Cold email agency for e-commerce
GrowthPal runs outbound for two kinds of e-commerce company: businesses selling into online brands, and brands or distributors selling to retail and trade buyers. The two share almost nothing operationally, so we pick one before a list is built. Lists come from store signals and public trade data, not from a bought file of store owners. Best fit is past 3M in ARR with a reachable market of roughly 30,000 contacts.
Two audiences, and they are not the same job
Selling to e-commerce brands
Platforms and app vendors, 3PLs and fulfillment providers, OMS, WMS and PIM software, retention and post-purchase tools, payment and fraud vendors, freight forwarders and customs brokers, performance and CRO agencies, packaging suppliers. Your buyer sits inside a D2C or multi-channel brand. Your list is built from store technology signals.
B2B and wholesale e-commerce selling
A brand with a wholesale arm, a distributor, a manufacturer selling trade, a private-label supplier, a B2B catalog seller. Your buyer is a retail buyer, category manager, merchandise planner or purchasing lead at a chain, group or independent retailer. Your list is built from store footprints, trade directories and show data, and your calendar bends around line reviews.
Who you are really selling to
| Role | What they care about | Where they pay attention | Best channel |
|---|---|---|---|
| Head of E-commerce or E-commerce Director (A) | Conversion rate, average order value, and site uptime through peak. Anything that moves revenue inside the current quarter and will not break in November. | Operator communities, platform partner circles, LinkedIn posts from other operators, a short list of trade newsletters. | Email first, LinkedIn as the warming layer. They read on a phone between vendor calls, so the first two lines decide it. |
| Director of Operations, Supply Chain or Fulfillment (A) | On-time shipping rate, returns processing time, inventory accuracy, and chargeback exposure from retail partners. | 3PL and carrier account managers, logistics peer groups, warehouse software documentation. | Email, with a phone follow-up that works better here than it does with marketing roles. Name a specific operational failure or do not write. |
| Founder or Managing Director at a small or mid brand (A and B) | Cash, inventory position, and whether a vendor becomes one more thing to manage. | Founder communities, Slack and WhatsApp peer groups, and the suppliers they already use. | Short email, one clear ask. LinkedIn DMs work when the profile is active. |
| Retail Buyer, Category Manager or Merchandise Planner (B) | Sell-through per linear foot, margin, lead time, minimum order quantity, EDI readiness, and whether you will cause chargebacks. | Trade shows, existing rep relationships, category trade press. | Email timed to the line review window. Outside that window the quality of the pitch does not matter. |
| Purchasing Manager or Procurement Lead (B) | Supplier qualification, terms, lead time reliability, single-source risk. | Formal supplier portals and RFQ processes. | Email that offers to be added to a supplier list. Route to a portal registration, not a calendar link. |
What makes outbound different here
Peak season freezes the calendar
A. Many merchants and their vendors run a code freeze and a publishing pause for roughly the two weeks before Black Friday, with build windows often closing in mid-October. Anything that needs implementation will not be approved inside that window, however good the email is.
What it decides: Pacing and the ask. Implementation-dependent angles land before the freeze or in January. The freeze weeks carry touches with no build attached.Retail buying runs on its own calendar, not on your quarter
B. Buyers work to line review windows and an open-to-buy budget, and many retailers plan on the 4-5-4 merchandising calendar rather than ordinary months. The National Retail Federation describes 4-5-4 as voluntary and does not publish how many retailers use it, so treat it as a question to ask, not an assumption.
What it decides: The list schema. Every row carries a category and a review-window field, and send dates come from that field rather than from a weekly sending plan. Pacing is seasonal, not steady.The e-commerce inbox is the most pitched inbox in B2B
A. Operators at D2C brands receive heavy agency and tool outreach, and near-generic personalization now reads as spam to filters and to people. Every published figure for how many pitches an operator gets per day traces to a vendor selling a cure, so we do not quote one.
What it decides: The angle. It has to contain something only someone who opened their store would know: a specific app, a checkout behavior, a line in their shipping policy. If the opener could go to a thousand stores, it gets cut. Volume is the wrong lever here.Trust is a supply-chain question, not a marketing one
A and B. Both audiences are being asked to hand over revenue-critical operations or shelf space. A 3PL switch, a payment vendor switch, a new supplier: all are downside-weighted decisions where the buyer loses more from a bad call than they gain from a good one.
