Digital Marketing Statistics 2026
By Axis Intelligence Research
Co-author: Elena Rodriguez | Last updated: August 11, 2026 | License: CC BY 4.0
Five publicly reporting platforms — Alphabet, Meta, Amazon, Pinterest and Reddit — booked $162.74 billion in advertising revenue in the quarter ended June 30, 2026, up 20.5% year over year. Of the $27.68 billion those platforms added versus the same quarter a year earlier, Google, Meta and Amazon took 98.27%.
Quick Answer
Digital advertising is growing fast and consolidating faster. Axis Intelligence Research finds that across the five largest platforms that disclose an advertising line, spend reached $162.74 billion in Q2 2026 (+20.5%), with a Digital Ad Growth Concentration Index (DGCI) of 98.27% — nearly every incremental dollar landed at Google, Meta or Amazon. Meta alone absorbed 46.24% of the increase.
Key Findings
- According to Axis Intelligence Research, the five largest publicly reporting digital advertising platforms generated $162.74 billion in the quarter ended June 30, 2026, up 20.5% from $135.06 billion a year earlier.
- Axis Intelligence Research finds the Digital Ad Growth Concentration Index (DGCI) reached 98.27% in Q2 2026, meaning Google, Meta and Amazon captured that share of all incremental advertising dollars across the panel.
- Meta’s advertising revenue rose 27% to $59.36 billion in Q2 2026 on ad impressions up 14% and average price per ad up 12%, per Meta’s July 29, 2026 results release.
- Axis Intelligence Research calculates Meta’s Monetization Intensity Delta (MID) at 23.78% for Q2 2026 — the share of its growth attributable to revenue per active person rather than audience growth of 3%.
- Amazon advertising services reached $19.81 billion in Q2 2026, up 26% from $15.69 billion, per Amazon’s Form 8-K exhibit filed with the SEC on July 30, 2026.
How Big Is Digital Advertising in 2026?
The honest answer is that nobody can size the whole market from primary sources, because most of it is private. What can be measured precisely is the part that files with regulators, and that part is large enough to set the price of everything else.
Across the five platforms in our panel, Q2 2026 advertising revenue reached $162.74 billion. That is a quarterly figure, not annual — it works out to roughly $651 billion on a simple annualized basis, though seasonality makes Q4 materially heavier than Q2, so treat annualization as an order of magnitude rather than a forecast.
| Platform | Q2 2026 ad revenue ($M) | Q2 2025 ad revenue ($M) | Change | Source |
|---|---|---|---|---|
| Alphabet (Search, YouTube, Network) | 81,629 | 71,340 | +14.42% | Alphabet Q2 2026 earnings slides |
| Meta Platforms | 59,363 | 46,563 | +27.49% | Meta Q2 2026 results release |
| Amazon advertising services | 19,809 | 15,694 | +26.22% | SEC EDGAR Form 8-K Ex. 99.1 |
| Pinterest (revenue, advertising-dominant) | 1,179.7 | 998.2 | +18.17% | Pinterest Q2 2026 results release |
| 762.0 | 464.6 (implied) | +64% | Reddit Q2 2026 results release | |
| Panel total | 162,742.7 | 135,059.9 | +20.50% | Axis Intelligence Research |
Alphabet’s own breakdown, published in its Q2 2026 earnings slides, splits three ways and the three tell different stories. Search and other advertising rose 16.76% to $63.27 billion. YouTube ads rose 12.85% to $11.06 billion. Google Network — the part Alphabet sells on other people’s inventory — fell 0.69% to $7.30 billion.
Elena Rodriguez: That Network line is the number a media buyer should sit with. Every quarter the story is that programmatic open-web inventory is being squeezed, and every quarter you can watch it happen inside a single issuer’s own segment disclosure. Owned-and-operated inventory grew double digits; the reselling business went backwards. If your plan still routes a meaningful share of budget through open exchanges on the assumption that supply and demand there are growing together, Alphabet’s own filing disagrees with you.
