Cold Email Templates, One Per Buying Signal

A signal-keyed cold email template is a fixed body shape with one variable that matters: a dated event at the recipient's store. The shape can be written in advance. The event cannot, and it is the event that decides whether the message reads as observant or as the fourth near-identical email that founder received this week.

The disclosure belongs at the top rather than in a footnote. We build Keaz Signals, which sells ecommerce leads with live buying signals attached, and one of the things it does is draft per-store copy from those signals. We are therefore a competitor to anyone selling you a template pack, and we have an obvious interest in you concluding that the observation is worth more than the wording. The argument should survive being read that way.

It is worth being exact about the claim. Nothing here says templates are useless. A template is a solved structure — somebody has already worked out where the ask goes and how long the thing should be, and repeating that work from scratch for every email is a waste. What a template cannot do is tell you why this store, this week. That gap is where most template packs quietly hand the problem back: they give you seven ways to say something and no reason to say any of it today.

So this post is organised the other way round. It starts from seven events we detect on Shopify and WooCommerce stores, and gives each one the body it supports — the full message, not just the opening line. The subject-line half of the same problem is covered separately in cold email subject lines driven by a buying signal; that post handles the sentence in the preview pane, this one handles everything after it.

Most of it is right, and it was right long before we existed. The standard cold email template is short, written in sentence case, makes one ask instead of three, and does not oversell what the next email can deliver. That is the accumulated result of a great many people sending a great many emails, and a buying signal changes none of it.

Four conventions in particular are load-bearing, and every template further down obeys all four:

• Length. A first email that runs past a phone screen is asking for a commitment before it has earned one. Brevity here is a constraint of the medium, not a stylistic preference.
• One ask. Two asks in a first message is a request that the reader do your prioritising for you, and the usual answer to that request is nothing.
• Register. It should read like an email a person wrote to another person. A body that reads like a campaign gets filed as one, and no amount of merge-field personalisation rescues it.
• Filter hygiene. Certain words, link counts and formatting patterns cost placement. Unglamorous, well documented by the people who run sending infrastructure, and worth respecting whatever the message says.

We follow all four and we invented none of them. The disagreement is not about craft. It is about what the craft gets applied to. A template pack takes the list as given and optimises the sentences, which is a reasonable division of labour, and for many senders it is the only division available, because the list genuinely is given. The rest of this post is about what changes when it is not — a distinction covered from the tooling side in what sales prospecting tools actually do.

Four parts, in order: the observation, the connection to what you sell, one small ask, and nothing else. The fourth part is a real part. Most weak cold emails are not badly written — they are correctly written and then continued past the point where they should have stopped, and the continuation is where the guessing starts.

1. The observation. One sentence naming a dated event at their store, in terms the owner would recognise as true. It may not contain an interpretation. "Your active ad count went from 4 to 19 in two weeks" is an observation; "you are clearly scaling" is a conclusion, and a conclusion invites an argument the reader did not ask to have. The date matters as much as the fact: an event from nine days ago explains why the email arrived today, and a stable attribute like "runs Shopify" explains nothing at all.
2. The connection. One sentence saying what you do and why the event makes it relevant now. This is the part most likely to overreach, because the honest version is often modest: not "your launch will fail without creative support" but "launch weeks are most of our work". Facts about your own business are safe to assert. Facts about theirs are not yours to assert.
3. The ask. One, small, and cheaper for them than a meeting. "Want me to send it?" costs a two-word reply. "Do you have 30 minutes Thursday?" costs a calendar decision about a stranger. Ask for the reply you can actually earn on a first message, and let the meeting be the second thing.
4. Nothing else. No credentials paragraph, no client logo list, no second link, no PS with a discount. Each of those is a small withdrawal from the credibility the observation just deposited. A body that stops early reads as considered; a body that keeps going reads as a brochure that happened to know your ad count.

One test catches most failures before sending. Delete the observation and read what is left. If the remaining message would still make sense sent to any store in your segment, then the observation was decoration and the email is a generic email wearing a fact. A signal-keyed body should become incoherent without its signal — that is the property that makes it impossible to mass-produce, and being impossible to mass-produce is most of why it works.

Seven events are live on the signal catalogue as of September 2026, and each supports a different body. Below, each one is paired with the template it licenses — bracketed parts are where the store's own detail goes. Every body is under 90 words, carries at most one link, and states no conclusion about the recipient's business. The full catalogue, including the two events marked planned rather than live, is the buying signal catalogue.

1. New Meta ads running. The store had no active ads and now has some. That supports the fact of the launch and nothing about how it is going.

Hi [first name], saw [store] started running Meta ads again on [date] — [n] live right now. We make [service] for [category] brands, and the first fortnight of a fresh ad set is usually when we get asked to look at creative. I can send over the two things we would test, no call needed. Want them?

