Playbook 02
How I build pipeline with ads and outbound
Paid ads and cold outreach run as one engine, built backwards from the revenue number.
The short version
- Count backwards from the revenue number.Revenue to deals, deals to meetings, meetings to replies and leads. That chain is the plan and the budget.
- Fix the list before you write a word.One company is one target, every address verified. The best email cannot save the wrong list.
- Open with a change in the buyer's market, not your product.True, recent and mostly unknown to the reader. That earns the reply.
- Answer every positive reply the same day.The sequence earns the reply. The reply earns the meeting.
- Tell the ad platform that a win is a lead.Not a click, not a page view. It buys exactly what you ask for.
- Read the numbers in order and fix the first broken step.Everything below a broken step is noise.
Start from the number
Most teams start with activity and hope it adds up. I start with the revenue number and count backwards until I know how many emails and leads it takes.
I built my pipeline at DJI Agriculture this way: paid ads plus a 16 mailbox outbound engine over a list of 10,000+ dealer contacts.
Example, round numbers: $100,000 a month at $20,000 a deal.
| The link | The math |
|---|---|
| Deals | $100,000 ÷ $20,000 = 5 |
| Qualified meetings | 5 ÷ 20% close rate = 25: 15 from outbound, 10 from ads |
| Positive replies | 15 ÷ 50% that book = 30 |
| Contacts to email | 30 ÷ 1.5% who reply with interest = 2,000 |
| Emails, mailboxes | 2,000 × 2 emails = 4,000. At 25 a day per mailbox, 8 mailboxes |
| Ad leads, budget | 10 ÷ 25% that book = 40 leads, $1,800 at $45 each |
The number people skip is the 2,000. That is new contacts every month, so a list of 10,000 lasts 5 months. If the list runs out before the year does, the plan is wrong, and week 1 is the cheapest time to learn it.
The mistake. Keeping borrowed rates. Every rate above is a placeholder. After 200 sends and a week of ad data, put in your own and run the chain again.
Build the list first
The list decides the result before a word is written. Four moves, in order.
- One company is one target. Group every row by company domain. A chain with 28 branches is one head office. Look it up 28 times and the same 2 people get 28 emails, the pattern that gets a domain blocked.
- Three people per company. The owner, a manager in sales or operations, and the general inbox.
- Verify every address. Only addresses confirmed deliverable are sent. Catch-all domains, which accept anything, are dropped, not saved for later. That keeps bounces under 3%.
- Pilot 50 companies. Measure how often you find a real decision maker and what each verified contact costs, then multiply by the list before you pay for the rest.
Before I load the list, I grade it: 100% verified, under 1% duplicates, 3 contacts per company at most, under 2% junk titles such as assistant or intern.
Example. A chain is the best pitch on the list, not a problem: "You have 24 locations. We can supply all of them: one supplier, one standard, rolled out at your pace."
The mistake. Rewriting the copy when the list is wrong. With the wrong people on it, no email works.
Write the messages
Three people at one company get three different emails. Colleagues forward each other mail, and identical copy reads as spam to them and to the filters. The ask shrinks with the title.
| Who | The angle | The ask |
|---|---|---|
| Owner or president | The market is changing, and what it means for them | A 15 minute call |
| Sales or operations manager | Margin and the day to day | "Worth showing to [owner's first name]?" |
| The general inbox | Neutral and businesslike | "Who handles new suppliers?" |
- Open with the market shift. A rule change, a deadline, a cost moving. Example: a regulator approves a new use for a product your buyers stock, and most of them have not heard.
- Under 100 words, 2 numbers at most, 1 ask, and the ask is a question.
- Claim nothing you cannot show in writing. Merge only verified fields: if one is empty, its sentence does not send.
Subject: {{the change}}, {{month}}
Hey {{first name}},
{{One sentence on the change in their market, with its date.}} Most {{their kind of company}} have not adjusted yet.
{{One sentence on what you do about it, with one number.}}
Worth 15 minutes next week to see if it fits {{company}}?
{{Your name}}
{{First name}}, one number I left out: {{the number, and what it means for them}}.
If this is not a fit, say so and I will close the file.
The mistake. A follow up that says "just checking in". It carries a new fact or it does not go. One follow up per contact is enough: with 3 people that is already 6 emails into one building. The last line, the easy way to say no, is the one that gets answered.
Send without burning the domain
A good email from a burned domain reaches nobody. These rules do not bend.
- Never from your real domain. 4 to 5 separate sending domains, with SPF, DKIM and DMARC, the 3 records that prove the mail is yours, checked before the first email.
- 25 emails per mailbox per day at most, 8 to 10 in week 1. New mailboxes need 10 to 14 days of warm-up, so order them first.
- Stagger each company. Owner on day 0, manager on day 2, general inbox on day 4, each from a different mailbox. Three emails into one company in one day is what filters catch.
- One reply stops the whole company. A no from the owner followed by an email to their manager is how complaints start. On a firm no, block the domain.
- Business hours in their time zone, weekdays only, with an unsubscribe link and a real mailing address in every footer. CASL in Canada and CAN-SPAM in the US both require them.