Advertising as a share of each seller’s business
Three companies, three completely different exposures to the ad market — which is why they behave differently when demand softens.
| Company | Q2 2026 advertising revenue ($M) | Total Q2 2026 revenue ($M) | Advertising share |
|---|---|---|---|
| Meta Platforms | 59,363 | 60,801 | 97.64% |
| Alphabet | 81,629 | 119,796 | 68.14% |
| Amazon | 19,809 | 200,606 | 9.87% |
Meta has essentially one revenue stream. Amazon’s advertising business is under a tenth of its top line, which is exactly why it can price aggressively for share — a point worth remembering when comparing rate cards.
Which Platforms Are Actually Capturing New Ad Dollars?
This is the question the growth-rate tables never answer, and it is the reason Axis Intelligence Research built the DGCI.
Reddit grew advertising revenue 64% in Q2 2026 — 3.1× the panel average. Pinterest grew 18%. Read a standard roundup and you would conclude that budget is flowing to the challengers. It is not. Reddit’s 64% growth represents $297 million of new revenue. Meta’s 27% growth represents $12.8 billion. In the market for incremental dollars, Reddit’s blowout quarter is worth 1.07%.
The Digital Ad Growth Concentration Index (DGCI)
DGCI — the Digital Ad Growth Concentration Index — measures the share of the total year-over-year dollar increase in advertising revenue, across a defined panel of publicly reporting sellers, that is captured by the three largest sellers.
Formula, in full:
DGCI = ( ΔAlphabet + ΔMeta + ΔAmazon ) ÷ ( Σ Δ all panel members ) × 100
where Δ = (advertising revenue, current quarter) − (advertising revenue, same quarter prior year)
Q2 2026 computation, with every input shown:
ΔAlphabet = 81,629.0 − 71,340.0 = +10,289.0
ΔMeta = 59,363.0 − 46,563.0 = +12,800.0
ΔAmazon = 19,809.0 − 15,694.0 = +4,115.0
ΔPinterest = 1,179.7 − 998.2 = +181.4
ΔReddit = 762.0 − 464.6 = +297.4
----------
Panel Δ +27,682.8 ($ millions)
DGCI = (10,289.0 + 12,800.0 + 4,115.0) ÷ 27,682.8 × 100 = 98.27%
Share of the incremental $27.68 billion, by platform:
| Platform | Incremental revenue ($M) | Share of all new dollars | Own growth rate | Source |
|---|---|---|---|---|
| Meta Platforms | +12,800.0 | 46.24% | +27.49% | Meta Q2 2026 release |
| Alphabet | +10,289.0 | 37.17% | +14.42% | Alphabet Q2 2026 slides |
| Amazon | +4,115.0 | 14.86% | +26.22% | SEC EDGAR |
| +297.4 | 1.07% | +64% | Reddit Q2 2026 release | |
| +181.4 | 0.66% | +18.17% | Pinterest Q2 2026 release |
What the DGCI is built on. Every input is a fact-table row traced to an issuer document — Alphabet’s earnings slides, Meta’s results release, Amazon’s SEC exhibit, Pinterest’s and Reddit’s results releases. All five are in the downloadable dataset with URLs and retrieval dates.
Two platforms we deliberately left out, and why. Snap reported Q2 2026 revenue of $1.599 billion against $1.345 billion, but its press release does not break out an advertising line, and mixing a total-revenue figure into an advertising-revenue panel would produce a number that means nothing. Microsoft discloses search and news advertising growth as a percentage without a dollar figure, which cannot enter a dollar-weighted calculation at all. Both exclusions raise the DGCI slightly, since both are challengers; we would rather publish a number we can defend than a more flattering one we cannot.
Scope of the reading. The DGCI covers publicly reporting sellers only. Retail media networks inside private retailers, connected-TV inventory sold by private streamers, and the entire long tail of direct-sold publisher advertising sit outside it. The index measures concentration among the platforms that set market pricing, which is a narrower and more defensible claim than measuring the global ad market.
