand: ad count, longest-running creative in days,
format mix, platform, site URL, category, and whether any video is
in the mix.

The format mix column is the second-best finding after fatigue.
A brand running only static in 2026 has a gap you can quantify.

Write to ecom/research/prospects.csv with a linkedin_url column for
the founder or marketing lead.

03. store-audit

---
name: store-audit
description: Produce the free audit from observable data only. This
  replaces the opener.
---
SIX FINDINGS maximum, each with a screenshot or URL:

1. Creative fatigue: the longest-running ad and its days live
2. Format gaps: what is missing from the mix
3. Feed health: any visible product feed error
4. PDP: reviews, schema, image coverage, mobile speed
5. Post-click: does the landing page match the ad promise
6. Retargeting: visible or not

THEN: the one you would fix first, and roughly what it is worth.

Quantify ONLY where the arithmetic is defensible from observable
data. Never invent a revenue figure.

FAIL CLOSED if fewer than three findings have evidence. Skip the
prospect and report why. A thin audit is worse than none.

Under 900 words. Six findings is the ceiling, not the target. Four
strong ones beat six thin ones.

04. feed-health-check

The finding nobody checks and everybody has.

---
name: feed-health-check
description: Check a product feed for the errors that silently cost
  money.
---
CHECK, from what is publicly visible:
- Products in Shopping results versus products on the site
- Missing GTIN or MPN where the category requires it
- Title structure: does it lead with brand or with the product term
- Missing or generic product descriptions
- Image quality and background compliance
- Price and availability mismatches against the live PDP
- Variants collapsing into one listing

OUTPUT per issue: the product, the error, and the likely effect on
delivery or disapproval.

THE HEADLINE NUMBER: how many products are effectively invisible.
That figure is the strongest single finding available in this
vertical, because it is money already spent on a catalogue that
cannot be shown.

Never claim a revenue figure from it. Say what is invisible, not
what it costs.

05. pdp-teardown

---
name: pdp-teardown
description: Tear down a product detail page against what actually
  converts.
---
CHECK, on the top three products:
- Above the fold: is the price, the main image and the add-to-cart
  visible on mobile at 375px without scrolling
- Reviews: present, surfaced above the fold, and with photos
- Image count and whether any show scale or use
- The first 40 words: do they answer "what is this and why this one"
- Schema: Product, Offer, AggregateRating present and valid
- Shipping and returns visible before the cart
- Load time on mobile, and layout shift

OUTPUT each as PASS, WEAK or FAIL with the evidence and one line on
what it costs.

END with the single change that would move conversion most. One
change, not a list. A list gets filed; one change gets done.

Never promise a conversion percentage. Describe the fix and the
reason.

06. lifecycle-gap-finder

The finding you can get without the ad library, by buying something.

---
name: lifecycle-gap-finder
description: Find the missing email and SMS flows from the outside.
---
THE STANDARD SET. Most brands are missing two of these:
- Welcome, on signup
- Abandoned cart
- Browse abandonment
- Post-purchase, with the second-order nudge
- Winback, at the category's repeat interval
- Back in stock
- Review request

TEST FROM OUTSIDE: sign up, add to cart and leave, browse without
adding, buy something small. Record what arrives and when.

OUTPUT: which flows exist, which are missing, the timing of each,
and what the missing ones typically carry in that category.

THE GAP IS THE FINDING. "Nothing arrived after I abandoned a cart"
is a specific, verifiable, uncomfortable finding, and it is true of
more brands than anyone expects.

Never estimate the revenue of a missing flow. Name the gap.

07. ecom-opener

---
name: ecom-opener
description: The message that delivers the audit.
---
Line 1  The single biggest finding, specific and evidenced.
Line 2  Offer the rest. Six findings, one page, no call needed.
Line 3  One question about whether they already know.

DO NOT attach the audit to message one. Offer it. The yes is the
engagement, and a cold attachment gets ignored.

Under 300 characters.

