By day 90, a fractional CMO should have fixed your measurement, shipped quick wins you can see live, put a written roadmap and a weekly decision cadence in place, and made at least one call to scale or cut spend. Discovery should take about two weeks, not two months. Here is the week-by-week plan I work to, the deliverables to check at days 30, 60 and 90, and a template to copy.
Why the first 90 days decide the engagement
Many fractional engagements run month-to-month or have a short minimum term. That makes the first quarter the real trial period, for both sides.
It also sets habits. A weekly meeting that starts as a status update stays one. A 40-page strategy document as the first deliverable sets the pattern for the next. The first 90 days show whether you hired an operator or an advisor.
Before day one, agree two things in writing:
- One primary metric. Revenue, qualified pipeline or contribution margin. Pick one; everything else is a supporting indicator.
- A baseline period. What you will compare against at day 90, usually the previous quarter, or the same quarter last year for seasonal businesses.
Without both, day 90 turns into a debate about what “better” means.
Days 1-14 (weeks 1-2): access, audit and baseline
Week 1: access and interviews
Nothing ships without access, and chasing logins is the most common way I see a first month slip. Request everything on day one:
- Ad accounts (Google, Meta, LinkedIn, TikTok) with admin rights
- GA4, Google Tag Manager and Search Console
- CRM (HubSpot, Salesforce or similar) and the store backend if you sell online
- Email and lifecycle tools such as Klaviyo
- Website CMS, plus a list of every marketing tool you pay for
- Finance data: monthly revenue and marketing spend by channel for the last 12 months
In parallel, run short interviews: founders, the sales lead, any agencies, and three to five recent customers or closed-won deal reviews. The goal is to learn why people buy and where deals stall, not to collect opinions about the logo.
Week 2: baseline audit
Build a one-page baseline: revenue, spend by channel, blended customer acquisition cost (CAC), conversion rates at each funnel stage, pipeline for B2B, and marketing efficiency ratio (MER) for ecommerce. Note every place where two systems disagree, such as ad platforms reporting more conversions than the CRM shows.
The marketing audit checklist covers what to inspect. The output that matters is a ranked issue list, ordered by impact and effort, with an owner on each item.
Day 14 deliverable: the baseline sheet and the ranked issue list. Not a presentation.
Days 15-30 (weeks 3-4): fix measurement and ship quick wins
Measurement first
You can’t make spend decisions on numbers nobody trusts. In this window, the basics should start working:
- Key events in GA4 and ad platform conversions match what actually happened in the CRM or store backend, within an agreed tolerance
- Browser and server events are deduplicated wherever you use the Meta Conversions API
- For long sales cycles, CRM lead stages feed back to Google Ads and other platforms, so bidding optimizes for qualified leads rather than form fills
- A UTM naming convention is written down and used
- A weekly report exists and leads with the primary metric
Quick wins: shipped, not recommended
A quick win counts when it is live, not when it is on a slide. Typical ones:
- Pausing campaigns, keywords or audiences that have spent for months with no conversions, once you’ve confirmed it isn’t a tracking gap
- Fixing broken forms, slow landing pages or a demo request that routes to nobody
- Turning on missing basics such as welcome and abandoned cart flows
- Canceling tools nobody uses
Each quick win should have a before number and a go-live date. As a made-up example: if the audit finds $3,000 a month going to campaigns with no attributed sales in 90 days, pausing them on day 20 frees that budget for the tests in month two. That’s the kind of arithmetic you should see written down.
Day 30 deliverable: tracking you trust, a working weekly report, and two to four quick wins live.
Days 31-60 (weeks 5-8): growth roadmap and team cadence
The roadmap
Draft the roadmap in week 5, get sign-off in week 6, and have the first bets live by week 7. A good one has one target and three to five bets, not thirty initiatives. For each bet, write down:
| Field | Example |
|---|---|
| Hypothesis | Rebuilding the demo page around the top objection from sales calls will lift demo requests |
| Owner | A named person, not “marketing” |
| Budget | Spend and hours committed |
| Success metric | Demo requests per 1,000 sessions, then qualified pipeline |
| Kill criteria | The threshold that ends the bet, agreed now |
| Review date | When the decision gets made |
Kill criteria are the part most roadmaps skip. Agree them before launch, when nobody is emotionally invested yet.
The cadence
A roadmap without a rhythm becomes a document nobody opens. The minimum operating cadence:
- Weekly growth meeting (45-60 minutes): numbers first, then decisions, then blockers. If a meeting ends with no decision, it was a status update.