Elena Rodriguez: Run this through a budget conversation and it lands differently than a share-of-market chart. If you are reallocating spend toward a fast-growing challenger because its growth rate looks like momentum, the DGCI says the auction you are leaving is where the money still is — and where the competition for your impressions is being priced. Diversification away from the big three is a defensible strategy for reach, audience quality or risk. It is not, in Q2 2026, a strategy you can justify by pointing at where budgets are moving, because in aggregate they are not moving.
Are Digital Ad Costs Rising, and by How Much?
Meta is the only platform in the panel that publishes both halves of the equation, and its Q2 2026 disclosure is unusually clean: ad impressions up 14%, average price per ad up 12%.
Multiply them and the reported revenue growth falls out:
1.14 × 1.12 − 1 = 27.68% vs. Meta's reported advertising revenue growth of 27%
The gap is rounding. Meta showed 14% more ads and charged 12% more for each one, simultaneously, at a scale of $59 billion a quarter. For a buyer holding budget flat, a 12% price increase buys roughly 89% of the prior impression volume — the arithmetic is 1 ÷ 1.12 = 0.893. Holding reach constant requires about 12% more budget.
The Monetization Intensity Delta (MID)
Most platforms do not publish a price metric. They publish revenue growth and user growth, and leave the buyer to guess how much of the increase is more people versus more money per person. MID — the Monetization Intensity Delta — closes that gap.
Formula:
MID = [ (1 + advertising revenue growth) ÷ (1 + active user growth) ] − 1
Q2 2026 readings:
| Platform | Revenue growth | Active user growth (own metric) | MID | Platform’s own published ARPU growth | Source |
|---|---|---|---|---|---|
| +64% | +18% (Daily Active Uniques) | 38.98% | +36% | Reddit Q2 2026 release | |
| Meta Platforms | +27.49% | +3% (Family DAP) | 23.78% | not published | Meta Q2 2026 release |
| Snap | +19% | +5% (DAU) | 13.33% | +13% | Snap Q2 2026 release |
| +18.17% | +11% (MAU) | 6.46% | +7% | Pinterest Q2 2026 release |
The three platforms that publish an ARPU growth figure let us check the formula against their own arithmetic: MID lands within 2 percentage points on Reddit, within 0.3 on Snap, within 0.6 on Pinterest. That agreement is the reason the Meta reading is worth something — Meta publishes no ARPU metric, and 23.78% is the closest defensible estimate of how hard it is monetizing an audience that grew 3%.
Read across the column and the pattern is stark. Pinterest is growing mostly by adding people. Meta is growing almost entirely by extracting more from people it already had. Reddit is doing both, aggressively.
Scope of the reading. MID compares each platform against its own user denominator, and those denominators are not interchangeable — Meta’s Family DAP, Reddit’s Daily Active Uniques, Snap’s DAU and Pinterest’s MAU count different things over different windows. We publish the readings side by side and do not average them into a composite, because a composite built on four incompatible denominators would be a number with no meaning. The percentage changes are comparable; the levels are not.
What a dollar buys, per platform, per user
The same caution applies to per-user revenue. Each figure below is the platform’s own published metric except Meta’s, which Axis Intelligence Research derived because Meta does not publish one.
| Platform | Revenue per active user, Q2 2026 | Denominator | Note |
|---|---|---|---|
| Meta Platforms | $16.49 | Family daily active people (3.60B) | Axis Intelligence Research calculation: $59,363M ÷ 3,600M |
| $6.18 | Daily Active Uniques | Company-published ARPU | |
| Snap | $3.25 | Daily active users | Company-published ARPU |
| Snap (North America) | $10.26 | Regional DAU | Company-published ARPU |
| $1.86 | Monthly active users | Company-published ARPU | |
| Pinterest (US & Canada) | $8.30 | Regional MAU | Company-published ARPU |
The regional splits carry more planning value than the global numbers. Pinterest monetizes a US or Canadian user at $8.30 a quarter against $1.86 globally — a 4.5× spread that tells you exactly where its advertiser demand sits.