What it decides: The ask. The first step is the smallest reversible one: an audit, a sample pack, a line sheet review, a single test lane. A demo of the whole thing is the second ask, not the first.Committee size varies more here than in most markets
A and B. A seven-figure D2C brand may have one decision maker who signs the same week. A retail chain routes a new supplier through buying, merchandise planning, supply chain compliance and finance, where the first reader is never the approver.
What it decides: Segmentation. Single-owner segments get a direct ask. Committee segments get a sequence that supplies a document designed to be forwarded internally.Buying triggers, and the angle each one earns
| Trigger | How we find it | The angle it earns | Proof the email needs |
|---|---|---|---|
| Platform migration or replatforming (A) | Storefront technology fingerprint change, a careers post for a platform-specific developer, an agency case study announcing the project, a new partner directory listing. | Every surrounding tool is being re-evaluated at once, so this is the window where switching costs are already being paid. | The same migration path, named, with the specific integration they need running on day one. |
| App install, uninstall or tier change (A) | Public app store review activity, storefront script tags changing, store-monitoring databases that track installed apps. An uninstall of a competitor is a displacement window. | Name the specific job the removed app was doing and the gap it leaves behind. | A before and after of the same swap at a comparable store, with the integration points listed. |
| Role hire or team change (A and B) | Careers pages, job listings, a new title appearing on a profile. A new Head of E-commerce, a first Retention Lead, a first Head of Wholesale. | New owners rebuild the stack and the vendor roster in their first ninety days. | What the first ninety days usually breaks, plus one artifact you can hand over in week one. |
| Funding, acquisition or ownership change (A and B) | Company registry filings, press releases, investor portfolio pages, Companies House or state filings. | New capital plus new reporting pressure forces a consolidation or expansion decision. | What post-raise reporting demands and how you cover it. |
| New market, channel or territory launch (A and B) | A new currency or country selector on the storefront, a new marketplace seller profile, a new language subfolder, a trademark filing in a new jurisdiction, a changed shipping policy page. | Cross-border duty, tax registration, local fulfillment and local payment methods all become live problems in the same month. | A named corridor you have shipped into, with the registration sequence in order. |
| Peak season stress and the post-peak review (A) | Public status pages, downtime reports, delivery complaints in reviews and social during peak, a code freeze notice on a vendor page. | January and February are when last peak's failure finally gets budget. The outreach lands after the pain, not during it. | A specific peak failure mode you fixed, described in operational terms, not marketing ones. |
| Trade show exhibitor or attendee listing (B) | Published exhibitor directories for apparel, gift, home, food and beverage and hardware shows, plus buyer attendee lists where they are published. | An exhibitor is opening wholesale accounts right now. A buyer attending is filling a category gap right now. | That you sell into their exact category and channel, with a sell-through reference from a comparable store format. |
| A D2C brand opens a wholesale channel (B) | A stockists, wholesale or trade page appearing on a D2C site, a wholesale marketplace brand profile going live, a wholesale application form, a published line sheet. | The brand has decided to sell trade but has no buyer list, no EDI capability and no trade terms sheet. | What retail buyers reject first, and the document set that fixes it. |
Example angles we would test
Examples of angles to race in this market, not results. Each one is a hypothesis with its proof assembled before a single send.
A. You moved platforms, and the thing that usually breaks next is subscription or loyalty data continuity.
- Proof it needs
- A named migration on the same path, with the data object that broke and how it was reconciled.
- The ask
- A thirty-minute review of their migration checklist.
A. You removed a tool that was doing a specific job. Here is what usually fills that gap, and what usually does not.
- Proof it needs
- A documented swap at a store of comparable size and category.
- The ask
- Send a short written comparison. No call.
A. January is when last peak's fulfillment failure finally gets attention.
- Proof it needs
- An operational account of a specific peak failure mode and the fix, written for an operations reader.
- The ask
- A review of their peak incident log.
A. Your store added a second currency, so duty, tax registration and local payment methods all land at once.
- Proof it needs
- A named corridor you have shipped into, with the registration sequence.
- The ask
- A one-page checklist for that specific corridor.
A. You are ninety days into owning this channel and you inherited a vendor roster you did not pick.
- Proof it needs
- A short audit artifact built from public store data before any call happens.
- The ask
- Permission to send the audit.
A. Your returns policy page implies a manual process at your order volume.
- Proof it needs
- A described workflow change and the operational metric it moved at a comparable merchant.
- The ask
- A look at their current returns flow.
