How to decide what campaign to launch next

Each campaign should improve the next one

After running over 1,326 cold email campaigns, there's one simple system I use to decide what campaign to launch next.

It comes down to using what you've already learned to find better prospects and write messages they have a reason to respond to.

Take a campaign you're running now. Pull up the replies and the companies behind them, then grab your sales-call notes. We're going to use those to build the next one.

Because Each campaign should improve who you target and what you say in the next one.

1. Pull out the interested leads and the objections

Open your sending tool and separate interested replies, replies with objections, and people who haven't answered.

Keep the message each person received next to their reply. You want to know what they were responding to when they said they were interested, or told you why it wasn't for them.

Then bring in your sales-call transcripts and ask whoever took the calls what they noticed.

Some of what you understood won't be in the transcript. Maybe a concern kept coming up, or you had to explain the offer differently before it made sense. Write that down too.

2. Research the companies that showed interest

Use those companies as the starting point for your next list.

Go through their sites and look at what they sell, who they sell to, and how their team is set up. You're looking for things you didn't think to target when you built the campaign.

For instance, what if you find out that 80% of your booked calls came from companies without sales leadership?

You weren't looking for that before, but now you've got a reason to go deeper. Do they have reps but nobody leading them? Is the founder still managing sales?

Check whether that's also common in the rest of your list. If it's much more common among the companies booking calls, try building a segment around it.

Now you can find more companies in that situation and write a message about the problem that comes with it.

You can also look for changes over time. Older pages in the Wayback Machine, when available, might show that a company used to list a sales leader. Check whether they've left or changed roles before treating that as a signal.

3. Build an angle for each buying situation

Audiences really segment themselves based on your offer.

My offer is very different for a company with a large sales team versus a company without one. So let's use that as the example.

For a company with reps, you could test helping them find better accounts to approach. For a founder without a sales team, you could test finding demand before they hire.

Now pull together the information that will help you write to each group:

• Internal information - what you've learned from customers, calls and previous campaigns
• Industry changes - where the old way is becoming a problem and your offer gives them another option
• Their own language - how they describe the pain on Reddit, Trustpilot and other review sites

Look for the complaint your offer can actually solve. Keep their wording, then connect it to the situation that makes your offer useful.

The founder doing their own selling and the manager trying to keep reps busy may describe very different problems. Write each of them their own insider message.

4. Answer recurring objections earlier

Go back to the replies and call notes. Find objections that keep appearing and that you can honestly address.

Say someone replies, "We already have a sales team."

If your offer helps existing reps, try a line like, "We find accounts for your reps to approach," in the next sequence.

That's a simple pre-objection handle. You're clearing up what you do before another person has to ask.

Keep the unanswered leads separate. They haven't told you why they didn't respond, but the research and objections from other replies can give you a better angle to test with them.

5. Turn those findings into the next campaign

You now have somewhere to go with both groups.

Find similar companies to the interested leads and test the message that earned that interest. If you've found a new buying situation, give that segment its own message.

For the remaining leads you can still contact, build a new campaign around another relevant angle. Respect opt-outs and keep bounced or suppressed contacts out of it.

Changing the copy doesn't fix a sending problem, so keep checking delivery health before you launch another round.

Write down who you're testing, what you're saying, and why you think it will matter to them. When the replies come in, keep them with that experiment so you can use them for the next one.

If you want to hand this process to ChatGPT or Claude Code, paste this whole email in and add the prompt below.

Use this newsletter to help me design my next campaign experiment. First, identify my sequencer and the available way to read its campaigns and replies, such as an API, CLI, MCP or export. Ask me for missing access. Gather my offer, internal campaign information and sales-call transcripts, then ask what I learned from the conversations that the transcripts missed.

Separate interested replies, objection replies and unanswered outreach. Research the interested companies through their websites and, where useful and available, historical Wayback Machine pages. Compare patterns with the rest of the audience. Suggest similar companies to find and new signals or segments to test, without treating engagement as proof of a sale.

Look for relevant industry changes and direct pain-point language from Reddit or review sites such as Trustpilot. Use those inputs to propose a specific message for each segment based on how my offer helps them. Identify recurring objections we can honestly address earlier in the sequence. Propose a new angle for eligible remaining leads, excluding unsubscribed or suppressed contacts.

Show the evidence behind each experiment, what is still an assumption, and what response would help us decide the next step. Ask for my approval before changing campaigns, adding contacts or sending anything.

- Mitch

P.S. Something I've been thinking about: games taught me a lot about getting into flow. The challenge gets harder as you get better, and there's a reward for progressing. I stopped playing, but that balance still interests me.