Zeta Global Holdings
A New York company that runs email, customer data and online ad buying for about 200 big consumer brands, with its own file of data on US adults.
The scores
Growth3/4 ↑
About 28% own growth and $1M customers up 17%, but a third of the headline was bought and resold social ads lead the gain.
Business quality3/4 →
Forrester Leader, 128% retention and a Gap win over Salesforce, but retention is inflated by LiveIntent and agencies own rival data stacks.
Financial position4/4 ↑
Net cash of about $110M, free cash of $202.6M a year and an undrawn $750M revolver; nothing needs outside money.
Execution record3/4 ↑
The 2026 guide rose in February, May and August and GAAP profit arrived, but the scaled-customer metric was dropped and the disclosure lawsuit survived dismissal.
Per share capture2/4 ↑
Share pay equals nearly all trailing free cash and acquisitions were paid partly in shares; the count still rose 5.5% despite buybacks.
Potential2/4 →
A $294.6B ad market, but signed obligations cover only $218.7M of next year and the partnerships carry no volume or term.
Risk profile2/4 ↓
Discovery in a lawsuit over the identity data the product rests on, one customer above 10% and a controlling chief executive.
Valuation2/4 ↓
About 164 times 2026 cash after share pay; 25 times in 2028 needs the plan beaten and share pay cut to about 5%.
Evidence quality3/4 →
Figures are filed and the ad market comes from IAB, but AI adoption, partnerships and buyer returns are management's word and Forrester was read second-hand.
The Verdict
We rate Zeta neutral. It runs email, customer data and online ad buying for big consumer brands, and it is growing fast on its own, about 28% without the company it bought. The business is mixed: the fastest-growing part is buying social media ads for agencies, where it keeps less of each dollar, and staff are paid so heavily in shares that almost no spare cash is left for owners. The price needs that share pay to fall sharply, so it is full, not cheap. The view hinges on share pay falling while growth stays above 20%.
The long caseBrands keep moving ad money online, fastest into social media; Zeta must keep growing on it while paying staff far less in shares.
The evidence under this - 6 claims
- Buybacks of $55.6M in the first half of 2026 already offset part of the dilution, so counting both the expense and the buyback cash would double-count.
- Share pay is falling as a share of revenue, from 13.6% in 2025 to 11.8% in Q2 2026.
- Most coverage and the company itself quote free cash flow before share pay, on which 29.0 times the 2026 guide for a company growing 28% on its own is not extreme.
- Net cash of about $110M lowers what a buyer pays for the business by about 1.5%.
- The $210M of 2026 share pay is my doubling of the first half; the company gives no share pay guide.
- At 23% a year from the 2026 guide, 2028 revenue would be about $2.75B, well above the $2.3B target, so free cash could beat $371M.
- Share pay has fallen from 13.6% to 11.8% of revenue in a year, and a further fall to 5% by 2028 is not impossible.
- The integrated share was higher, at 30%, in 2024 and fell to 26% in 2025, so the mix swings rather than trends.
- Adjusted operating profit margin rose 1.7 points in Q2 2026 despite the mix, so costs elsewhere are falling faster.
The company
A big retailer or bank wants to know who its customers are and reach them at the right moment. Zeta gives it one system for that: it stores the customer records, sends the email and text messages, and buys ads on websites, streaming TV and social media.
What sets it apart is data. Zeta owns a file on more than 245M US adults, so it can recognise people a brand has not identified yet. Forrester ranks it among the three leading email vendors, and Gap moved to it from Salesforce.
The evidence under this - 4 claims
- Gap has not disclosed the agreement and its term and value are not stated anywhere.
- One large win shows Zeta can displace Salesforce; it does not show how often it does.
How it makes money
Brands pay a platform fee, plus charges that rise with how many messages they send and how much data they use. When Zeta buys ads for them, it bills the media too and passes most of it on to the publisher.
The money is concentrated: 197 customers pay at least $1M a year, averaging $1.8M a quarter, and one unnamed customer is more than a tenth of revenue.
