The B2B SaaS marketing audit checklist, pipeline first

Most marketing audits start with channels: is the SEO healthy, are the ads optimized, is the email cadence right. That's backwards. A perfectly tuned channel feeding the wrong definition of pipeline still misses the number.

This checklist starts where the board starts, with pipeline, and works outward. It's built for a CEO or CRO to run without an agency. Score each item 0 (no), 1 (partly) or 2 (yes, and you can show it). Most sections take an hour or two.

1. Pipeline and measurement

The foundation. If this section scores low, fix it before you trust anything below it.

• There's a written definition of qualified pipeline, agreed by sales and marketing, with a date. (Template here.)
• The CRM is the single source of truth for pipeline and revenue.
• Marketing-sourced and marketing-influenced pipeline are reported separately, with written rules for edge cases.
• Pipeline by source can be shown for at least the last four quarters.
• Stage-to-stage conversion rates (lead → meeting → opportunity → won) are tracked by quarter.
• The demo form asks "How did you hear about us?" and the answers are reviewed monthly.

Red flag: marketing's pipeline number in the board deck can't be traced to opportunities in the CRM.

2. ICP and positioning

• The ICP is based on closed-won data (win rate, cycle length, deal size, retention), not opinion.
• There's a written list of disqualifiers: segments that look right but stall or churn.
• The team can name the alternative customers most often choose instead of you (often "do nothing" or a spreadsheet, not a competitor).
• Positioning states the unique value against that alternative, with proof.
• A stranger can say what you do and who it's for after five seconds on the homepage.
• Sales and marketing describe the product the same way.

Red flag: the ICP covers most of the market. If everyone is a fit, the targeting isn't doing any work.

3. Channel economics

• Every program is labeled as demand capture (search, review sites, retargeting) or demand creation (social, events, content, podcasts).
• Capture channels are judged on cost per opportunity; creation channels on reach into the ICP and what new pipeline reports in "how did you hear about us."
• Fully loaded CAC (people, tools, agencies, media) is known, and CAC payback is known by channel where the data allows.
• Ad platforms receive opportunity-stage conversions, not just form fills.
• No single channel or person produces most of the pipeline, or there's a written plan to reduce that dependence.
• Budget moves monthly based on pipeline per dollar, and the moves are written down.

Red flag: the team can't name the channel they'd cut first.

4. Website and conversion

• The homepage headline names the buyer's problem or outcome, not the category.
• Every call to action has proof within one scroll (a number, a logo, a named customer).
• There's a transitional offer for buyers who aren't ready to talk to sales.
• The demo form asks only what sales needs to route and qualify.
• Demo requests are followed through to meetings held and opportunities created, not just counted.
• Paid landing pages match the ad that sent the visitor (same promise, same words).

Red flag: form conversion went up and meetings held went down.

5. Search and AI-search visibility

• Search Console is connected; indexing errors are reviewed monthly.
• Pages render in plain HTML without JavaScript, so search and AI crawlers can read them.
• robots.txt deliberately allows (or deliberately blocks) AI search crawlers like OAI-SearchBot and PerplexityBot.
• The company is described the same way on the site, LinkedIn, review sites and directories.
• There are pages for the questions buyers ask before they buy: comparisons, alternatives, pricing, use cases.
• A fixed set of buyer prompts is run monthly in ChatGPT, Perplexity and Gemini, and you know whether you're named. (How to get recommended by AI.)

Red flag: you've never checked what AI assistants say when buyers ask for options in your category.

6. AI workflows and output quality

• You know which marketing tasks the team runs through AI, and how much of the output gets rewritten.
• There's one source of truth for ICP, positioning, voice and proof that AI tools can read.
• Standards for good work (messaging rules, banned words, proof requirements) are written down.
• AI-assisted work passes a review before it ships.
• The workflows are documented well enough that a new hire could run them.

Red flag: the team has AI tools, and nothing ships faster than it did before. That's the gap the harness closes.

7. Team and rhythm

• Someone owns the pipeline number, by name.
• There's a weekly metrics review with written takeaways.
• Decisions (what was cut, what was doubled, why) are logged.
• Channel playbooks are written down.

Red flag: the marketing system lives in one or two people's heads.

How to read your score

There are 39 items, worth up to 78 points.

Score	What it usually means
60–78	The system works. Look for the two or three channels with the best pipeline per dollar and scale them.
40–59	Typical for a growing B2B SaaS company. One or two sections are dragging the rest down; fix those first.
Under 40	The numbers can't be trusted yet. Start with section 1 before spending more.

Don't average across sections. A low score in pipeline and measurement makes every other score less reliable, so fix it first even if another section scores lower.

When to bring in help

Run this yourself first. If the low scores cluster in pipeline definition, ICP or channel economics, an outside operator can usually find the leaks faster, because they're not defending last year's plan.

That's what the Fear Audit is: this checklist and a lot more, run against your real CRM, analytics and ad data in two to three weeks, with a ranked fix list, the first AI workflows built, and a 30-day plan. Fixed fee.

Frequently asked questions

What should a B2B SaaS marketing audit include?

Start with pipeline definitions and measurement, then ICP and positioning, channel economics, website conversion, search and AI-search visibility, AI workflow quality and team rhythm. Auditing channels before pipeline definitions leads to optimizing the wrong things.

How long does a marketing audit take?

A CEO can score this checklist in a day. A full audit with CRM, analytics and ad-account data, interviews and fixes typically takes two to three weeks.

When should a SaaS company run a marketing audit?

Before raising the marketing budget, hiring a marketing leader or agency, rebuilding the website, or preparing for a raise or an exit. Also whenever spend is rising and qualified pipeline is flat.

Should the marketing team audit itself?

It's a good start, and this checklist is built for that. The limit is that teams rarely score their own definitions and channel choices harshly. An outside review is most useful for sections 1 to 3.