Content Marketing ROI: How to Calculate It and Make It Credible

Content marketing ROI is (value attributed to content − content cost) ÷ content cost × 100. The formula is easy. The hard part is deciding what counts as cost, what counts as return, and how much of a deal your content can honestly claim. A number that survives a finance review needs a stated cost base, a stated attribution rule and a stated time window. A number without those is a guess with a percent sign.

This guide gives you the formula, a fully illustrative worked example, a way to grade your return by evidence, and a short diagnostic to run before you quote any ROI figure to leadership. It is written for B2B and ecommerce teams worldwide, so money examples are labeled with their currency and nothing here assumes one country’s rules.

The standard version, as Ahrefs states it, is ((return from content − cost of content) ÷ cost of content) × 100. Its own example: $10,000 in sales from content that cost $2,000 to create gives 400%. Ampere adds the reading guide: 100% means the content returned twice what it cost, 0% means break-even, and a negative figure means cost exceeded return in the period you measured.

Two mistakes show up constantly. The first is reporting a benefit-to-cost ratio as ROI. A ratio of 5:1 is a 400% ROI, not 500%, and Mindstamp flags this confusion as one of the common errors. The second is treating correlation as cause: traffic rose, revenue rose, so content did it. The same source notes that causal claims need comparison and timing evidence.

You can run the formula at three scopes, as G2 describes: the whole content program, a single campaign or a single piece. Choose the scope before you collect numbers. Program-level ROI is the most defensible with finance. Piece-level ROI is the most useful for editorial decisions, and also the noisiest.

Northbeam draws a distinction worth keeping: reporting tells you what happened, while measurement helps explain why and what to do next. It also separates three layers: activity metrics (how much you published), performance metrics (time on page, scroll depth, return visits) and business outcomes (pipeline influence, conversion lift, retention). ROI lives only in the third layer. A dashboard full of sessions and shares is reporting. It becomes ROI only when it connects to money.

Our analysis: most of the “content ROI” slides that fail in budget meetings fail because they stay in layers one and two. Pageviews are a fine diagnostic and a poor justification.

The numbers below are invented to show the mechanics. They are not a benchmark and not a client result. The currency is euros, chosen arbitrarily; swap in your own.

Imagine a B2B software company that spends twelve months on a content program.

Now the return side. Suppose the CRM shows 14 closed-won deals where content was the first or last recorded touch, with an average first-year gross profit of €13,000 each.

Now add a second tier. Say 20 further deals had a buying-group member read tracked content inside a defined window, worth €260,000 in gross profit combined. If your team agrees in advance to count 10% of influenced value, that adds €26,000. Then:

Notice what changed. Nothing about the content moved. The ROI shifted 32 points purely because of a counting rule. That is why the rule must be written down before the results come in, and why the report should show both figures.

If the same deals were counted at revenue and the average deal carried a 50% gross margin (again, illustrative), the attributed revenue would be €364,000 and the ROI would show 355%. The program did not get better. The measurement just stopped subtracting the cost of delivering the product. Mindstamp argues for gross profit where possible for exactly this reason: revenue overstates the return.

The most common way to inflate ROI is to leave out cost nobody invoices. Ampere recommends including loaded internal people costs, subject-matter expert time, freelancers, production, tools, distribution and maintenance. Sitecore likewise says investment can include the overhead to create, distribute and reuse content assets.

A checklist to apply before calculating:

On the return side, decide whether you count direct revenue only or also customer lifetime value. Shopify notes that content can create value for existing customers too, which you can express through lifetime value. That is legitimate, but it must be a declared choice, not an afterthought added when the first number looked weak.

When conversion data is thin, Ahrefs describes an approximation: monthly traffic value multiplied by the content’s lifetime in months, based on what you would have paid for the same visits through ads. We treat this as a floor-of-effort estimate for ranking content, not as revenue. It values visits, and a visit is not a lead. Use it to compare articles against each other, not to claim return to a CFO.