The evidence under this - 4 claims
Where the money comes from, stream by stream
| Stream | Share of sales (Q2 2026) | How it earns it |
|---|---|---|
| Direct platform | 72% | Subscription, usage and service fees earned entirely on Zeta's own platform: email and messaging, the customer data system and its identity data. |
| Integrated platform | 28% | Fees for running customers' marketing through channels Zeta does not own, such as social media, largely resold ad spending bought for agencies. |
Who runs it
David Steinberg co-founded Zeta and has run it since 2007; Chris Greiner has been finance chief since 2020 and Steven Gerber president since 2009. Steinberg owns about a tenth of the company but holds a majority of the votes, so outside shareholders cannot overrule him. Bosses' bonuses ride on sales and a profit measure that ignores pay in shares, and their long-term shares on the stock price; nothing rewards cash left over after share pay.
The evidence under this - 3 claims
- Adjusted EBITDA adds back stock pay, so the bonus's profit test does not charge the cost that takes nearly all the free cash (C08, C09); the PSUs pay on the share price, and nothing pays on free cash after share pay or on per-share results.
- In December 2025 the vesting of 205,674 of Steinberg's restricted shares was brought forward two to six months to fund tax on other awards.
Why it wins, and who could beat it
Why customers come back
Once a brand's customer records, audiences and years of campaign rules live inside Zeta, leaving means rebuilding all of it and losing Zeta's data. That is why existing customers spent 28% more in 2025 than in 2024.
That figure flatters, though: it counts customers of a company Zeta bought in late 2024 as if they were old ones. And the claim that customers earn $6 to $7 back per $1 spent comes from Zeta, not from any buyer.
The evidence under this - 4 claims
- Gap has not disclosed the agreement and its term and value are not stated anywhere.
- One large win shows Zeta can displace Salesforce; it does not show how often it does.
- None of these figures is filed or defined in a filing, and the heavy AI users may simply be the largest customers already.
- The return-on-spend figure comes from a study Zeta commissioned, and no buyer publishes one.
The threat
Salesforce sells the same system bundled with its sales software and AI agents. Braze and Klaviyo sell the messaging part, cheaper. The quietest threat is the agencies, about a fifth of Zeta's revenue: Omnicom now owns Acxiom, a data company, and can keep that work in-house.
The evidence under this - 4 claims
- Adjusted operating profit margin still rose 1.7 points to 20.7% in Q2 2026, so the lower-margin media has not yet dragged the overall margin.
- Management expects agencies to add higher-margin channels after social, which would reverse the mix.
The rivals: CRM, BRZE, KVYO, OMC, TTD.
Who else does this, rival by rival
| Ticker | Rival | How far along | Priced against it |
|---|---|---|---|
| CRM | Salesforce Marketing Cloud with AgentforcePriced against it: Different product | On sale | Different product |
| BRZE | BrazePriced against it: Yes | On sale | Yes |
| KVYO | Klaviyo B2C CRMPriced against it: Yes | On sale | Yes |
| OMC | Omnicom's Omni platform with AcxiomPriced against it: Different product | On sale | Different product |
| TTD | The Trade DeskPriced against it: One segment only | On sale | One segment only |
How big it could get
How big this could get
US companies spent $294.6B on online ads in 2025, 13.9% more than the year before, and social media grew fastest. Zeta's $1.57B of yearly revenue is about half a percent of that, but nobody measures the software part on its own, so the true size of its market is unknown.
The evidence under this - 2 claims
Who has signed
Very little is locked in. Signed work for the next 12 months is $218.7M, against about $1.82B of revenue expected this year. Gap's deal is multiyear with no size given, and the OpenAI, Palantir and Snowflake partnerships carry no amounts at all.
The evidence under this - 3 claims
- Gap has not disclosed the agreement and its term and value are not stated anywhere.
- One large win shows Zeta can displace Salesforce; it does not show how often it does.
