Earned media ROI: measure outcomes without ad-value proxies

Earned media ROI compares the contribution assigned to earned coverage with the full cost of earning and measuring it. Match referral and sales evidence, state how much credit earned media receives, and deduct program spending before calculating the return. Keep attributed results separate from claims about additional profit caused by the coverage.

For a communications director and finance partner, the difficult work is the evidence connecting an editorial mention to a business outcome. The marketing ROI versus ROAS guide owns the general contribution formula. Here the task is deciding which earned-media outcomes can enter it and how to show the uncertainty.

Replace advertising-value proxies with outcome evidence

Advertising-value equivalency estimates what editorial space or airtime might have cost as advertising. It is neither revenue nor contribution. AMEC's Barcelona Principles 4.0 reject advertising-value equivalents and direct measurement toward communication outcomes and impact.

Separate coverage outputs from business outcomes. A published article is an output. A qualified inquiry is a response. A completed, matched sale can support a financial contribution calculation. Awareness and reputation may be valid objectives, but do not convert them into dollars without an explicitly supported economic method.

Agree the outcome before looking at results. For a sales objective, choose invoiced revenue net of returns rather than mixing inquiries, open pipeline and completed sales. For an awareness objective, use the appropriate audience evidence and say that a financial return has not been established.

Build an evidence register from existing exports

Use a coverage register with publication, URL, publication date, campaign and any referral identifiers. Match it to website referral records, CRM contacts and opportunities, then to the finance outcome using approved identifiers. Retain the evidence category for each match: observed referral, customer-reported source or another documented basis.

Keep one outcome identifier per sale. Multiple mentions of the same deal are evidence records, not additional revenue rows. Choose the credit policy across all channels before allocating dollars; the marketing attribution guide explains those allocation choices.

Google's attribution documentation defines attribution as assigning credit across interactions. That is the limited use of attribution in this example: an approved allocation of observed results, not a controlled estimate of earned-media lift. Covirage is not affiliated with Google.

Also export agency invoices, internal labor allocations, production, monitoring and research costs. Earned placement does not mean the program costs nothing. Record shared-cost allocations and separate paid placements from earned coverage, while including any supporting costs within the agreed program boundary.

Calculate attributed return for a US manufacturer

A fictional US equipment manufacturer reviews a completed communications program. Its observation window runs from publication through the agreed sales follow-up cutoff. The following three closed sales have finance-matched revenue and delivery costs. The earned-credit fractions are illustrative policy inputs approved for this review, not percentages inferred from the revenue.

Enter inputs in columns A through D. Columns E through G are calculated:

Use these formulas in row 2 and fill down through row 4:

E2 = B2*D2
F2 = (B2-C2)*D2
G2 = (B2-C2)*(1-D2)

The matched program cost register is:

Attributed earned media ROI = (assigned contribution − program cost) / program cost

With program cost entered in J2, the return formula is:

=IF(J2=0,NA(),(SUM(F2:F4)-J2)/J2)

Assigned contribution is $49,500. Deducting $20,000 leaves $29,500, and $29,500 / $20,000 gives 147.5%. Percentages are displayed to one decimal; calculations use unrounded values. This result applies only to the matched sales, cost boundary and credit policy described above.

Reconcile the allocation and test the policy

Full contribution is $130,000 − $57,000 = $73,000. Earned-media contribution of $49,500 plus other-channel contribution of $23,500 must equal that $73,000. In the sheet, this check must return zero:

=SUM(F2:F4)+SUM(G2:G4)-(SUM(B2:B4)-SUM(C2:C4))

Then ask how dependent the return is on attribution judgment. Change only the earned-credit fractions to 25%, 50% and 100%. Assigned contribution becomes $9,000 + $11,000 + $15,000 = $35,000, and attributed ROI becomes ($35,000 − $20,000) / $20,000 = 75.0%.

Both policies produce a positive attributed return. Neither establishes incremental profit. A suitable experiment or defensible comparison may support that separate conclusion; the association and cause guide explains why observational matches alone do not.

Where it goes wrong

• Audience reach becomes revenue. Potential readership does not show who read the coverage, acted or bought.
• Every touch claims the full deal. Sum the allocations across channels for each outcome before calculating channel returns.
• Unmatched sales receive invented credit. Keep them visible as unmatched. Missing tracking is uncertainty, not evidence of either zero impact or full credit.
• Short follow-up hides later sales. Compare programs at equally mature cutoffs and retain open pipeline separately from realized outcomes.

The communications owner should confirm coverage evidence, sales operations should approve the matching and finance should sign off on revenue and costs. Together they can decide whether the result supports renewing the program or first improving measurement.

Evaluate your own earned-media records in Covirage

Bring the coverage, referral, CRM and cost files with your approved matching and credit rules. Covirage maps the columns, and its tools compute and check the supplied figures; the external AI model explains the result with every figure traced to its rows. Create a free Covirage account and bring a sample of your own data.

See customer profitability for the contribution behind matched sales.