2. Active ad count rising. A number moved, and the movement is the fact. A count the recipient can check beats any adjective describing it.

Hi [first name], [store]'s active Meta ad count went from [x] to [y] over [period]. We do [service] for [category] brands. When the count climbs that fast, the thing we normally get called about is [the specific bottleneck you fix] — which may or may not be where you are. Happy to send two examples from a brand your size. Useful?

3. Product launch. Something new appeared in the catalogue. Naming the specific product is what demonstrates a person looked at the store rather than at a row in a database.

Hi [first name], the [product] page went live on [store] on [date]. We do [service] for [category] brands and launch weeks are most of our work. I wrote down three things I would change on that page — happy to send them across, no meeting required. Worth a look?

4. Email marketing tool installed. Klaviyo, Brevo or another retention tool went in. The install is observable; whether the setup is finished is not, so the template asks instead of assuming.

Hi [first name], [tool] went live on [store] around [date]. We build [flows/retention work] for [category] brands, usually in the first month after an install while the setup is still open. If yours is already finished, ignore this. If not, I can send the three flows we would build first — want them?

5. Social growth. Followers or posting cadence moved on Instagram or TikTok. Whether it was paid or organic is not visible to us, and the template says so rather than guessing.

Hi [first name], [store]'s [Instagram/TikTok] following moved from [x] to [y] in [month], and posting went to [n] a week. We turn that kind of attention into [what you do] for [category] brands. I have no idea whether it was paid or organic, and it does not change what I would suggest. Want the short version?

6. Newsletter activity. The sending rhythm changed — started, stopped, or shifted. The change is the fact; the cause belongs to the reader.

Hi [first name], [store]'s newsletter went from [cadence] to [cadence] since [month]. We run [service] for [category] brands. A change like that usually means either someone new owns email or the same person got busier. I do not know which, and only the second one is a problem I can help with. If it is the second one, say so and I will send what we would take over.

7. Store and site changes. The storefront was rebuilt or materially altered. The most self-evidently observed of the seven, and the hardest to fake.

Hi [first name], [store]'s storefront changed on [date] — [new theme / new checkout / new app]. We do [service] for [category] brands, and rebuilds are when [the specific thing you fix] tends to slip. I can send a short post-launch checklist of what I would look at first. Want it?

Two things are true of the set as a whole. None of the seven contains a claim about the recipient's business, and none of them survives being sent to a store where the event did not happen. Delete the first sentence from any of them and what remains is a generic email — which is the test from the previous section, applied. Three of these events have a page of their own explaining what is and is not visible in them: new Meta ads running, product launch and email marketing tool installed. How agencies work the first of them in practice is in how Meta ads agencies find stores worth pitching.

Because a template is the wrong container for the thing that matters. The seven bodies above are printed here to show what a signal-keyed message looks like, but the product does not hand you seven files to fill in. It drafts each message from the signal plus a knowledge base you write once, which is a different mechanism with a different failure mode.

The reason is a cost curve, not a philosophy. Writing a specific first line stopped being expensive some time ago. On our own sending, across 660,000 emails and as of August 2026, the language model cost worked out at roughly $0.007 per message — the drafting is close to free. What is not free, and never became free, is finding the data point worth writing about. A template pack optimises the cheap half of the problem and leaves the expensive half with you.

What keeps drafting-at-volume from becoming volume for its own sake is a set of rails that apply to every message, published on the sales agent page and worth reading as constraints on the templates above rather than as features:

• At least 90 percent of each message specific to that store.
• No more than 90 words.
• At most one link.
• No spam trigger words.
• Capacity-bounded sending, and an automatic stop on reply.

The 90-word cap is the one that shapes everything above it. A body that short has no room for a credentials paragraph, which is why "nothing else" appears as a structural part rather than as advice. The 90-percent rule is what a static template structurally cannot satisfy: a file with three brackets in it is mostly the same words sent to everyone, and the brackets do not change that arithmetic. And the per-market access cap follows from the same place — volume without these constraints is how a sending reputation dies, and on this setup it would be the customer's reputation rather than ours, since campaigns run in the customer's own Instantly workspace.

The pool the events come from is 4.1 million Shopify and WooCommerce stores across Europe and North America, re-scraped weekly, with 41,200 carrying a fresh signal in the last seven days — figures from the counter on our own site, last updated 22 August 2026 and retrieved 5 September 2026. Those numbers move, so the date is part of the number. What the records contain is described on the ecommerce leads database page. Where drafting sits relative to the rest of an outbound stack is covered in what an AI SDR actually automates, and the sending layer itself in what a sales engagement platform does.

On data protection the position stays deliberately narrow. We are GDPR-conscious by design and we will not claim more than that; responsibility for what you send stays with you. A template you paste into your own tool and a message drafted inside ours are the same obligation wearing different clothes. That is a structural point about where the duty sits, not a claim about anyone's compliance.

