Launch checklist template for B2B SaaS

The launch checklist with the rows a subscription business needs before demand generation starts: the lead route tested end to end, the account list loaded and versioned, sales told what to expect, and the gated asset actually delivering.

Everything else is the general checklist, unchanged. One row per check, with an owner by name, a due date relative to the start, an honest blocking flag, a verification method and evidence kept.

Use it before every quarter's demand generation and before any campaign that produces leads a human has to follow up. The failures here are downstream of the ad account, in the form, the routing and the sales process, and none of them is visible in a platform's reporting.

Use it again whenever the routing or the form changes, which is more often than anybody expects.

In this category the campaign rarely breaks. What breaks is everything between the click and the conversation: the form, the routing, the sales system, the person who was supposed to call.

None of that appears in a platform's reporting. The ads will show a healthy cost per lead while the leads themselves sit unassigned in a queue, lose their source fields to a redirect, or arrive at an owner who is on leave. Four weeks later the campaign is judged on a conversion rate it did not cause.

So the checks that matter here point outside the ad account, and most of them are tests rather than inspections. The rule is the same one that makes any checklist work: a check you can tick without doing anything is not a check.

The lead route row is the most important in the list, and it is verified by submitting the form as an outsider would and then following the record.

Every part of the journey is a place it has failed somewhere. The form submits but the record is not created. The record is created but has no owner. It has an owner who left. The source fields are empty because a redirect stripped the parameters. The record lands in a queue that is reviewed weekly rather than daily. The notification goes to a channel nobody watches any more.

Do the whole thing, keep the screenshot, and repeat it whenever the form or the routing changes — which is more often than anybody expects, because both are edited by people who are not thinking about this campaign.

Where a campaign works from an account list, the check is that the list is loaded, that it matches the version named in the plan, and that the version is recorded at launch.

The reason is that lists change quietly. A refresh mid-flight alters who the campaign reaches without altering anything visible in the account: same campaign, same budget, same creative, different market. Weeks later the cost per lead has moved and nobody can explain it.

Recording the version at launch is what makes that detectable afterwards. It costs one cell.

The sales readiness row is the cheapest high-value check in this document.

It is not a process: it is a message saying what is starting, which segment it reaches, roughly how many leads to expect and when the first ones will arrive. Sent before the launch, it means the leads are handled by somebody expecting them. Not sent, it means the first week's leads are called late by whoever is free, and the campaign is judged on that.

Where a campaign will deliberately produce nothing for a fortnight — an awareness phase, a long-form asset — say that too. Silence is interpreted as failure by people who were told something was starting.

Two checks here and both get missed.

First, the gated asset actually arrives. Complete the form and confirm the document reaches an ordinary inbox, with the right file, from an address that does not look like a system. A campaign whose asset silently fails to deliver produces contacts who received nothing and a support problem nobody attributes to the campaign.

Second, where the brief promised an open summary, confirm it is genuinely open. Gating creeps: somebody optimising a page adds a form to the summary, and the campaign that promised an open read becomes one that demands an email for the abstract. That is the fastest way to lose a first-time reader, and it is usually done by somebody who never saw the brief.

A quieter check worth carrying: confirm that the reporting window somebody will look at is long enough to contain anything.

In a business with a long cycle, a default window will show a campaign producing almost no accepted leads for its first month, which is true and is also a fact about the window rather than the campaign. Agreeing before the launch which report will be read, over which period, prevents a well-run campaign being cut at week three by somebody looking at the wrong view.

Write it in the row, and send the same view to the people who will ask.

A quiet category of failure in this business is access: an ad account owned by an agency that has been replaced, a payment method on a card belonging to somebody who left, a page with one administrator on leave, a tracking property nobody can edit.

None of these stop a launch on the day it is planned. They stop it on the day something has to change, which is usually the day something is going wrong.

So the list carries an access row a week out: confirm that at least two current people can administer each account, page and property the campaign touches, and that the payment method will not expire during the flight. It is dull, it takes ten minutes, and it is the difference between a fixable problem and a two-day one.

Seven checks. Every row has an owner by name and a due date relative to the start. A test lead has travelled the entire route and arrived with its owner and its source intact. The account list is loaded, versioned and recorded. Sales knows what is coming and when. The gated asset delivers to an ordinary inbox, and anything promised as open is open. The reporting view and its period are agreed. And every completed row has evidence attached, not only a status.