Where Does Retail Media and Commerce Fit?
Amazon’s advertising business is the clearest read on retail media, because it is the only one of scale that files a dollar figure. Amazon’s Form 8-K exhibit discloses the full quarterly series.
| Quarter | Amazon advertising services ($M) | Growth ex-FX |
|---|---|---|
| Q1 2025 | 13,921 | +19% |
| Q2 2025 | 15,694 | +22% |
| Q3 2025 | 17,703 | +22% |
| Q4 2025 | 21,317 | +22% |
| Q1 2026 | 17,243 | +22% |
| Q2 2026 | 19,809 | +26% |
Trailing twelve months to Q2 2026: $76.07 billion (Axis Intelligence Research: sum of the last four quarters above). Growth accelerated from a steady 22% to 26% in the most recent quarter, though Amazon moved Prime Day into Q2 this year, which inflates the comparison.
Amazon also disclosed something rare — a measured performance claim about its own AI tooling. Advertisers using its Ads Agent see 8% lower cost-per-impression and 6% lower cost-per-acquisition than advertisers who do not. That is company-reported, not independently audited, and the comparison is between self-selecting groups rather than a controlled test. It is still the most specific published number on AI-assisted campaign management from a platform of that size.
The commerce backdrop is measured by the government rather than the platforms. The U.S. Census Bureau put e-commerce at 16.9% of total U.S. retail sales in Q1 2026 on a seasonally adjusted basis, with unadjusted e-commerce sales of $302.3 billion for the quarter. For full-year 2025, e-commerce was 16.4% of total retail on sales of $1,233.7 billion.
Elena Rodriguez: Hold those two series next to each other and the tension in retail media becomes obvious. Amazon’s advertising line grew 26% year over year. The share of American retail that happens online moved from 16.4% for full-year 2025 to 16.9% in Q1 2026. The online shopping pie is barely widening; the fee charged to be visible inside it is compounding at more than a quarter a year. That gap is not a growth story for the merchant paying it. It is a margin transfer, and it belongs in the contribution-margin model before the next annual plan gets signed, not after.
How Many People Work in Digital Marketing, and What Do They Earn?
The Bureau of Labor Statistics published its May 2025 Occupational Employment and Wage Statistics with national counts for every marketing-adjacent occupation code. Axis Intelligence Research summed the eight codes that make up a working digital marketing function.
| Occupation (BLS OEWS, May 2025) | Employment | Mean annual wage |
|---|---|---|
| Market research analysts and marketing specialists | 899,580 | $89,490 |
| Marketing managers | 395,240 | $177,770 |
| Public relations specialists | 283,380 | $84,120 |
| Graphic designers | 197,830 | $70,560 |
| Web and digital interface designers | 113,330 | $117,490 |
| Advertising sales agents | 91,700 | $80,670 |
| Public relations managers | 74,850 | $164,760 |
| Advertising and promotions managers | 21,470 | $154,280 |
| Total (Axis Intelligence Research) | 2,077,380 | — |
Two million American jobs sit inside the function. The wage spread inside it is wider than the headline suggests: a marketing manager’s mean of $177,770 is roughly double the market research analyst’s $89,490, and the analyst category is more than twice as large. The median hourly wage for marketing managers is $80.19, against $37.87 for analysts and specialists.
Note the structural detail buried in the table. Advertising and promotions managers — the classical ad-agency job title — number 21,470 nationally. Marketing managers number 395,240. The org chart moved in-house and the BLS codes recorded it.
What Compliance Rules Now Carry Financial Risk for Marketers?
The Federal Trade Commission’s Rule on the Use of Consumer Reviews and Testimonials, codified at 16 CFR Part 465, became fully effective on October 21, 2024. The FTC’s own warning letter template, issued December 22, 2025, states the exposure directly: a court could impose civil penalties of up to $53,088 per violation.