B. Your stockists page went live, and retail buyers will ask for three documents you probably do not have yet.
- Proof it needs
- The document set itself: a line sheet template, a trade terms sheet, a sell-through reference.
- The ask
- Review their line sheet.
B. Your category has a visible gap in this store format, and we supply into it.
- Proof it needs
- Sell-through evidence from a comparable format, with lead time and minimum order quantity stated up front.
- The ask
- A sample pack, not a meeting.
B. You exhibited at the show. Here is the one thing we do that nobody there covered.
- Proof it needs
- A specific, verifiable supply capability tied to the show's category.
- The ask
- A fifteen-minute slot before the next buying window.
B. You just signed a vendor purchasing agreement and now need to be EDI-capable quickly.
- Proof it needs
- The exact documents you support, 850, 855, 856, 810 and 997, and the connectivity method.
- The ask
- A readiness check against that retailer's published vendor guide.
How the list gets built in this market
Lists here are built from dated signals on public surfaces, then verified. The store itself is the richest source and the least used one.
Selling to e-commerce brands
- Storefront technology fingerprints, and direct reading of a store's own markup for platform and app signatures.
- Platform partner and app directories, including app review pages with dated reviews.
- Marketplace seller profiles across the major marketplaces.
- Public traffic and ranking estimates, used as a rough size band only, never as a fact about the business.
- Careers pages and job posts, for role triggers.
- Company registries, for ownership and filing changes.
- The store itself: shipping policy, returns policy, country selector, subscription widget, checkout flow.
Selling to retail and trade buyers
- Trade show exhibitor and attendee directories, by category.
- Wholesale marketplaces as a discovery surface. Only verified retail accounts can buy on them, which is itself a qualification signal.
- Retail store locators and chain store lists, read from retailers' own sitemaps.
- Distributor and buying group member directories.
- Retailer supplier and vendor portals, which publish their onboarding requirements openly.
- Trade association member lists for the category.
- Local business registries and licensing data, for independent retail.
How a row gets verified
- Confirm the company still trades: a live storefront, recent review or order activity, a dated public signal inside a sensible window.
- Confirm the role still holds, against two sources, because e-commerce roles churn fast.
- Verify the address technically: syntax, domain MX record, catch-all detection, and a verification pass. Catch-all domains go into a separate, slower segment rather than into the main send.
- Suppress existing customers, prior opt-outs, anything on a do-not-contact list, and sole trader or partnership addresses in jurisdictions that treat those as individuals.
- Record the trigger and its public source on the row. A row with no trigger is not ready to send.
What makes a bad list here
- Any store, any size, any country, filtered only by the platform it runs on. Platform alone is not a qualifier.
- Rows with no dated signal, which force the angle to be generic.
- Role addresses treated as a decision maker. At a small brand they might reach the founder. At a chain they reach nobody.
- Dropshipping shells, dormant stores and test stores, which technology scrapes pull in by the thousand.
- Retail buyer lists with no category field, so a hardware buyer receives a beauty pitch.
- Scraped show or marketplace data used without checking the company still exhibits or still trades.
- A bought file of store owners. Those files are resold widely, go stale fast, and the addresses were burned by the last twenty senders.
Compliance and trust constraints
Nothing here is legal advice. Check your own position with a qualified adviser and read each source directly, because rules and thresholds change.
CAN-SPAM makes no exception for business email
The FTC's compliance guide states the law covers all commercial email, including business to business. That means accurate header and routing information, a subject line that is not deceptive, a clear disclosure that the message is an advertisement, a valid physical postal address, and an opt-out honored within ten business days. The FTC also states you cannot contract away responsibility for email sent on your behalf, which applies directly to an agency sending for a client.
Read the source
In the UK, a sole trader is not a company
The ICO treats corporate subscribers, meaning companies, limited liability partnerships, Scottish partnerships and government bodies, differently from individual subscribers, who need consent unless the soft opt-in applies. Sole traders and some partnerships count as individuals. This matters more in e-commerce than in most markets, because independent retailers and small D2C brands often are sole traders, so the same list can contain both categories.
Read the source
Mailbox provider rules decide whether any of this is read
Google's sender guidelines apply above a daily volume threshold to Gmail accounts and have been in force since February 2024. They require SPF, DKIM and DMARC, correct forward and reverse DNS, TLS, standards-compliant formatting, one-click unsubscribe for marketing mail, and a spam rate kept low. E-commerce operators and retail buyers sit heavily on Google Workspace and Microsoft 365, so a domain that fails these rules is not read regardless of what it says.
