The outbound scope nobody writes down
Most outbound engagements fail in the same boring place: nobody wrote down what was included, so both sides spend the retainer arguing about it. The list was "included". The verification was "obviously included". The tracking was "the agency's job". Nobody owned the definition of a lead, so nothing could be judged.
A scope is not a sales document. It is the list of things that must exist on day one for a number to exist on day thirty. Here is the version I work from, with the parts people usually leave out.
1. The list, and who owns the file
The most common hidden cost in outbound is a list rented by the month. When the contract ends, the pipeline ends — you cannot re-run the segment, cannot split it, cannot hand it to a new vendor.
So the first line of the scope is ownership. Who holds the file, in what format, and does it survive the engagement. A working list is not a spreadsheet of names. It is: company, domain, contact, role, the source the contact came from, the date it was verified, and the reason this company fits.
Two of those fields are usually missing: the source and the verification date. Both are the difference between a segment you can rebuild and a pile you cannot defend.
2. Verification, and what happens to the ones that fail
A verification pass is not a filter you run once. It is a decision written before the first send:
- what happens to a catch-all domain,
- what happens to a role address (
info@,sales@) versus a named person, - what happens to a contact verified 40 days ago,
- and what the bounce ceiling is before the campaign pauses.
Without those four decisions, deliverability is a matter of luck and the first bad batch is discovered by the recipients, not by you.
The failure mode is arithmetic, not opinion: if the address book is not verified at entry, you pay for the same cleaning twice, and the second time it lands in the reputation of your sending domain.
3. Deliverability, before volume
Domain warm-up, SPF/DKIM/DMARC, sending limits per mailbox, and one rule that is worth more than the rest: a fixed number of sends per day per mailbox, never increased because a batch "felt good".
Volume is where an outbound programme stops being a test and becomes a subscription. Write the ceiling down before the first send, and treat any exception as a decision with a name attached.
4. The events that count
"If someone is interested, we will know" is not measurement. The scope should name the events in the order they happen: delivered, opened (with the caveat that open rates are unreliable), replied, meeting booked, qualified per the definition both sides signed, and closed.
The definition of qualified is the one number that must be agreed in writing, because it is the number every channel gets judged by afterwards. Two teams with different definitions of a lead will read the same dashboard and reach opposite conclusions.
5. The read date
A test with no read date is a hobby. Before the work starts, write the date the first verdict is due and the threshold that makes it a verdict.
The arithmetic is unforgiving and worth doing early. A strict two-variant test at a 3% base rate needs 13,914 contacts per arm — 27,828 in total — at 80% power and alpha 0.05. At 500 contacts a day that is roughly 56 days of sending. A campaign that reports a winner on day 12 has not found a winner; it has found noise.
That does not make small campaigns pointless. It makes them something else: a diagnostic. A 300–500 contact pass tells you whether the offer, the subject line and the segment are alive — it just cannot tell you which of two variants is better. The sample-size planner runs the same two-proportion formula in the browser, so the deadline in the scope and the deadline in the plan agree.
6. The stop rule
The cheapest line in any scope. Written before the spend, naming the channel, the threshold and the person allowed to switch it off.
For a paid test the rule is blunt: 1–3× target cost per qualified conversation with no readable return in 48–72 hours. For owned channels it is the read date plus a threshold. A channel that can never be switched off is not being tested, it is being funded.
7. What is not included
A scope that never says no is a scope that will be renegotiated. The exclusions I write down in advance:
- co-managing a round of fundraising, or anything that requires being in the room with your board;
- writing product positioning from scratch — the scope can sharpen a claim, not invent one;
- press and influencer relations as a line item; if a launch needs them, they are their own project;
- being the CRM administrator of record. Configuration yes, ongoing data janitorial work no;
- "and community" — an unbounded channel with no read date is how the retainer quietly becomes 24/7.
The exclusions are not modesty. They are what keeps the read date honest: every hour spent on out-of-scope work is an hour not spent on the number the invoice is supposed to move.
8. The handover
The last line, written on day one: what remains if we stop.
Pages, tracking, the automation, the credentials, the definitions, and the written record of what was killed and why. The test of a scope is whether the next person can pick it up without a call to you.
What it costs, in public
Being specific is cheaper than being vague. Fixed and published: Sprint $900 for one focused build in five working days, Engine $1,900/month for a running queue of experiments with a monthly read-out, Full Build $2,900 for site architecture, the page set, the machine-readable layer, tooling, distribution and handover. The full scope line by line is on the scope page and the cost breakdown explains what lands inside each one.
Copy the eight lines above into your own scope document, whoever you end up hiring. The engagement that survives is the one where both sides can point at the same sentence.
Top comments (0)