The number that matters operationally is not the ceiling but the unit. A violation is not a campaign. The rule reaches fake or false reviews and testimonials, buying positive or negative reviews, and buying fake indicators of social media influence — and it reaches the advertiser’s agencies, not only the advertiser. A single incentivized-review push touching fifty listings is fifty potential violations before anyone has argued about intent.
Elena Rodriguez: Every operator I have watched run this calculation gets the same surprise: the review-generation line item is usually the cheapest thing in the marketing budget and now carries the largest tail risk in it. The rule has been fully effective for nearly two years and the FTC has moved from writing it to sending letters about it. Price the exposure into the vendor contract, or drop the tactic. Those are the two defensible positions.
Methodology
Collection. Every figure in this article was traced to a primary document fetched during production on August 11, 2026: Alphabet’s Q2 2026 earnings slides, Meta’s Q2 2026 results release, Amazon’s Form 8-K Exhibit 99.1 as filed with the SEC, Pinterest’s and Reddit’s Q2 2026 results releases, Snap’s Q2 2026 results release, the U.S. Census Bureau’s Quarterly Retail E-Commerce Sales report, the BLS Occupational Employment and Wage Statistics Table 1 for May 2025, and the FTC’s Consumer Review Rule warning letter template. No aggregator or secondary summary was used for any figure. Source URLs and retrieval dates for all 85 observations are in the downloadable dataset.
DGCI formula. DGCI = (ΔAlphabet + ΔMeta + ΔAmazon) ÷ (Σ Δ all panel members) × 100, where Δ is the year-over-year change in quarterly advertising revenue. Panel: Alphabet (sum of Google Search and other, YouTube ads and Google Network), Meta (advertising segment), Amazon (advertising services), Pinterest (total revenue, advertising-dominant), Reddit (advertising revenue). Q2 2026 reading: 98.27%.
MID formula. MID = [(1 + advertising revenue growth) ÷ (1 + active user growth)] − 1, computed per platform against that platform’s own published active user metric. Validated against the three panel members that publish an ARPU growth figure; agreement within 2 percentage points in all three cases.
Derived values. Three figures in this article are Axis Intelligence Research calculations rather than direct disclosures, and each is flagged as such in the dataset with its method note: Reddit’s Q2 2025 advertising revenue of $464.6 million (implied from Reddit’s reported 64% growth), Alphabet’s total advertising revenue (sum of its three disclosed advertising lines), and Meta’s revenue per daily active person of $16.49 (advertising revenue ÷ Family DAP, because Meta publishes no ARPU).
Currency and periods. All financial figures are U.S. dollars as reported, not adjusted for foreign exchange unless noted. Quarters are calendar quarters ended June 30. BLS wage data reflects May 2025, the most recent OEWS reference period; the Census e-commerce reading reflects Q1 2026, with the Q2 2026 estimate scheduled for August 18, 2026.
Verification. All arithmetic was recomputed independently of drafting. Every prose figure was matched against a dataset row before publication.
About This Dataset
digital-marketing-statistics-2026.csv contains 85 observations covering platform advertising revenue, user metrics, per-user monetization, DGCI and MID components, U.S. e-commerce penetration, marketing occupation employment and wages, and advertising compliance penalties. Each row carries source_org, source_document, source_url, retrieved_date, is_primary, axis_calculated and method_note.
Coverage: Q1 2025 through Q2 2026 for platform financials; May 2025 for occupational data; Q1 2026 for U.S. e-commerce.
Geography: global for platform revenue, United States for e-commerce, employment and regulatory data.
License: CC BY 4.0. Free to use, share and adapt with attribution.
Citation line: Axis Intelligence Research, Digital Marketing Statistics 2026, 2026.
Cite This Research
APA Axis Intelligence Research, & Rodriguez, E. (2026, August 11). Digital marketing statistics 2026: Ad spend, platform share and cost data. Axis Intelligence. https://axis-intelligence.com/digital-marketing-statistics/
MLA Axis Intelligence Research, and Elena Rodriguez. “Digital Marketing Statistics 2026: Ad Spend, Platform Share and Cost Data.” Axis Intelligence, 11 Aug. 2026, axis-intelligence.com/digital-marketing-statistics/.
