Outbound to data and analytics leaders: a playbook

This playbook starts from an assumption, not a finding: that a job post may name the tools a data team uses today. Nothing here shows that job posts reliably reveal a team's stack or its typical problems, so test the idea on a small batch. If a post does name a tool, a specific first email becomes possible: "I saw this tool on this page on this date."

This playbook covers how to do that without overclaiming. A job post is one page, written on one day. It is an observation, not a diagnosis. The rest of this guide is about keeping that difference visible in your research, your email and your follow-up.

Who buys

Start with your own closed deals, not a list of industries. Outlevo can import past deals from a CSV or your CRM and run a win/loss analysis that proposes ICP hypotheses [[1]](#sources). A CSV import takes up to 200,000 rows [[2]](#sources). The HubSpot connection reads pipeline, stage, amount and closed-lost reason from your deals [[3]](#sources). The Salesforce connection reads accounts and opportunities, and is a pilot that needs the Outlevo package installed in your org [[4]](#sources).

Two rules from the win/loss analysis apply to any "who buys" list:

It counts companies, not deals. Fewer than 10 won companies is labeled very low confidence [[5]](#sources).
Segments proposed by AI are labeled experimental until a person approves them [[5]](#sources).

If you have fewer than 10 won companies, treat everything below as a hypothesis to test in a small batch.

Which countries you contact is your decision. Outlevo asks which countries you sell into and never proposes markets [[5]](#sources).

This page does not list the industries or company sizes that buy from data teams. Your won and lost deals are the evidence to use.

Roles to reach

Build the title list from your own CRM, not from a template. Here is one process:

Open the won companies from your win/loss analysis [[5]](#sources).
Look at the contacts you dealt with at each one, and write down the data and analytics job titles that appear. HubSpot contacts carry a job title field [[3]](#sources).
Use those titles, and only those, as your first list. Add a title later only when a won or lost deal shows it.

If you import contacts from HubSpot, the fields include email, name, job title and email opt-out. An admin has to turn contact import on and state a purpose first [[3]](#sources).

Finding the people costs credits. Contact discovery costs 4 credits, volume contact discovery costs 2 credits, and email verification costs 1 credit [[6]](#sources).

Keep one rule for every role on the list: you need a reason for contacting that person that you can write down. A title alone is not a reason.

Triggers to watch

One vendor in outbound tooling lists its signal categories as "Job changes, funding rounds, executive hires, tech stack shifts, website visitors and G2 buying signals" [[7]](#sources). That shows tech-stack shifts are a recognized trigger category. This playbook works on the assumption that job posts are one place to look for them.

Here is how each trigger maps to this playbook:

Tech stack shift. A tool appears on a careers page that you did not see there before. This is the trigger the angle is built on. Record the page and the date.
Job change or executive hire. You may choose to write when someone new takes the role. If you do, keep the signal in the same dated record.
Funding round. Treat it as context for timing. It says nothing about stack.
Website visit. Outlevo's website-visitor script shows which companies visit, which pages they read and whether they fit your ICP. It never identifies a person and sets no cookie [[8]](#sources). It costs 1 credit per company identified per day, however many pages that company reads [[8]](#sources). A company visit is a reason to review the account. It is not a reason to email a named individual as though you watched them.

If you buy signals, a signal check costs 1 credit and a person signal costs 1 credit [[6]](#sources). For a view of vendors that sell this kind of data, see intent data tools: pricing compared and data and enrichment tools: pricing compared.

Name the tools you saw, and keep the proof

The angle only works if every tool you name is true. Use the same standard Outlevo applies to its own research.

Keep a fact only if you can quote it. Outlevo keeps a fact read from a website only when its quote is found on the page [[9]](#sources). Do the same by hand. If the job post says a tool by name, copy the line and the URL. If you inferred it from a skill list, do not write it as seen.

Label guesses as guesses. Outlevo labels guesses as hypotheses and never presents them as findings [[9]](#sources). Your email should do this in plain words. "I saw X" is a finding. "My guess is Y" is a hypothesis.

Require three dated observations before you treat an account as researched. A Readiness Card needs at least three observations, each with a source and a date [[9]](#sources). For this playbook, that could be:

A job post that names a tool.
A second page, such as a team page or a product page, that supports or contradicts it.
A trigger from the list above, with its date.

An account with one observation can still go on a list. It should not get a first email that claims knowledge of the stack.

Say less about problems than you want to. Write your problem statement from what your won customers told you, and phrase it as a question. Check it against your first batch before you rely on it.

Rules before you send

Read the primary text of each rule, and ask your counsel about your own case. These are the words from each source, checked on 2026-10-22.

United States. The FTC's CAN-SPAM guide says: "The law makes no exception for business-to-business email." [[10]](#sources)
United Kingdom. The ICO's PECR guidance says: "You can email or text any corporate body (a company, Scottish partnership, limited liability partnership or government body)." [[11]](#sources)
GDPR. Article 6(1)(f) covers "processing is necessary for the purposes of the legitimate interests pursued by the controller or by a third party" [[12]](#sources). Recital 47 adds: "The processing of personal data for direct marketing purposes may be regarded as carried out for a legitimate interest." [[13]](#sources)

LinkedIn needs separate attention. Its User Agreement, checked 2026-10-22, lists among the things you must not do: "Use bots or other unauthorized automated methods to access the Services, add or download contacts, send or redirect messages" [[14]](#sources). Read the agreement before you turn LinkedIn sending on. In Outlevo, each person connects their own LinkedIn account, and the invitation note is at most 300 characters [[15]](#sources). LinkedIn sending is switched on per campaign [[15]](#sources).

