How Should You Build an Onboarding Email Sequence for B2B Trial Users?

What should your trial onboarding emails actually say?

Each one should get the user to do exactly one thing that makes the product work for them, and nothing else. Most onboarding sequences fail because they describe features in the order the product team built them rather than in the order a new user needs them, which is almost never the same order.

A trial is a short window in which someone decides whether your product is going to be part of their week. The emails are not marketing during that window. They are a second interface, running alongside the product, doing the job your empty states cannot do alone.

Here is how I structure the sequence, what has to be in each message, and the deliverability and legal requirements that will quietly sink the whole thing if you skip them.

What is the job of an onboarding sequence?

To get one person to one moment of real value, as fast as their situation allows. Not to explain the product. Not to build a relationship. One action, then the next one, each chosen because it is the next thing that has to be true for the product to be useful to them.

This means you have to know what that moment is, expressed as something the user does rather than something they feel. Imported their data. Connected their calendar. Invited a teammate. Ran their first report. If your team cannot agree on which action that is, the sequence cannot be written, and no amount of copy craft will cover the gap. Nor will an automation on top of it, which is the argument I made in when to keep a human in the loop.

Everything else people put in onboarding emails is optional. The founder's welcome note, the case study, the webinar invitation: none of these are wrong, and all of them are competing with the one message that actually determines whether this trial converts.

How many emails should there be, and what should trigger them?

Fewer than you think, triggered by behavior rather than by the calendar wherever you can manage it. A time-based sequence sends the same email to someone who has already succeeded and someone who has not opened the product, and it is wrong for both.

The structure I use is small. One email on signup that names the single next action. One that fires when they complete it, confirming and pointing at the next thing. One that fires when they do not complete it within a sensible window, asking a question rather than repeating the instruction. And one before the trial ends that is honest about what happens next.

Behavior-based branching sounds like more work and is usually less, because each email gets shorter when it only has to serve people in one state. If your tooling makes branching genuinely hard, then send fewer time-based emails rather than more, and make the first one carry the weight.

The trigger design here is the same discipline as any other routing problem, and it fails in the same way when the underlying data is not populated. I worked through that in detail in routing inbound leads to the right owner in HubSpot.

What goes in each email?

One action, one reason, one link. The reason is the part people cut and it is the part that works. Telling someone to connect their data source is an instruction. Telling them that connecting it is what makes the dashboard show their own numbers instead of a demo is a reason, and reasons survive a distracted reading.

Write the subject line for the inbox, not for the email. A new trial user has your product name fresh in their mind, so the subject can be plain and specific rather than clever. Clever subject lines are for people who have forgotten who you are, which is a different problem in a different sequence.

Keep the sender human and consistent. From a person, replying to an address a person reads. The single highest-value thing I have seen in trial onboarding is a reply from a confused user reaching someone who can actually help, and that only happens if the reply-to is real.

What are the deliverability and legal requirements you cannot skip?

Three sets, and none of them are optional. On deliverability, Google's sender guidelines ask all senders to keep spam rates reported in Postmaster Tools below 0.3 percent, and to set up SPF or DKIM authentication for their sending domains.

Google defines bulk senders as those sending more than 5,000 messages per day to Gmail accounts, and requires more of them: SPF and DKIM together, plus DMARC for the sending domain. Bulk senders must also support one-click unsubscribe on marketing and subscribed messages, with a clearly visible unsubscribe link, implemented through the List-Unsubscribe-Post and List-Unsubscribe headers.

On the legal side, the Federal Trade Commission's CAN-SPAM compliance guide states that you must honor a recipient's opt-out request within 10 business days, and that your opt-out mechanism must be able to process requests for at least 30 days after you send the message. It also requires accurate From, To, Reply-To and routing information, a subject line that accurately reflects the content, disclosure that the message is an advertisement where that applies, and a valid physical postal address.

The consequences are not theoretical. The same FTC guide states that each separate email in violation of the law is subject to penalties of up to $53,088. That figure is per email, not per campaign, which is the detail that tends to focus attention.

Why does the unsubscribe experience matter more than you think?

Because the alternative to an easy unsubscribe is a spam complaint, and spam complaints are the thing that actually damages you. A person who cannot find the unsubscribe link does not resign themselves to your emails. They press the button that tells Gmail you are a problem.

Set against Google's 0.3 percent threshold, that arithmetic is unforgiving. It does not take many frustrated recipients to move a rate expressed in tenths of a percent, and once deliverability degrades it degrades for everyone on your list, including the customers who wanted to hear from you.

So make unsubscribing genuinely easy, implement the one-click headers even if you are below the bulk sender threshold, and resist every dark pattern anyone proposes. Preference centers that require a login, unsubscribe links in four-point grey, confirmation flows with three steps: all of these trade a small retention gain for a deliverability risk that compounds.

How do you write emails people actually open?

By earning it in the previous email. Open rates are largely a function of whether the last message was worth reading, which means the first email in a trial sequence is doing more work than any subject line optimization you will ever run.

Write short. Not artificially clipped, but genuinely short, because a trial user is reading this between other tasks and a long email is a task they will defer. If you cannot say it in a hundred and fifty words, the email is trying to do more than one job.

And write specifically. Generic onboarding copy is interchangeable across every product in your category, which is a strange thing to send someone who is currently deciding whether your product is different. Name the actual screen, the actual button, the actual outcome.

What should you measure?

Completion of the action, not engagement with the email. An email with a low open rate that moves people to connect their data source is doing its job. An email with a high open rate that moves nobody is a well-written distraction. The same instinct applies to signup forms, which I covered in what actually makes a signup convert.

Track the proportion of trial users who reach the key action, split by whether they received each email. That is the only measurement that tells you whether the sequence is causing anything, and it is very different from the numbers your email platform shows by default.

Then watch your spam rate and your unsubscribe rate as guardrails rather than as goals. Across 70 plus projects for 25 plus clients over 6 plus years, the sequences that got quietly worse were the ones where nobody was watching the guardrails because the conversion number looked fine.

What should you do next?

Write down the single action that predicts whether a trial converts, in the language of something the user does. Then check what proportion of your trial users currently do it, and how long it takes them. Those two numbers define the entire sequence, and most teams have never looked at them together.

Then check your authentication and your unsubscribe implementation before you write a word of copy. SPF, DKIM, DMARC, a visible unsubscribe link, one-click headers, a real reply-to, a physical address, and an opt-out process that completes well inside 10 business days. A sequence that does not arrive cannot convert.

I build lifecycle and onboarding automation for founders and marketing teams, including HubSpot work through Zapier for Kismet Health, on fixed fees, with most projects landing between 1,000 and 10,000 dollars. If your trial users are signing up and going quiet, reach out and let's chat.