Chicago Axis Intelligence Research and Elena Rodriguez. “Digital Marketing Statistics 2026: Ad Spend, Platform Share and Cost Data.” Axis Intelligence, August 11, 2026. https://axis-intelligence.com/digital-marketing-statistics/.
Copy any format above and paste directly.
Frequently Asked Questions
Does the DGCI mean challenger platforms are not worth buying?
No, and the index should not be read that way. DGCI measures where incremental dollars land in aggregate, not where a given advertiser gets the best return. Reddit’s 64% advertising growth on 18% audience growth says advertisers who tested it are spending more per user, which is a demand signal. What the index rules out is the argument that budget is broadly migrating away from the big three. In Q2 2026 it was not — 98.27% of new dollars went to three sellers.
If Meta’s price per ad rose 12%, why did my CPMs move differently?
Meta’s figure is a global blended average across every delivered ad, every format, every country and every optimization objective. Your account sits in one vertical, one geography and one auction. Meta’s own 10-Q defines average price per ad as total advertising revenue divided by ads delivered, which makes it a platform-wide monetization measure rather than a CPM. Use it to sanity-check direction and magnitude; plan from your own account data.
Why exclude Snap from the panel when it clearly sells advertising?
Because Snap’s Q2 2026 press release reports total revenue of $1.599 billion without an advertising breakout, and its revenue now includes a fast-growing subscription business. Dropping a total-revenue figure into a panel of advertising-revenue figures would overstate Snap’s advertising and corrupt the index. Snap’s numbers are in the dataset, marked as total revenue, so anyone can recompute the panel with Snap included and see the effect.
What is the difference between MID and ARPU growth?
Almost nothing, arithmetically — and that is the point. Where a platform publishes ARPU growth, MID reproduces it within two percentage points. MID exists because Meta, the single largest seller in the panel, publishes no ARPU metric at all. The formula lets a buyer compute the equivalent number from the two figures Meta does publish, and the validation against Reddit, Snap and Pinterest is what makes the Meta reading trustworthy.
Is the Google Network decline a signal about programmatic open-web inventory?
It is one data point, and a company-reported one. Alphabet’s Google Network segment fell 0.69% year over year to $7.30 billion while its owned-and-operated Search and YouTube inventory grew 16.76% and 12.85%. That divergence is consistent with budget concentrating on first-party inventory, but Alphabet does not disclose why the segment declined, and one issuer’s reselling business is not the whole open web. Treat it as corroborating evidence rather than proof.
How much should a US marketing team budget per head in 2026?
BLS mean annual wages for May 2025 give the floor: $177,770 for a marketing manager, $89,490 for a market research analyst or marketing specialist, $117,490 for a web and digital interface designer. Those are wages, not loaded cost — employer taxes, benefits, tooling and overhead typically add a meaningful multiple on top, and the OEWS reference period is May 2025, so a year of wage movement sits between the data and today’s offer letter.
Does the FTC review rule apply to my agency or only to my brand?
Both. 16 CFR Part 465 reaches advertisers and the parties acting for them, which is why the FTC’s December 2025 warning letters instructed recipients to forward them to whoever manages review practices across affiliated offices. If a vendor generates incentivized reviews on your behalf, the exposure does not stop at the vendor. The maximum is $53,088 per violation, and violations are counted per review, not per campaign.
Where does connected TV and retail media outside Amazon show up in these numbers?
It largely does not, and that is a stated boundary of the panel. Amazon’s advertising services line includes its own video inventory, and YouTube sits inside Alphabet’s disclosure. Retail media networks run by private retailers, CTV inventory sold by privately held streamers, and direct-sold publisher advertising are outside the five reporting sellers we can trace to filings. The panel measures the platforms that set market pricing, not total advertising spend.
Related Axis Intelligence research: Social Media Statistics 2026 · LinkedIn Statistics 2026 · CRM Statistics 2026 · Shopify Statistics 2026 · Influencer Marketing Statistics 2026