Set up sending

Mailbox rules apply to any sender. Google requires that "email senders who send more than 5,000 messages per day to Gmail accounts must meet the requirements in this section. Set up SPF and DKIM email authentication for your domain." [[16]](#sources)

With Outlevo and Smartlead, Outlevo writes and schedules the emails and Smartlead sends them from your mailboxes [[17]](#sources). Before a campaign sends, every mailbox must be connected and warmed up. Its domain must also have SPF, DKIM and DMARC checked within the last 7 days [[17]](#sources).

If you would rather not use your main domain, Outlevo can buy sending domains, create the mailboxes through Zapmail, set up DNS and warm them up [[18]](#sources). A new mailbox sends nothing until it is warm. That takes at least 14 days and usually about 3 weeks [[18]](#sources). Builder includes 1 domain and 2 mailboxes, and Growth includes 2 domains and 4 mailboxes [[18]](#sources). Plan your research while the mailboxes warm.

Other tools list similar deliverability features. Apollo.io, for one, lists "Domain warming, inbox rotation, and reputation monitoring" [[19]](#sources).

Outlevo's approval model applies to all of it. Nothing is sent outside a policy that a person approved. Policies set who can be contacted, on which channels, how often and with what budget [[20]](#sources). Every external action is checked against the approved policy before it runs and is recorded in a ledger [[20]](#sources). Kill switches pause a campaign, a channel or the whole workspace at once [[20]](#sources).

Sample first email (example)

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

Subject: [Tool name] on your careers page

Hi [First name],

Your careers page lists [tool name] in the [role title] posting I read on [date]. I'm writing because [one sentence on why your product relates to that tool or role].

My guess, and it is only a guess: [one-sentence hypothesis about a problem, phrased as something you are checking]. Is that close, or is the real issue somewhere else?

If it's useful, I can send [one specific thing: a short note, a checklist, a comparison]. If not, reply "no" and I won't write again.

[Your name]

[Your company]

Why it is built this way:

The first sentence is a finding. It names a page, a tool and a date, all of which you can quote back.
The second paragraph is labeled as a guess, so a wrong guess costs less trust.
It asks one question and offers one thing.
Every bracket must be filled from your dated record. If a bracket needs an invention to fill, the account is not ready for a first email.

After the first email

Replies need a fast, plain answer. Outlevo can answer interested replies with your Calendly link or your Cal.com link, with the person's name and email filled in [[21]](#sources) [[22]](#sources). On autopilot the answer goes at once. Otherwise it waits in Replies for you to send [[21]](#sources). Each booking with one of your contacts counts as a meeting booked and is written to your CRM when one is connected [[21]](#sources).

If a LinkedIn step is part of the sequence, Outlevo sends one message after the person accepts, and a reply stops that lead's emails [[15]](#sources). Alerts for hot leads and approvals can go to your phone through Telegram, and every alert also appears in the app [[23]](#sources).

For CRM hygiene, Outlevo writes its status, fit score and next step to fields it owns. Your standard HubSpot fields are never edited, and nothing is written until an owner or admin approves the field mapping [[3]](#sources).

If your team wants it in code

Some teams want outbound wired into their own pipelines. Everything in the app is available in the REST API, with a TypeScript SDK and an MCP server [[24]](#sources). Test keys run against a sandbox with 100 fake companies, so you can build without sending anything or spending credits [[25]](#sources). Every write takes an Idempotency-Key, so a retried request never runs twice [[25]](#sources).

Webhook deliveries carry an Outlevo-Signature header with a timestamp and an HMAC-SHA256 of the body [[26]](#sources). A failed delivery is retried after 1 minute, 5 minutes, 30 minutes, 2 hours and 12 hours [[26]](#sources). Each delivered event costs 1 credit. Failed attempts and test pings are free [[26]](#sources).

For comparison with a spreadsheet-and-enrichment approach, see Outlevo vs Clay. For a similar playbook aimed at another audience, see cold email to CISOs and outbound to HR and people leaders.

What it costs

Outlevo has four plans [[6]](#sources):

Free: USD 0 per month.
Builder: USD 29 per month with 320 credits a month.
Growth: USD 149 per month with 1650 credits a month.
Managed: from USD 999 per month, on a contract.

Qualification costs 3 credits, and deep research costs 30 credits [[6]](#sources). Add the contact discovery and verification costs from the roles section to estimate one account. Connecting and syncing a CRM uses no credits [[3]](#sources). Sending, follow-ups and the mailboxes your plan includes use no credits [[17]](#sources).

Next step

Before you write to anyone, find out what your own site and your own deals say about who buys. The free assessment 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 [[27]](#sources). It also asks which countries you sell into [[28]](#sources).

Run it at the free assessment, then compare its segment hypotheses with your won and lost deals before you build a data-leader list.