Outbound for developer tools: a cold email playbook

Selling developer tools by email comes down to one question. What can you say to a stranger that is specific enough to be worth reading? This playbook builds the answer from evidence you can check. It sets up sending so the email arrives and keeps the legal basics in view.

It does not claim to know how developers or engineering managers behave in their inboxes. Your own reply data will settle that. What follows is the process, and the process is the same whoever you write to: one observation, one ask, and a record of where the observation came from.

If you want the product-side view first, see Outbound for developer tools companies.

The approach: one observation, one ask

Write about one thing you can point to. Read it on the prospect's public page, with the page open in front of you. Quote it or paraphrase it closely, and ask one question about it.

Treat a guess as a guess. If you think a company is hitting a problem but cannot show it, phrase it as a question rather than a statement. Outlevo works the same way: facts read from a website are kept only when their quote is found on the page, and guesses are labeled as hypotheses, never presented as findings [[1]](#sources).

That rule is also a writing rule. An email built on one verified detail is short by default. It has nothing to pad.

Who buys

This guide cannot tell you who buys your tool. Your closed deals can.

Start there if you have them. You can import past deals from a CSV or your CRM and get a win/loss analysis that proposes ICP hypotheses [[2]](#sources). A CSV import takes up to 200,000 rows [[3]](#sources). HubSpot and Salesforce are the CRM routes:

HubSpot: Outlevo reads your companies, deals and owners [[4]](#sources).
Salesforce: Outlevo reads accounts, opportunities and users [[5]](#sources). This is a pilot. The connection needs the Outlevo package in your org, which Outlevo installs with you [[5]](#sources).

The analysis counts companies, not deals. It also labels small samples: under 10 won companies is very low confidence [[6]](#sources). If you have fewer won customers than that, read any pattern as a hypothesis. Segments proposed by AI are labeled experimental until a person approves them [[6]](#sources).

If you have no deal history yet, the free assessment is the other starting point. It reads up to 5 public pages of your website. It returns facts labeled as verified observations or AI inferences, and up to 5 experimental segment hypotheses [[7]](#sources). It also asks which countries you sell into [[8]](#sources). Outlevo never proposes markets [[6]](#sources). That choice stays with you.

Roles to reach

There is no title list here, because none can be backed up. Your won deals show which roles signed, evaluated or blocked. Use them.

Some practical points on getting role data:

A HubSpot import can include contact email, name, job title and email opt-out. It only does so after an admin turns contact import on with a stated purpose [[4]](#sources).
Hot leads go to the right person: HubSpot owners tell Outlevo who owns each company [[4]](#sources), and Salesforce users route leads to the account owner [[5]](#sources).
Finding contacts costs credits. Contact discovery is 4 credits, and volume contact discovery is 2 credits [[9]](#sources). Email verification is 1 credit [[9]](#sources).

Verify before you send.

If your product also reaches revenue or finance teams, the structure of Outbound to RevOps leaders at SaaS companies and Cold email to finance leaders may help you plan a second list.

Triggers to watch

A trigger is a reason to write this week rather than next quarter. Which triggers matter for developer tools is something your own deals will show. The ones below are what Outlevo can observe.

A company visits your site. Add one script and Outlevo shows which companies visit, which pages they read and whether they fit your ICP. It never identifies a person [[10]](#sources). It looks up the organization behind a visitor's network, such as their employer, and sets no cookie [[10]](#sources). Browsers that send Do Not Track or Global Privacy Control, and automated browsers, send nothing [[10]](#sources). The cost is 1 credit per company identified per day, however many pages it reads [[10]](#sources).

A competitor changes. Competitor tracking keeps quoted facts from competitors' public pages, records changes, shows win/loss by competitor and adds competitor signals to accounts [[11]](#sources). A competitor search costs 15 credits [[9]](#sources).

A fresh signal on an account. A signal check costs 1 credit [[9]](#sources).

Whatever the trigger, it only helps the email if you can name it in one plain sentence. If you cannot, do not use it.

Build the evidence before you write

A Readiness Card needs at least three observations, each with a source and a date [[1]](#sources). Hold your own notes to the same bar. Before writing to an account, you should have:

One observation you can quote, with the page and the date you read it.
One reason it relates to what you sell.
One question that a person could answer in a line.

Three levels of research cost different amounts of credits. Light qualification costs 1 credit, qualification costs 3 and deep research costs 30 [[9]](#sources). A sensible order is to qualify widely and cheaply, then spend deep research only on accounts that pass.

Set up sending before the first email

See the deliverability entry for the general idea. The specifics below come from the sending setups Outlevo supports.

If you bring Smartlead. Connect your key and Outlevo builds each campaign in Smartlead, sends from your mailboxes and reads replies and bounces back by webhook [[12]](#sources). Before a campaign sends, every mailbox must be connected and warmed up. Its domain must also have SPF, DKIM and DMARC checked in the last 7 days [[12]](#sources). See the Smartlead integration.

If you want managed mailboxes. Outlevo buys sending domains, creates the mailboxes through Zapmail, sets up DNS and warms them up, so you never touch your main domain [[13]](#sources). A new mailbox sends nothing until it is warm: at least 14 days, usually about 3 weeks [[13]](#sources). Plan the wait into your launch date. Builder includes 1 domain and 2 mailboxes. Growth includes 2 domains and 4 mailboxes [[13]](#sources). See managed mailboxes.