- No volume, value or term has been disclosed for any of the three partnerships.
- OpenAI is also a supplier Zeta pays, and its ad platform lets brands buy directly, so the partner can also compete.
The results
The numbers
Sales grew 36% over the twelve months to Jun 2026, faster than the 30% in the year to Dec 2025.
Operating margin is 1.6%, up from −23% in the year to Dec 2023.
Running costs take 58¢ of each sales dollar, down from 116¢ in the year to Dec 2021.
Shareholders' money lost 0.3% a year, up from −256% in the year to Dec 2022.
It turned 14% of sales into free cash, $224M over the last twelve months.
It holds $310M of cash against $197M of debt, $112M more cash than it owes.
The share count rose 195% in 4.5 years, and pay in shares costs 12% of sales.
All the numbers
Filed figures from the company's 10-K and 10-Q, read 2026-10-04. Fiscal years end in December.
| Dec 2021 | Dec 2022 | Dec 2023 | Dec 2024 | Dec 2025 | 12 mo to Jun 2026 | |
|---|---|---|---|---|---|---|
| Size and growth | ||||||
| Sales | $458M | $591M | $729M | $1.01B | $1.30B | $1.57B |
| Sales growth | +25% | +29% | +23% | +38% | +30% | +36% |
| Profit | ||||||
| Gross margin | 62% | 64% | 62% | 60% | 61% | 59% |
| Operating margin | −54% | −44% | −23% | −6.8% | 0.4% | 1.6% |
| Net margin | −54% | −47% | −26% | −6.9% | −2.4% | −0.1% |
| What growth costs, per sales dollar | ||||||
| Research and development | 14% | 12% | 10% | 9.0% | 9.0% | 9.4% |
| Sales and marketing | 50% | 51% | 40% | 31% | 26% | 24% |
| General and admin | 41% | 36% | 28% | 20% | 18% | 17% |
| Pay in shares | 57% | 51% | 33% | 19% | 14% | 12% |
| Capital spending | 2.1% | 3.8% | 2.8% | 2.6% | 1.1% | 1.1% |
| Cash | ||||||
| Cash from operations | $44.3M | $78.5M | $90.5M | $134M | $199M | $241M |
| Free cash flow | $34.8M | $56.3M | $70.0M | $108M | $185M | $224M |
| Balance sheet | ||||||
| Cash and short-term investments | $104M | $121M | $132M | $366M | $320M | $310M |
| Debt | $184M | $184M | $184M | $196M | $197M | $197M |
| Net cash | −$79.8M | −$62.8M | −$52.4M | $170M | $123M | $112M |
| Returns and shares | ||||||
| Shares (diluted) | 86.9M | 139M | 157M | 186M | 221M | 257M |
What growth costs
Growth is making Zeta more profitable on paper: selling, office and research costs take a shrinking slice of each sales dollar. Most of that improvement is pay in shares fading from its peak after the listing, not cheaper growth. The cost of delivering the service does not shrink, and rises as more sales come from reselling social media ads.
The evidence under this - 3 claims
The research and development line is the same figure on the income statement and in the filer's tag, but stock pay sits inside every line, so the fall in each line is partly the share awards made around the 2021 listing rolling off rather than cash costs growing slower than sales.
Even taking stock pay out entirely, the operating cost lines still fell several cents per sales dollar, so some of the leverage is real cash leverage (C44).
The last report (4 Aug)
Quarter ended June 2026, filed figures.
Q2 2026 looked like another clean beat. Revenue was $442.8M, Zeta made a small profit under the official accounts, and it raised its forecast for the year again.
Underneath, the mix shifted. The cost of delivering revenue grew 54.7%, faster than sales, because agencies bought more social media ads through Zeta, and the share of revenue earned that way rose from 25% to 28%. Shares handed to staff still cost $52.1M, about 12 cents of every revenue dollar.
What it did to the thesisIt confirmed the growth is real, even without the business it bought. It also showed the two drags the view rests on: lower-margin resold ads are a growing share of revenue, and pay in shares still takes a large slice, so owners get less than the headline suggests.