Most visitors do not buy from the article they land on. Ahrefs notes that return typically depends on conversion tracking, such as form submissions or trial sign-ups. Google’s own documentation describes attribution as assigning credit for key events to the clicks and factors along a user’s path, and its reports offer data-driven, paid and organic last click, and Google paid channels last click models. Note what that implies: the default tooling is built around ad and click paths, and a buyer who read three of your guides, then asked a colleague, then typed your brand name into a search box leaves a thin trail.

The B2B problem has three parts.

Long cycles. ZoomInfo’s guide says B2B cycles run weeks to months, that standard 30-day windows miss early touches, and that a practical test is to check how many closed-won accounts had their first touch more than 90 days before opportunity creation. Heeet makes the same point bluntly: a 90-day window is no help when the average cycle is 365 days.

Buying groups. Several people from one account read different things at different times. Credit assigned to a single anonymous session misses this, so account-level views matter.

Two systems. Engagement data sits in analytics and revenue sits in the CRM. Cometly describes the needed work as connecting anonymous sessions to named lead records, mapping touchpoints across sessions and channels, and linking them to pipeline stages and revenue events. Marketing MBA argues the CRM should be the system of record for revenue because it holds deal amounts, close dates and stages.

All of these are vendor-published methods, not independent studies, and they have an interest in selling attribution tooling. The logic is sound, and the specific thresholds (30, 90, 365 days) are illustrations of why you should look at your own data rather than adopt someone’s default.

The most useful idea we found is Ampere’s evidence tiering. Rather than one blended figure, report three lines:

This is a method from one vendor, not a standard, but we think it is the right instinct. It keeps the defensible number defensible and still gives credit to effects you can see but cannot prove.

A related discipline: keep AI mentions, citations, clicks, visits, enquiries and revenue as separate rows. An AI answer naming your brand is a mention. A link inside that answer is a citation. A person following it is a click, and only an enquiry tied to a deal is revenue. Merging them produces a story that sounds good and measures nothing.

ROI is the end of a chain. The earlier links tell you where the chain is weak. Northbeam suggests tracking awareness, engagement, consideration, conversion influence and retention or expansion. Search Engine Watch proposes a similar top, middle and bottom view for B2B. Combined into one working table (our synthesis):

CMI’s 2025 B2B benchmark report shows why the stage matters. Among 980 surveyed B2B marketers, 87% said content helped create brand awareness, 74% said it helped generate demand or leads, and 49% said it helped generate sales or revenue. In plain terms, content is most often credited at the top of the funnel and least often at revenue. If your ROI formula looks only at revenue, you are measuring the stage where content gets the least direct credit.

Search this and you will find confident numbers. One roundup claims an average return of $7.65 for every $1 spent and a median SEO ROI of 748% over three years. Others publish industry ranges, such as 400 to 600% over three years for SaaS. We do not use these as targets, for four reasons.

In the same vein, the claim that content generates three times the leads of outbound at 62% lower cost circulates widely, and we could not trace it to a primary study. Treat it as marketing copy until you can.

Our analysis: a “good” ROI is one that is positive on your own cost base, improving over successive quarters and above what the same budget would earn in your next-best channel. That third comparison is the one a CFO actually cares about.

A few habits we think deserve gentle mockery.

Work through these in order. If you cannot answer one, fix it before presenting an ROI.

A small caveat on incrementality: it needs enough volume to produce a meaningful comparison. For a team closing a handful of enterprise deals a year, a clean control group may simply not exist. In that case, lean on the tiered report and on trend over time rather than forcing a test the data cannot support.

If your team can run the diagnostic above in-house, do. If the blockers are structural, such as content produced without a strategy, analytics disconnected from the CRM or no one owning maintenance, a managed engagement can help. Our SEO/GEO measurement and performance tracking work sits alongside strategy, production and optimization, and scope, deliverables and budget are defined in a proposal after initial scoping. Pricing is available on request. You can describe your situation through our contact page.