By Mehdi HassanHow do you measure content marketing ROI?
You measure content marketing ROI by comparing the pipeline and revenue content contributes to with the full cost of producing and maintaining it. The formula is simple: (value attributed to content minus content cost) divided by content cost. The hard part is the value side, because B2B buyers rarely read one post and buy.
That's why traffic is a poor stand-in. TripleDart's Q1 2026 study of 28 B2B SaaS companies found general blog posts drew 12,000 sessions and zero leads, while homepages got 81.7% of AI-attributed leads. Lots of sessions can mean very little revenue. The metrics below measure what content actually does for a buying decision, and they connect to the plan in our B2B content strategy guide.
Which metrics show whether content is working?
Five metrics show whether content is working for a B2B site: assisted conversions, direct conversions from content pages, self-reported attribution, AI citations and referrals, and sales use. Together they cover both the trackable clicks and the research that happens where analytics can't see.
Metric | What it tells you | Where to get it |
|---|---|---|
Assisted conversions | Content viewed before a demo or contact request | GA4 conversion paths, CRM first-touch and multi-touch |
Direct conversions | Content pages that lead straight to the offer | GA4 key events by landing page |
Self-reported attribution | Where buyers say they heard of you | A free-text field on forms |
AI referrals and citations | Visits from AI tools and whether they cite you | GA4 AI channel group, monthly prompt audit |
Sales use | Pages reps send to prospects | A shared link list or tracked links |
Keep the dashboard to these five, reported per cluster. Impressions and clicks from Search Console are still useful as early signals, but they sit underneath these numbers, not beside them. For the wider analytics setup, see our guide to marketing website analytics.
How do you track assisted conversions from content?
Track assisted conversions by setting your demo, contact and signup actions as key events in GA4 and checking which content pages appear in the paths that lead to them. In your CRM, store the first page and the content pages each lead viewed, so you can tie content to deals, not just form fills.
Don't expect a perfect model. Multi-touch attribution looks precise but rests on assumptions about how much credit each visit deserves, and it misses everything that happens off your site. Use it to compare clusters with each other, not to calculate exact revenue per post. Our article on multi-touch attribution for B2B covers when the extra effort is worth it.
How do you measure AI citations and AI referral traffic?
Measure AI visibility in two ways: track referral visits from AI tools in GA4, and run a monthly prompt audit to see whether engines cite or mention you. Referrals show the clicks; the audit shows the influence that doesn't produce a click.
Create an AI channel group in GA4
Add a custom channel group with a regex on source matching chatgpt.com, perplexity.ai, claude.ai, gemini and copilot. Place it above Referral so those visits aren't lumped in with other sites.
Pick 20 to 30 buyer prompts
Choose the questions buyers ask before they shortlist you, such as comparisons, costs and "best option for" prompts.
Run them monthly
Run each prompt in ChatGPT, Perplexity and Google, and record whether your brand is mentioned and which of your pages is cited.
Report the trend per cluster
Show the share of prompts where you appear, and which pages earn the citations. Refresh the pages that should appear and don't.
Expect engines to differ. Semrush's AI Visibility Index, covering 126 million prompts, found ChatGPT cites about 15 sources per answer against about 3 for Gemini. Our guide to B2B attribution when buyers research in ChatGPT goes further on tying this to pipeline.
Why does sales use matter as a content metric?
Sales use matters because it measures whether a page helps a real buying conversation. If reps send a page to prospects every week, it's answering a question that blocks deals, whatever its traffic says.
Keep a shared list of pages by the question each one answers and ask reps to use tracked links in follow-up emails. Each month, note which pages were sent, which prompted replies and which questions still had no page. Pages that sales never uses and that bring no conversions are candidates for a content refresh or removal.
Hypothetical example: a quarterly content ROI report
For example, suppose a SaaS team reports on its "migration" cluster for the quarter. The content cost is the writer's time, the designer's time and the expert interviews. On the value side: the cluster appears in the conversion paths of a set of demo requests in GA4 and the CRM; several form answers name the migration guide; the brand is cited in a growing share of the 25 tracked prompts; and reps sent the comparison page to prospects most weeks.
The report doesn't claim exact revenue per post. It shows the cluster is influencing pipeline, names the two pages that aren't pulling their weight, and puts both in next quarter's refresh slots.
How long before content shows a return?
Content usually takes months, not weeks, to show a return, and it compounds when pages keep earning visits and citations after they're published. Judge it per cluster over at least two quarters rather than per post over a few weeks.
Decision-stage pages usually show results first, because they meet buyers who are already comparing vendors. Top-of-funnel pieces take longer and pay back mostly through mentions, links and brand recall. If a cluster shows no movement in impressions, citations, assisted conversions or sales use after two quarters, revisit the questions it targets before writing more of it.