Control. Nothing is sent outside a policy that a person approved [[14]](#sources). Policies set who can be contacted, on which channels, how often and with what budget [[14]](#sources). Kill switches pause a campaign, a channel or the whole workspace at once [[14]](#sources). Read more on how Outlevo works.

Rules that apply to B2B email

This is a summary, not legal advice. Check your own situation with counsel.

CAN-SPAM. The FTC's guide says: "The law makes no exception for business-to-business email." [[15]](#sources)
Unsubscribe. Google's sender guidelines say: "Marketing messages and subscribed messages must support one-click unsubscribe, and include a clearly visible unsubscribe link in the message body." [[16]](#sources)
GDPR. Article 6(1)(f) covers processing that is "necessary for the purposes of the legitimate interests pursued by the controller or by a third party" [[17]](#sources). Recital 47 says: "The processing of personal data for direct marketing purposes may be regarded as carried out for a legitimate interest." [[18]](#sources)
UK PECR. The ICO's guidance says: "You can email or text any corporate body (a company, Scottish partnership, limited liability partnership or government body)." [[19]](#sources)
LinkedIn. LinkedIn's User Agreement lists among its don'ts: "Use bots or other unauthorized automated methods to access the Services, add or download contacts, send or redirect messages" [[20]](#sources). Read that clause before you turn on automatic sending.

Each of these has conditions beyond the lines quoted. The quotes show where to start reading, not the full rule.

Sample first email (example)

Example only: written for this playbook, not sent to anyone.

Subject: Your docs page on [feature name]

Hi [First name],

I was reading your docs page on [feature name]. It says "[exact line from the page]". I read it on [date].

That made me wonder how your team handles [one specific, related question].

We build [one plain sentence about your product]. If the question is live for you, I can send a two-line example of how we approach it.

If not, no reply needed.

[Sender name]

[Sender role], [Your company]

Unsubscribe: [one-click unsubscribe link]

Why it is built this way:

The subject names the page. It points at the thing you read, not at a benefit.
The quote is exact and dated. It matches the evidence rule above, so the reader can check it in seconds.
There is one question and one small offer. No call is requested in the first email.
The unsubscribe link is in the body. That follows the Google guidance quoted earlier [[16]](#sources).
Nothing in it is a claim you cannot support. Fill each bracket only from your notes.

Follow-ups and a second channel

Keep the sequence short and make each step add something new. A follow-up that restates the first email adds nothing. A second observation from the same account does.

If you add LinkedIn, Outlevo sends a connection invitation with a short note, then one message after the person accepts [[21]](#sources). The invitation note is at most 300 characters [[21]](#sources). A reply stops that lead's emails [[21]](#sources). You sign in on Unipile's page, and Outlevo never sees or stores your LinkedIn password [[21]](#sources). Each person connects their own account [[21]](#sources).

Automatic LinkedIn sending is included on Growth and Managed. On other plans the invitations and messages are drafted for you to send by hand [[21]](#sources). See the LinkedIn integration.

When someone is interested, Outlevo can answer with your Calendly or Cal.com link, with their name and email filled in [[22]](#sources) [[23]](#sources). On autopilot the answer goes at once. Otherwise it waits in Replies for you to send [[22]](#sources).

Setting expectations for replies

Belkins' 2026 cold email study covers 7.5M B2B emails sent in 2025, with reply rate defined as replies divided by emails sent [[24]](#sources). Across industries, 0.45% of emails get a reply [[25]](#sources).

This guide has no developer-tools figure to offer. Measure your own, by segment and by trigger, and compare each batch to the last one. Try changing one variable at a time, so you can see what moved the result.

Working in code

If your team wants the workflow in its own stack, Outlevo's developer hub covers this:

Everything in the app is available in the REST API, with a TypeScript SDK and an MCP server [[26]](#sources).
The public API has 177 paths across resources that include campaigns, companies, contacts, signals and webhooks [[27]](#sources).
Test keys act on a sandbox with 100 fake companies, so you can build without sending anything or spending credits [[28]](#sources).
Every write takes an Idempotency-Key, so a retried request never runs twice [[28]](#sources).
Metered calls quote their credit cost first and run only when confirmed [[26]](#sources).
Webhooks are signed with an HMAC-SHA256 header and retried for up to 12 hours [[29]](#sources). A delivered event costs 1 credit, while failed attempts and test pings are free [[29]](#sources).

If you would rather not write code, n8n, Zapier and Make connect through the same API and signed webhooks [[30]](#sources) [[31]](#sources) [[32]](#sources).

What it costs

Plans are Free at USD 0 a month, Builder at USD 29 a month with 320 credits, and Growth at USD 149 a month with 1650 credits. Managed is priced per contract [[9]](#sources). Connecting and syncing a CRM uses no credits [[4]](#sources). Sending, follow-ups and the mailboxes your plan includes use no credits [[12]](#sources). Full detail is on the pricing page.

Next step

Run the free ICP assessment. It reads up to 5 public pages of your site and returns facts labeled as verified observations or AI inferences, plus up to 5 experimental segment hypotheses [[7]](#sources). It also asks which countries you sell into [[8]](#sources). Use the hypotheses to pick your first list. Write to it with one observation per account.