BANNED, burnt in this vertical specifically:
"I help Shopify brands scale"
"I noticed your ads"
"I was looking at your store"
"Quick question about your Meta account"
"Are you happy with your current ROAS"

The first one is the single most-sent opener in ecom outreach. Using
it puts you in a folder with two hundred others.

08. ecom-sequence

---
name: ecom-sequence
description: The four-touch sequence for ecom founders.
---
1  Opener, biggest finding plus the offer
2  On reply, send the audit. No pitch inside it.
3  Wait 2 days after sending. One question about finding two.
4  Wait 5 days. Close-out.

IF NO REPLY to touch 1, send the finding anyway at touch 2. Giving
it away unasked converts better than chasing permission.

TIMING NOTE specific to this vertical: avoid the two weeks before
and the week after Black Friday entirely. Nobody is evaluating
vendors, and a message then is remembered badly.

Reply stops the sequence. No wait node after a connection request.

State the daily send split against the account ceiling.

The spec maps node for node onto the Prosp builder, and the ceiling it checks against is real: 20 connection requests a day and roughly 100 messages a week, account-wide rather than per campaign.

One ecom-specific use worth knowing: ecom founders cluster in the comments of the big operator accounts and the platform pages. Prosp imports commenters and likers straight from any post URL, so a post about Shopping feed errors on a large account is a pre-qualified list of people with that problem.

09. ecom-objection

---
name: ecom-objection
description: The objections ecom operators actually raise.
---
"Our current agency handles this"
-> Ask when they last changed creative. The fatigue finding answers
   it without you attacking anyone.

"Our margins are too thin for agency fees"
-> A real objection, not a brush-off. Ask for AOV and current ROAS.
   If the maths does not work, say so and walk. That honesty gets
   referrals in a small industry.

"We are seasonal, not now"
-> Capture the date. Seasonality is real here and a diarised
   follow-up converts unusually well.

"We tried an agency and it did not work"
-> Ask what the reporting looked like. It is almost always that.

"We do it in-house"
-> Ask who owns the feed. Usually nobody does, and that is the
   opening.

Never answer a margin objection with features. Work the arithmetic
with them or leave.

10. ecom-diagnostic

---
name: ecom-diagnostic
description: Weekly. Stop at the first failure.
---
1 ACCEPTANCE LOW
  The list includes brands that already have agencies. Tighten the
  footer check.
2 REPLIES LOW
  The finding in line one was generic. "Your ads could be better"
  is not a finding.
3 REPLIES FINE, AUDITS UNOPENED
  The audit is too long. Cut to four findings.
4 AUDITS READ, NO CALLS
  The audit pitched instead of helping.

STAGE 4 is the one to watch. An audit that sells does not get
forwarded internally, and forwarding to the founder is what closes
these when you started with a marketing lead.

ALSO track: which finding type produced the most replies. Feed
errors and missing flows usually outperform creative fatigue,
because fewer people have mentioned them.

ONE change per week.

10 skills for the brand side. Wholesale buyers, suppliers, partnerships and hiring.

This is the half most ecom brands ignore entirely, because they think of LinkedIn as somewhere their customers are not. Their buyers, suppliers and next hires all are.

Wholesale and retail: 4 skills

---
name: retail-buyer-finder
description: Find the actual category buyer at a target retailer.
---
THE TITLE VARIES by retailer and it matters. Buyer, Category
Manager, Merchandise Manager, Assistant Buyer, Head of Buying.

PER RETAILER capture:
- The buyer for YOUR category specifically, not the head of buying
- Their assistant buyer, who often does the first screen
- Whether the retailer runs a formal supplier portal
- Their range review cycle, if published

THE ASSISTANT BUYER is the one to reach. Buyers are inundated;
assistants are not, and they control what reaches the buyer.

THE RANGE REVIEW CYCLE is the single most important field. Pitching
outside it wastes the contact, because there is no decision to be
made and you do not get a second first impression.