- Best list first. Real replies build sending reputation faster than any warm-up tool.
Example budget, round numbers: 16 mailboxes at about $10 a month each, a sending platform at $100 to $200, a month of a data tool at about $200. About $500 in month 1, $300 a month after. I use Clay for the data and Instantly to send.
The mistake. Buying mailboxes to fix weak replies. Mailboxes add volume. They do not fix the list or the message.
Answer replies the same day
I judge a campaign on positive replies, not replies. 20 unsubscribes is a reply rate too.
| The reply | What I do |
|---|---|
| Interest, or a question | Answer it in the first line, then offer the next step. Same business day, within the hour when I can |
| "Talk to someone else" | Email that person the same day and name who sent me |
| An objection | One honest answer: concede what is true, add the fact they are missing |
| "Not now" | Log the date they gave and come back on it |
| No | One gracious line. Stop the whole company. Never argue |
| Unsubscribe | Off every list, for good. That is the law |
The mistake. Answering interest with a calendar link and nothing else. They asked a question. And when one objection arrives 3 times, the copy or the offer is wrong, not the prospects.
Run the ads
On Meta you are not choosing an audience. You are teaching the algorithm what a win looks like, and it chases exactly that.
Example. Tell it to buy page views and it hands you 1,000 page views and 2 leads. It did its job. So I optimize for the lead, and if the lead event does not fire every time, I fix that before I spend a dollar.
- Feed the learning phase. A new ad set is guessing until it gets about 50 results in a week, and every edit restarts it. Fewer ad sets, more budget each, no edits for 7 days.
- A message you must test gets its own ad set and budget. A shared budget goes to whatever moves first, not to what matters.
Rules, so the same decision is not argued every week
| Action | Only when |
|---|---|
| Wait | The ad has spent less than the target cost per lead. It has not lost. It has not been tested |
| Cut an ad | It has spent 3 times the target with 0 leads, and other ads in its ad set got leads in the same period. Both |
| Scale an ad | 10 or more leads at under 70% of target over 7 days. Raise the budget 20% at most, every 3 days at most |
| Stop a campaign | Over 3 times target for 7 days, and the cause is the objective, the page or the audience. New creative does not fix a broken funnel |
What wins. The ad that answers the buyer's real question. On an expensive purchase that is rarely the specification. It is what the whole thing costs, and who handles the parts they do not understand. So show the complete package under one headline, with real footage of it working. When the ad and the page disagree, fix the page.
The mistake. Replacing an ad because you are bored of it. You tire of it long before the market does. Refresh when each person sees it more than about 3 times a week.
Read the numbers in order
Top down. Stop at the first broken step. Fixing anything below it is wasted effort.
Outreach
| Metric | Healthy | Act if |
|---|---|---|
| Bounce rate | under 3% | Over 5%: pause. The list is dirty |
| Open rate | 40 to 60% | Under 25%: an inbox problem, not a copy problem |
| Reply rate | 3 to 8% | Under 2%: a copy problem, not an inbox problem |
| Positive replies | 1 to 3% | The number the calculator runs on |
| Spam complaints | under 0.1% | Over 0.3%: stop everything |
Ads
- Is the event real? Lead events match the leads received, and no ad is rejected.
- Is it converting? Cost per lead against target.
- Where does it break? Page views to leads. Rule out the objective and the audience before you blame the page.
- Is the ad earning attention? A click rate under about 1% says no.
Then each engine feeds the other. The outbound list becomes an ad audience, so the people I email also see the ad. The message that wins in ads, where data comes in days, opens the next email. Questions from calls become the next ads.
What does your number take?
Put in your own target and rates. The starting values are placeholders, not benchmarks. You get what the target takes each month, and the first thing to do about it.
The target
Outbound
Ads
Start this week
- Run the calculator with your own target and order the mailboxes it asks for. They need 10 to 14 days of warm-up.
- Find the market shift. One true, recent change your buyers have not adjusted to. It opens every email and ad.
- Clean the list and pilot 50 companies. Group by domain, find 2 decision makers each, verify every address.
- Check the ad tracking. Submit a test lead and confirm the platform records exactly 1 lead event.
- Write email 1 and the follow up for each of the 3 roles. Once the mailboxes are warm, send 200 and read the numbers in order.
What this is built on
I never stop learning. This playbook is built on the people below, then tested in my own work.
- Alex Hormozi, $100M LeadsHis Core Four: warm outreach, content, cold outreach and paid ads. I run cold outreach and paid ads as one engine, and his Rule of 100 sets the floor: 100 actions a day, for 100 days, before judging anything.
- Jeb Blount, Fanatical ProspectingPipeline is a daily habit, not a campaign. His 30 day rule: what I prospect this month pays over the next 90 days. When my calendar looks full, I keep prospecting.
- Josh BraunOutbound that opens with the buyer’s problem, not my product, and makes it easy to say no. A message about them gets read. A message about me gets deleted.
Want this engine running in your business? Message me on LinkedIn.
I build and run it end to end: the list, the copy, the ads and the follow up.
Ali Hamou · London, Ontario · Methods drawn from campaigns I run. Not legal or compliance advice for your market.