Read the source
Retailer vendor compliance is a trust gate, not a legal one
Retailers publish their own supplier requirements. These commonly include GS1-licensed GTINs at item level, the EDI document set, and shipping labels carrying an identifier that matches the advance ship notice. Asserting readiness you do not have creates chargebacks later, so an email to a retail buyer should only claim what it can back. Always read the retailer's own current vendor guide rather than a summary of it.
Read the retailer or regulator source directly.A targeting brief for this market, in full
GrowthPal fills in this document before a single row is built. It is published whole rather than as a screenshot, because the shape of it is the argument. Every field is filled from public sources before any call happens.
| Field | What goes in it |
|---|---|
| Audience | A or B, chosen, not both. Everything below changes with this answer. |
| Segment | Category, company size band, and channel mix. For wholesale, the store format too, because sell-through evidence only transfers between comparable formats. |
| Trigger | The dated event that earns the email, plus the public URL it was read from. A row without this is not ready. |
| Signal on the row | The one specific thing about this company that could not be written about any other: an app, a checkout behavior, a line in their shipping policy, a stockists page that went live. |
| Calendar field | Freeze window for brand-side rows. Line review window and open-to-buy timing for wholesale rows. Send dates are derived from this, not from a weekly plan. |
| Angle | The pain hypothesis in one line, in the buyer's vocabulary, not ours. |
| Proof the email needs | The specific evidence that makes the angle credible to this role. Named, comparable, and assembled before the send. |
| The ask | The smallest reversible step. Named explicitly so the sequence cannot drift back to asking for a demo. |
| Disqualifiers | Dormant stores, dropshipping shells, role addresses, catch-all domains routed to a slower segment, prior opt-outs, and sole trader addresses in jurisdictions that treat those as individuals. |
| What would tell us this is wrong | Written down before sending. A hypothesis with no failure condition cannot teach you anything, because no result would ever count against it. |
We publish the brief rather than a dashboard screenshot because a screenshot proves a number and a brief proves a method. Copying this means copying how we work, which is the point.
How the engagement runs
Three stages and nine steps, the same ones in every market: build the list from public sources, race message angles against it, scale the one that wins. The nine steps from kickoff to booked meetings are written out in full.
What changes for this market
- Week one changes first. Before any list is built we settle which of the two audiences you sell to, because the list source, the calendar and the proof are different for each. That decision is the engagement, not a preliminary to it.
- List building reads the store, not just a database. Shipping policy, returns policy, country selector, subscription widget and checkout flow carry signals no vendor file contains, and a row without a dated trigger does not get sent.
- The send calendar is set from the buyer's calendar. For brand-side outreach that means pacing around the pre-peak code freeze and the January post-peak review. For wholesale it means line review windows, so send dates come off a review-window field on the row.
- The first ask is sized down. In a market where the buyer is handing over revenue-critical operations or shelf space, angles are built toward an audit, a sample pack or one test lane, not a demo.
What GrowthPal has published
| What was published | What the figure counts | Source |
|---|---|---|
| 4M+ cold emails sent for 40+ B2B companies | Emails sent, cumulative across all clients. Not one campaign. | Read it |
| 183 positive replies from 43,043 emails, April to June, across North America and Europe | Emails sent. | Read it |
| 78 positive replies from 53,508 emails over three months, reaching D2C brands in the US, UK and Canada | Emails sent. | Read it |
| A study of 412 campaigns across 47 companies over 18 months: campaigns scoring above 0.7 on the composite message-market fit score averaged an 11.4 percent reply rate, against 1.1 percent for campaigns below 0.3. Volume, warm-up and time of day explained less than 8 percent of the variance. | Total reply rate, measured against messages in those campaigns. This is a different measure from interested replies per contact and the two must never be compared on the same page. | Read it |
How these are counted
An interested reply means a human reply expressing interest in continuing the conversation. Out-of-office messages, unsubscribes, referrals to someone else and automated responses are not counted. A figure measured against contacts reached and a figure measured against emails sent are different measures: the column above says which each one is, and we do not convert between them or add them together.
The full counting method, and what an interested reply is.Who this is not for
- Consumer e-commerce marketing. This is business to business outbound. If you want more shoppers buying from your store, you want paid media, lifecycle email and retention work, not cold outbound to businesses.
- A market smaller than roughly 30,000 reachable contacts. Testing needs volume to tell a message that works from one that got lucky. Under that size, outbound still works, but learning is slower and a smaller, more manual approach usually beats a system.