The quarter in detail
Sales growth from the previous quarter, filed figures.
Earnings per share $0.21 against $0.20 expected; $0.03 as filed, the gap being what the company adds back.
The evidence under this - 6 claims
- Adjusted operating profit margin still rose 1.7 points to 20.7% in Q2 2026, so the lower-margin media has not yet dragged the overall margin.
- Management expects agencies to add higher-margin channels after social, which would reverse the mix.
What management did with the money
Insiders in 2026Q2: 3 purchases and 11 sales, from their Form 4 filings.
The price, and where we disagree
In short. Full. A fair price for the growth if you ignore pay to staff in shares, but owners only get paid if that share pay falls sharply.
The price rose 83.2% in six months, and most of that was buyers paying more for each dollar of sales rather than sales growing. Two tests are close. On 3 November Zeta reports Q3 2026, where it has forecast growth without acquisitions of 23% to 24%. And the lawsuit over its data is now in discovery, where documents come out.
The evidence under this - 3 claims
- The Q3 own-growth guide of 23% to 24% is below the 28% Q2 just delivered, so a slowdown is built into management's own numbers.
- Management says it builds a 2% to 5% cushion into guidance, so the guide is a floor it expects to beat, not a forecast.
How we read the multiple
Buyers pay about 29 times the spare cash Zeta expects this year, the cash left after running and investing in the business. After paying staff in shares, very little is left: about 164 times.
For the price to look fair by 2028:
- Cash left after share pay must reach about $296M.
- Over the last year it was about $8M.
- Getting there needs Zeta to beat its own 2028 plan and cut share pay to around 5% of sales.
Against its own history. Buyers now pay more than twice what they usually have for each dollar of Zeta's sales, so the recent rise was mostly enthusiasm.
The evidence under this - 4 claims
- Most coverage and the company itself quote free cash flow before share pay, on which 29.0 times the 2026 guide for a company growing 28% on its own is not extreme.
- Net cash of about $110M lowers what a buyer pays for the business by about 1.5%.
- The $210M of 2026 share pay is my doubling of the first half; the company gives no share pay guide.
- At 23% a year from the 2026 guide, 2028 revenue would be about $2.75B, well above the $2.3B target, so free cash could beat $371M.
- Share pay has fallen from 13.6% to 11.8% of revenue in a year, and a further fall to 5% by 2028 is not impossible.
- The $2.3B target implies only about 12.5% a year from the $1.82B 2026 guide, so the plan is either conservative or already assumes a sharp slowdown.
- The targets were set before the 2026 guide was raised three times and have not been restated since February 2026.
- Buybacks of $55.6M in the first half of 2026 already offset part of the dilution, so counting both the expense and the buyback cash would double-count.
- Share pay is falling as a share of revenue, from 13.6% in 2025 to 11.8% in Q2 2026.
Priced against its rivals
The competitors its own annual report names, each confirmed against the rival's filing. Measured on price for each $1 of spare cash, cheapest first:
| Ticker | Company | price for each $1 of spare cash |
|---|---|---|
| KVYO | Klaviyo | 19.3× |
| BRZE | Braze | 32.8× |
| ZETA | Zeta Global Holdings | 32.8× |
The middle of the other 2 is 26.0. Zeta Global Holdings Corp. is 2nd cheapest of 3.
Rivals' figures come from a data provider, so read the spread of the group rather than any single pair.
Braze and Klaviyo, its closest rivals, cost less for each dollar of sales, and they also pay staff heavily in shares.
Also checked and set aside: Adobe (a substitute, not the same product).
Where we disagree with the market
Zeta is an AI platform that will keep growing above 20% a year, so it deserves a much higher price than a year ago.
The growth is real, but resold ads and share pay mean each share gets far less of it than revenue suggests.
The market is paying for growth and the new AI tools. On the last call nobody asked about share pay, about resold ads or about the lawsuit.