- Weekly report: the same format every week, sent before the meeting.
- Monthly review with the CEO: budget reallocation and anything that needs founder sign-off.
- Written briefs for every agency and freelancer, so they work to the roadmap rather than to their own retainer scope.
This is where a fractional CMO earns the title: running the team and the agencies to one plan, not adding a layer of meetings on top of them.
Day 60 deliverable: the written roadmap, at least two bets live, and four or more weeks of consistent weekly reports.
Days 61-90 (weeks 9-13): scale what works, cut what doesn’t
Month three is about decisions: verdict reviews in weeks 9 to 11, budget moves in week 12, the quarter review in week 13. Each bet gets one of three verdicts:
- Scale: it hit its success metric. Increase budget in steps and check that efficiency holds, rather than doubling spend overnight.
- Iterate: leading indicators are positive but the result is not yet clear. Change one variable and set a new review date.
- Cut: it missed its kill criteria after a fair test. Stop it, write down what you learned, and move the budget.
Cutting is the proof of judgment. A quarter where every channel “shows promise” is a quarter where nobody made a call.
Before day 90, the knowledge should also leave the fractional CMO’s head. Channel playbooks, tracking documentation and the reporting setup belong to you, whether the engagement continues or not.
Day 90 deliverable: a quarter review against the baseline, a verdict for each bet, next quarter’s budget allocation, and the next 90-day plan.
Checkpoints at day 30, 60 and 90
Use these as a scorecard. If most of a row is missing, raise it that week.
| Checkpoint | You should see | Metrics to review | Question to ask |
|---|---|---|---|
| Day 30 | Baseline sheet, tracking fixed, weekly report, 2-4 quick wins live | Tracked vs actual conversions, wasted spend removed, funnel conversion rates | “Which numbers do you trust now that you didn’t on day one?” |
| Day 60 | Written roadmap, cadence running, at least two bets live | Leading indicators per bet, CAC or cost per qualified lead trend | “Which bet is most likely to fail, and when do we decide?” |
| Day 90 | Scale and cut decisions made, budget moved, next plan agreed | Primary metric vs baseline, blended CAC or MER, CAC payback | “What did we stop doing, and what did it free up?” |
Don’t expect the primary metric to have moved much by day 30. Do expect measurement and wasted spend to have moved.
Red flags that the engagement is drifting
- Still “discovering” in week 8. After week two, discovery runs alongside execution, not instead of it.
- Nothing you can click. Every deliverable is a document; no page, campaign, flow or dashboard has changed.
- Meetings without decisions. The weekly call reviews activity and ends with “let’s keep an eye on it.”
- More spend before working tracking. Budget increases proposed while conversion data is still unreliable.
- The agreed metric has gone quiet. Reports now lead with impressions, traffic or engagement.
- Nothing has been cut by day 75.
- Agencies left unmanaged. Your requests get forwarded to agencies without a brief or a deadline.
- Unclear time. You don’t know how many hours you’re getting or when.
You can screen for most of these before you sign; these questions to ask a fractional CMO show how.
Copyable 90-day plan template
Paste this into your project tool and assign an owner and date to each line.
Before day 1
- Primary metric agreed: ______
- Baseline period agreed: ______
- Time commitment and weekly availability confirmed
Days 1-14: access and baseline
- Admin access to ad accounts, GA4, GTM, CRM, store backend and email platform
- Interviews: founders, sales, agencies, 3-5 customers
- Baseline sheet: 12 months of revenue, spend by channel, CAC, funnel conversion rates
- Ranked issue list with owners
Days 15-30: measurement and quick wins
- Conversions reconciled against CRM or store data
- UTM convention published and weekly report live
- 2-4 quick wins shipped, each with a before number and go-live date
Days 31-60: roadmap and cadence
- Roadmap with 3-5 bets: hypothesis, owner, budget, metric, kill criteria, review date
- Weekly growth meeting and monthly CEO review scheduled
- Agency and freelancer briefs rewritten against the roadmap
- At least two bets live
Days 61-90: decisions
- Each bet marked scale, iterate or cut, with the reason
- Budget reallocated for next quarter
- Playbooks and tracking documentation handed over
- Quarter review against baseline delivered and next 90-day plan agreed
Get it built
If you’d rather have this plan run inside your company than read about it, that’s how I work. Engagements usually start with a fixed-price Growth Audit covering the access and baseline work, credited if we continue. See pricing or get in touch to talk through where your first 90 days should start.