EXCLUDE retailers whose category you cannot supply at their volume.
A buyer who likes you and finds you cannot fulfil is worse than no
conversation.

---
name: category-fit-brief
description: Build the one-page case for why this retailer should
  stock you. Before any message.
---
WHAT A BUYER ACTUALLY NEEDS TO BELIEVE:
1. This fills a gap in my current range, not a duplicate
2. It will sell at my price architecture
3. The brand can supply at my volume, reliably
4. There is demand I can see without taking a risk

OUTPUT one page answering each, with evidence:
- The gap: what they currently stock in the category, and what is
  missing from it
- Price architecture: where you sit against their existing range
- Supply: your current volume, lead times, MOQ
- Demand proof: DTC sell-through, repeat rate, waitlist size,
  any regional concentration near their stores

POINT 4 IS THE ONE THAT WINS. A buyer risks shelf space. DTC data
showing demand in their catchment removes the risk, and almost no
brand brings it.

Never lead with brand story. Lead with the gap.

---
name: line-sheet-opener
description: The first message to a retail buyer.
---
Line 1  The gap in their range, named specifically. A product type,
        a price point, or a customer need their current range does
        not cover.
Line 2  One demand proof point. A number, with the timeframe.
Line 3  Ask whether the range review is open.

Under 300 characters.

NEVER attach a line sheet to message one. Buyers receive dozens and
open none. Offer it.

NEVER open with your founding story. Every brand has one and buyers
have read them all.

BANNED:
"We'd love to be stocked in"
"I think we'd be a great fit for"
"Our products are loved by customers"

The gap-first opener works because it tells them you looked at their
range rather than their logo.

---
name: buyer-sequence
description: The sequence for retail buyers. Longer delays than any
  other in this guide.
---
1  Opener, the gap plus one proof point
2  Wait 7 days. The line sheet and the category fit brief, on reply
   or unprompted.
3  Wait 10 days. One new demand data point. Sell-through, a
   regional spike, a comparable retailer.
4  Wait 14 days. The close-out, referencing their next range review
   date rather than asking for a decision now.

BUYERS MOVE ON CYCLES, not on urgency. Compressed delays read as
desperation and they have seen it a thousand times.

If the range review is six months out, say so and diarise it. A
follow-up that lands two weeks before the review, referencing the
first conversation, converts better than anything you can send now.

Reply stops the sequence.

Supplier and 3PL: 2 skills

---
name: supplier-sourcing
description: Find and qualify a manufacturer or supplier through
  LinkedIn rather than a directory.
---
DIRECTORIES give you sales contacts. LinkedIn gives you the people
who run the floor, and they answer differently.

TARGET: production managers, operations directors, BD leads at
manufacturers in your category.

QUALIFY BEFORE MESSAGING:
- Do they already make something in your category, visibly
- What is their apparent scale, from headcount and posts
- Do they serve brands your size, or only much larger
- Any certifications your category requires

THE OPENER: name what you saw them produce, state your category and
rough volume, and ask one question about their minimum.

Lead with volume honestly. A supplier who finds out your real MOQ
after three calls remembers, and the category is small.

---
name: threepl-shortlist
description: Build and qualify a 3PL shortlist before the sales calls
  start.
---
QUALIFY ON, in order:
1. Do they serve your category and your unit economics
2. Warehouse locations against your customer concentration
3. Integrations: your platform, natively or via middleware
4. Their client size band. Being their smallest client is bad;
   being their largest is worse.
5. Peak capacity. Ask directly what happens in Q4.

THE OPENER should reference a specific operational detail, not a
generic "looking for a 3PL". Their location relative to your
customers, or an integration you need.

QUESTION 5 is the one that separates them. Every 3PL says yes to
capacity in January. Ask what their Q4 dispatch times looked like
last year, specifically.

Never choose on price alone. The cost of a bad peak is larger than
any per-pick saving.