We may be wrong if share pay keeps falling as quickly as it did this year, from 13.6% of revenue to 11.8%, while growth stays above 20%. Then the cash left for owners would grow far faster than revenue.
What analysts say, and the evidence
The evidence under this - 5 claims
- Buybacks of $55.6M in the first half of 2026 already offset part of the dilution, so counting both the expense and the buyback cash would double-count.
- Share pay is falling as a share of revenue, from 13.6% in 2025 to 11.8% in Q2 2026.
- The integrated share was higher, at 30%, in 2024 and fell to 26% in 2025, so the mix swings rather than trends.
- Adjusted operating profit margin rose 1.7 points in Q2 2026 despite the mix, so costs elsewhere are falling faster.
What could go wrong, and how we would know
The risks
- The data lawsuit. A court let a lawsuit over Zeta's data gathering go ahead in July 2026. Anything damaging found in discovery hits the data Zeta sells.
- Lower-quality growth. More growth comes from buying social ads for agencies, which leaves Zeta less of each dollar than its own software.
- One big customer. An unnamed customer is more than a tenth of revenue, and most customers are not locked into long contracts.
- Owners' slice keeps shrinking. Shares outstanding rose 5.5% in a year despite buybacks, partly to pay for acquisitions.
- One holder controls the votes. The chief executive holds most of the votes through ten-vote shares, so outside owners cannot outvote him.
The evidence under this - 6 claims
- A motion to dismiss tests only whether the complaint is plausible, not whether it is true, and no regulator has acted on the data claims.
- Revenue grew 29.7% in 2025 after the report, so customers did not leave on it.
- The integrated share was higher, at 30%, in 2024 and fell to 26% in 2025, so the mix swings rather than trends.
- Adjusted operating profit margin rose 1.7 points in Q2 2026 despite the mix, so costs elsewhere are falling faster.
What we still do not know
- How much of revenue is ad space Zeta buys and passes on, rather than its own software and data. It does not say.
- Who the customer above a tenth of revenue is.
- What discovery in the lawsuit shows about how Zeta collects consumer data.
- What share pay will be for 2026 as a whole; Zeta gives no forecast for it.
The evidence under this - 4 claims
What would change our view
- Neutral→AttractiveThe price falls on an unchanged business
It needs all of:
- ○The market value falls to 25 times the past year's spare cash before share pay, from about 36.5 today.about $20.17 today, 32% below the last close
- ○Growth without acquisitions or election ads stays at 20% or more.checked at the next results, due by 3 Nov
- Neutral→AttractiveShare pay falls while growth holds
It needs all of:
- ○Shares paid to staff fall below 8 cents of every revenue dollar in a quarter, from 11.8 now.checked at the next results, due by 3 Nov
- ○Growth without acquisitions or election ads stays at 20% or more.checked at the next results, due by 3 Nov
- Neutral→UnattractiveOwn growth stalls or the data case goes against Zeta
It needs any of:
- ○Growth without acquisitions or election ads falls below 15%.checked at the next results, due by 3 Nov
- ○A court or regulator finds Zeta gathered consumer data without proper consent.watched as it happens
We would also be wrong if
- The cost of delivering revenue passes 45% of sales for a full year: resold ads are taking over.
- Share pay stays above 10% of revenue through 2027.
What is coming
- 3 November 2026 Q3 2026 results: Whether own growth holds at 23% or more with cost of revenue at 41% of sales or less, and whether net dilution stays inside the 3% to 4% target.
Every dated event ahead
- 3 November 2026 Q3 2026 results: Whether own growth holds at 23% or more with cost of revenue at 41% of sales or less, and whether net dilution stays inside the 3% to 4% target.
- February 2027 Q4 2026 results and the 2027 guide: Whether 2027 share pay falls as a share of revenue toward the 2028 plan, and whether growth holds once political candidate revenue drops out.
Since we wrote this
- 6 OctNewsZeta agrees to buy Senso, an AI engineering firm, price unread. View heldZeta Global agrees to acquire Senso.