- Before product-market fit. Outbound finds the people with a problem you already solve. It does not tell you what to build. If the offer is still moving every month, testing angles against it measures the wrong thing.
- No one ready to take the calls. Replies arrive in days once sends begin. If nobody owns the follow-up inside a day, interested buyers go cold and the spend is wasted.
- You want us to pick the audience for you. Audience A and audience B need different lists, different calendars and different proof. If the answer to which one you are is still open, that is a positioning conversation, not an outbound engagement.
Answered straight
Which of the two e-commerce markets are you selling into?
Both exist and they share almost nothing operationally. If you sell software, logistics or services to online brands, your list is built from store technology signals and your pacing bends around peak season. If you are a brand or distributor selling to retail and trade buyers, your list comes from store footprints, trade directories and show data, and your pacing bends around line review windows. Pick one before anyone writes a list.
Should we pause outbound during Black Friday and Cyber Monday?
Do not pause. Change the ask. Many merchants and vendors run a code freeze and publishing pause for roughly two weeks before Black Friday, with build windows often closing in mid-October. Anything requiring implementation will not be approved in that window. Keep sending, swap the implementation ask for a touch with no build attached, and schedule implementation-dependent angles to land either before the freeze or in January.
Our prospects are retail buyers. Why did a good campaign get no replies?
Usually timing rather than copy. Retail buyers work to line review windows and an open-to-buy budget, and many plan on the 4-5-4 merchandising calendar rather than ordinary months. A pitch outside the review window has nowhere to go even when the buyer likes it. The fix is a category field and a review-window field on every row, with send dates driven from that field instead of from a weekly sending schedule.
Can we just buy a list of store owners on one platform?
You can, and it is the most common way these campaigns fail. The platform a store runs on does not tell you its size, its category, whether it still trades, or whether anything changed recently. Prebuilt files are resold widely, go stale, and the addresses have been burned by earlier senders. We build from a dated trigger instead, and we keep that trigger and its public source on the row.
What proof do retail buyers actually want in a first email?
Three things, stated plainly. Sell-through evidence from a comparable store format. Your commercial terms, including minimum order quantity and lead time. Your compliance readiness, meaning GS1-licensed GTINs at item level and the EDI documents that retailer requires, with labeling that matches the advance ship notice. Claiming readiness you do not have creates chargebacks after the win, which is worse than losing the pitch.
What should the first ask be for an e-commerce buyer?
The smallest reversible step. Both audiences are deciding about revenue-critical operations or shelf space, so a demo of the whole thing is too large a first commitment. For software and services, ask to send a written audit or comparison. For logistics, ask for one test lane. For wholesale, ask to send a sample pack or have a line sheet reviewed. A meeting is the second ask, not the first.
When will we know if the message is working, given peak season?
Campaigns are sending two to four weeks after kickoff, and the gap is warmup, not setup drag. Expect to know your winning message around week eight. In e-commerce that clock has a seasonal caveat: if week eight lands inside the pre-peak freeze or outside a line review window, you are measuring the calendar rather than the message.
Our products are seasonal. Does outbound still work?
Yes, but the calendar does the work that volume usually does. For brand-side selling, implementation asks land before the pre-peak freeze or in January, and the freeze weeks carry touches with no build attached. For wholesale, send dates come off a line review window on each row. Seasonality is a reason to put a date field on the list, not a reason to go quiet for a quarter.
Related reading
Sources for this page
- FTC, CAN-SPAM Act: A Compliance Guide for Business (primary source). Accessed 2026-10-23.
- ICO, Guide to PECR: electronic mail marketing (primary source). Accessed 2026-10-23.
- Google Workspace Admin Help, Email sender guidelines (primary source). Accessed 2026-10-23.
- Shopify Help Center, B2B on Shopify (primary source). Accessed 2026-10-23.
- US Census Bureau, Quarterly Retail E-Commerce Sales (primary source). Accessed 2026-10-23.
- National Retail Federation, the 4-5-4 calendar (primary source). Accessed 2026-10-23.
Last reviewed 2026-10-23. A plain-text version of this page is at /ecommerce-cold-email-agency.md, and the site index for language models is at /llms.txt.
We score your market, map the signals we would build the list from, and tell you honestly whether outbound is the right tool for it. If you would rather see the mechanics first, the nine steps from kickoff to booked meetings are written out in full.