The first 30 days with a new client: a week-by-week script

The first 30 days define the whole relationship. Not through performance — month 1 is rarely the best — but because this is when it gets established what will count as success and which data the conversation will happen on.

The temptation is to start advertising on day 1 to show activity. It is the most expensive onboarding mistake: you will spend money producing a number nobody will be able to interpret afterwards.

Week 1 — understand and verify

What to ask for on day one

Admin access to the ad account, the pixel and the page. Read access to the CRM. And a direct contact for whoever handles the leads.

That last one is the most forgotten and the most important: half of month 2's performance problems sit in lead handling, and you need a relationship with that person before you need them.

The questions that change everything

What is the margin per sale? Without it, you do not know what CPA is acceptable and you will optimize blind.

What is the real target? "More sales" is not a target. "30 sales a month at up to $800" is.

Who handles the lead and how fast? The honest answer is usually worse than the official one.

What has already been tried? It prevents repeating last year's failed test.

What is the seasonality? December and January are not alike in almost any sector.

The measurement check

Before any ad, walk the whole path as a lead: click, navigate, submit, and check that it reached the CRM with the source attached.

If it did not, fixing that is week 1's job. Advertising with broken tracking is buying a month of useless data.

Week 2 — structure and first launch

With measurement verified, build:

• Standardized naming from the very first campaign, before there is history to break
• UTMs derived from the name, with matching fields in the CRM
• Base audiences: remarketing, lookalike from the customer list, exclusions
• Two or three campaigns, not eight — a small structure learns faster

And launch the first creatives. Week 2's goal is not performance, it is generating data.

Tell the client, in these words: this week the number will swing, it is learning, and the first valid reading is in week 4.

Week 3 — observe and correct

The temptation to tinker daily is strong and is the classic mistake. A campaign touched every day never exits learning.

What to do this week:

• Verify leads are still arriving with their source
• Check quality with whoever handles them, not just CPL in the dashboard
• Pause anything that spent two or three target costs with zero conversions
• Produce the next creatives

What not to do: change budgets, switch the optimization event, restructure campaigns. None of that before the first reading.

The conversation with whoever handles the leads is the most valuable part of this week. It answers, with no dashboard at all, whether the audience is right — and three days of conversation save three weeks of analysis.

Week 4 — first reading and agreement

Now there is enough data for a real conversation.

MONTH 1 · ACME

WHAT RAN
Spend           $18,000
Leads           94
Sales           7
CPL             $191
Cost per sale   $2,571

WHAT WE LEARNED
· Testimonial creative: 30% higher CPL, 3x better close rate
· The 35-50 audience converted better than 25-34
· Median response time 3h12 — the under-1h band closed 2.4x more

WHAT CHANGES IN MONTH 2
· Budget concentrated on the testimonial angle
· 25-34 audience reduced
· Proposal: lead response within 30 minutes (with sales)

MONTH 2 FORECAST
Leads 110 to 140 · Sales 11 to 16 · Cost per sale $1,300 to $1,800
Assumptions: close rate held at 7.4%, response under 1h

The "what we learned" section is what justifies the whole of month 1. Even with modest results, the client leaves with three things they did not know about their own business.

And the forecast with a range turns month 2 from a vague expectation into an explicit agreement.

The four classic mistakes

Advertising before verifying measurement. Produces a month of useless data and an impossible conversation.

Promising numbers in month 1. The first month is learning. Promising results in it is betting against yourself.

Building too large a structure. Eight campaigns on $18,000 splits the budget into pieces too small for any of them to learn.

Not talking to whoever handles the leads. You will discover in month 3 that leads had been waiting four hours since day one.

What the client should have on day 30

A clear reading of month 1, three learnings about their own business, a forecast with a range for month 2, and the certainty that the number they see is the one you see.

If that is in place, month 2 starts with the right conversation — and that is the difference between a client who renews in month six and one who was lost in week 2 without anyone noticing.

For the audit script when the account comes from another agency, see auditing an inherited ad account. For week 4's forecast, see forecasting results for clients.