Often, and in more situations than we would like. A signal-keyed template is only available when a signal exists, and there are several common cases where it does not. In those cases the conventional template pack is the whole toolkit, and it is a decent toolkit. Six cases where it is the right answer and we are not:

• You sell to anyone other than ecommerce stores. The pool is Shopify and WooCommerce and nothing else. No event we detect helps you write to a dental practice, a law firm or a SaaS company, and no amount of store data changes that.
• Your market sits outside Europe and North America. We do not cover it, and we are not going to describe a thin pool as a pool.
• You work LinkedIn. We have no LinkedIn signals, no LinkedIn sending, and no plans to add either. If that is your channel, this is not a close call — and the template you need is a connection request, which is a different object entirely.
• You need the records outside the system. We do not export. Leads move into campaigns and stay there. If your plan is to take a list into your own sequencer and run these seven templates there, we are the wrong supplier, and you should treat that as a real dealbreaker rather than a detail.
• The event you want is one we have not shipped. Registry and firmographic triggers, and hiring signals, are marked planned on our own site — the polite way of saying they do not exist today. If your best opening line is "saw you are hiring a retention lead", we cannot give you that one, and an eighth template here would be fiction.
• No signal fires often enough in your segment. This is the one people discover late. If your buyers are a slice of stores where little changes month to month, seven excellent templates produce a trickle, and a trickle does not fill a calendar. Build the segment, look at the live count, and find that out in week one rather than month three.

There is also a limit no template touches, ours included. Relevance gets a message read. It does not create the capacity to handle what comes back. The clearest sign this approach fits is paid sales capacity that is not fully booked — if your calendar is filled by referrals, better cold email templates are not the purchase you need yet. The adjacent version of that question, buying people instead of software, is worked through in SDR outsourcing versus a signal-led agent.

A post with seven cold email templates in it is expected to end in a table of reply rates. There is none here, and the absence belongs in the text rather than in a footnote, because the reasoning is part of the argument.

We did not run the test. No A/B split across signal types, no measured lift from naming a date, no per-template reply rate. We could have produced a number that looked like one — the genre is full of them and very few carry a method — and it would have been the most persuasive dishonest sentence in the post.

Even a figure we did hold would not do the job being asked of it. A rate produced by our copy, sent to our market, from our sending setup, is evidence about those three things. Your offer, your audience and your infrastructure are all different, and a vendor controls at most one of the three. Presenting our number as your forecast would be a category error dressed as data. The same applies to the deliverability question one step downstream: we make no guarantee about inbox placement, and anyone who does is selling you something they do not control either.

There are no comparisons against other products here either — no claim that anyone else's templates perform worse, no coverage or freshness percentages about other databases. We have not measured them. A table of those figures either comes from someone who did the work and can show it, or from someone repeating a number they liked the look of, and ours would have been the second kind.

What survives the subtraction is a structural claim: a body the reader can verify outperforms one they cannot, because verification is the thing being borrowed when a stranger opens an email. The useful property of a structural claim is that you can test it yourself in a fortnight without trusting anyone's published figure. Signing up gives you 1,000 credits to spend inside segments you build yourself, as of September 2026 — enough to see whether a signal fires in your niche at a rate that matters, and to read the messages the agent drafts before any of them go anywhere. What each tier includes is on the pricing page.

• Keaz Signals product pages, retrieved 5 September 2026: /ecommerce-leads-database, /signals, /sales-agent and /pricing. Source for the seven live signal types, the two marked planned, the sending rails quoted in this article, the weekly re-scrape, the per-market access cap and the export limitation.
• Own pool figures, from the counter on keaz.ai, last updated 22 August 2026 and retrieved 5 September 2026: 4.1 million Shopify and WooCommerce stores under watch across Europe and North America, 41,200 carrying a fresh buying signal in the last seven days, 0 burned domains across all sending on our infrastructure. These numbers move; the date is part of the number.
• Drafting cost, from /sales-agent, as of August 2026: roughly $0.007 per message in language-model cost across 660,000 emails on our own sending. Our own figure, our own sending — not a benchmark for anyone else's.
• 1,000 credits on signup, to spend inside segments you build yourself. Stated on keaz.ai, retrieved 5 September 2026.
• Instantly, the platform our sending runs through: instantly.ai, retrieved 5 September 2026. Named here only as our own sending path; no measurement of their product is reported.
• Related reading on this blog: cold email subject lines driven by a buying signal, what sales prospecting tools actually do, what a sales engagement platform does, what an AI SDR actually automates, SDR outsourcing versus a signal-led agent and how Meta ads agencies find stores worth pitching.
• Measurement disclosure: no open rates, reply rates, A/B results, deliverability guarantees or competitor performance figures are cited anywhere in this article. We did not run those measurements, so we do not report them. The reasoning is in the section on the reply rates this post does not contain.