Brand partnerships: 2 skills

---
name: brand-partnership-finder
description: Find complementary brands worth a partnership.
---
COMPLEMENTARY, NOT COMPETING. The test: would a customer plausibly
buy both in the same month, without one replacing the other.

SCORE each candidate:
- Audience overlap: same customer, different need
- Size parity. A brand ten times your size will not reciprocate.
- Channel overlap: do they have a list, and is it active
- Brand safety: would you be comfortable being associated

TARGET the marketing lead, not the founder, for anything
operational. Founders for anything strategic.

SIZE PARITY is the one people ignore. Partnership outreach to a
much larger brand is a sponsorship request wearing a different hat,
and they can tell.

---
name: partnership-pitch
description: The message that proposes a specific trade.
---
NEVER propose "exploring a partnership". Propose the specific thing.

STRUCTURE
Line 1  The overlap, named. Their customer and yours, and why the
        same person buys both.
Line 2  ONE concrete trade. A list swap, a bundle, a giveaway, a
        co-created product.
Line 3  What they get, quantified. Your list size, your engagement,
        your reach, with real numbers.

LEAD WITH WHAT THEY GET. Every partnership message leads with what
the sender wants, which is why most are ignored.

Be specific about your own numbers. A list size with an open rate
attached is credible; "a highly engaged audience" is not.

Under 350 characters.

Hiring: 2 skills

---
name: operator-sourcing
description: Find ecom operators worth hiring, from observable work.
---
THE ROLES ecom brands actually compete for: media buyer, retention
or lifecycle lead, ecom manager, CRO specialist, creative strategist.

SOURCE FROM OBSERVABLE WORK, not from job titles:
- People posting teardowns or breakdowns in your category
- Media buyers whose named brands you can see and check
- Agency operators three or more years in, who are the most
  reachable group in ecom hiring
- People commenting substantively on operator content

SOURCE 3 is the one to work. Agency operators have seen many
accounts, are often ready to move in-house, and almost nobody
approaches them directly because everyone assumes they are happy.

CAPTURE: the work you can actually see, with the URL. That is what
the opener references.

---
name: hiring-outreach
description: The message to a passive candidate. Not a job ad.
---
Line 1  The specific work of theirs you saw, and what was good
        about it. Not "your profile", not "your experience".
Line 2  What you are building and why it needs that exact skill.
Line 3  One question. Never "are you open to opportunities".

NEVER paste the job description. Never lead with the salary band,
and never hide it either if they ask.

PASSIVE CANDIDATES are not job hunting. The message has to be
interesting on its own terms, because there is no need being met.

BANNED:
"I have an exciting opportunity"
"Are you open to new opportunities"
"I came across your profile and was impressed"

If they say no, ask who they would recommend. That question
produces more hires in ecom than the original message does.

All four of these surfaces run on the same account and share the same ceiling. Prosp handles the split: separate campaigns for buyers, suppliers, partnerships and hiring, each with its own daily allocation, all inside the 20 requests a day that are account-wide rather than per campaign.

Turn on skip-leads-contacted-by-another-campaign. Retail buyers and partnership contacts overlap more than you would expect, and the same person receiving two different pitches from one brand in a week is a bad look in a small industry.

The calendar problem

Ecommerce has a seasonal shape no other vertical in this library has, and it governs all twenty skills.

Do not run outbound to brands in the two weeks before Black Friday or the week after. Nobody is evaluating vendors and a message then is remembered badly.

Retail buyers run on range review cycles, often six to nine months ahead of the season. A spring pitch for spring is late by roughly half a year.

January and late Q1 are when brands review their stack, their agency and their 3PL. That is the window for the selling-to-brands skills, and it is short.

Hiring outreach works best in January and September, for the obvious reasons.

Plan the send budget against that calendar rather than spreading it evenly. Twenty requests a day in mid-November is twenty wasted.

20 skills find the conversations. Prosp runs them 💜

↳ Start a 14 Day Free Trial 💜 ↳ Watch Full Tutorial ▶️