- 5 OctThesis reviewThe long case now sits under the verdictAfter the five-name pilot, the owner asked for the long case on every thesis.
- 4 OctThesis reviewOwner's checklist: revenue streams, cost of growth, management, catalystsThe owner's checklist found the method silent on who runs the company, what growth costs, where the money comes from and what is dated ahead; this pass fills those four blocks from the proxy and filings.
- 27 SepThesis writtenNeutral, medium convictionGrowth is real and mostly Zeta's own, but share pay takes nearly all the free cash and the fastest-growing part is resold ads, so the price is full.
Also checked 4 times with nothing new to act on.
Sources 20
Sources - 20
Every claim above cites one of these, or states the reasoning that derived it. The quote is frozen at the date it was read: a link can rot, and an agent fetch returns a reading of a page rather than the page. Backs counts the whole claim ledger, so it can name a claim this page does not surface.
- S01primarycompany filing2026-02-25Item 1 business description, customers by industry, super-scaled share of revenue, competition (classes only), risk factors, legal proceedings naming Culper Research, direct and integrated revenue mix 2023-2025, net revenue retention, customer concentration note, dual-class voting, headcountbacksC01C06C15C16C17C19C21C26C39C43C44C49
- S02primarycompany filing2026-08-05MD&A revenue and cost bridge, Marigold contribution, media costs, direct/integrated mix, super-scaled metrics, remaining performance obligations, acquisition liabilities, credit facility, statement of equity and share issuance, legal proceedings updatebacksC01C02C05C06C08C10C16C17C18C20C22C23C26C40C43C44C49
- S03primarycompany filing2026-08-04statements, adjusted EBITDA and free cash flow reconciliations, revenue excluding acquisitions by quarter, Q3 and 2026 guidancebacksC02C03C04C08C10C40
- S04primarytranscript2026-08-04Q2 2026 earnings call transcript (FMP)data/cache/transcripts/ZETA-2026Q2.jsonown growth, AI adoption figures, Gap win, partnerships, agencies at about 20% of revenue, social channel driving cost of revenue, buyback and dilution target; 17 analyst turns from 8 analysts, none on share pay, the mix or the lawsuit
For Q2, revenue came in at $443 million up 44% year over year or 28% excluding M&A.
backsC03C07C24C28C32C33C42 - S05primarycompany filing2026-02-242025 free cash flow, share pay, adjusted EBITDA, February 2026 guidance and the Zeta 2028 targetsbacksC04C08C12C40
- S06primarycompany filing2026-07-27$250M term loan A and $750M undrawn revolver replacing the $550M facility; stated uses include acquisitions and buybacksbacksC22
- S07secondaryvendor data2026-09-27ZETA research pack (FMP market tier, FINRA, SEC XBRL), frozencontent/ZETA/research-pack.jsonprice, price-move decomposition, valuation history, consensus, short interest, filed cohort multiples for Braze and Klaviyo, headcount seriesbacksC10C13C14C26C34C35C39C41
- S08tertiarynews2024-11-13the date and substance of Culper Research's report as reported at the timebacksC15
- S09primaryindustry data2026-04-16US internet ad revenue, social $117.7B up 32.6% and programmatic $162.4B up 20.5%, confirmed on the PR Newswire copy of the same release
reaching $294.6 billion in 2025, reflecting a 13.9% year-over-year increase
backsC25 - S10primarycompany filing2026-03-02Agentforce agents across sales, service, marketing and commercebacksC29
- S11primarycompany filing2026-03-25customer engagement platform; rivals named: Adobe, Salesforce, Iterable, KlaviyobacksC30
- S12primarycompany filing2026-02-10B2C CRM for consumer companies, over 193,000 businesses; rivals named: Mailchimp, Braze, Adobe, SalesforcebacksC30
- S13secondarycompany release2026-03-23Zeta: named a Leader in The Forrester Wave, Email Marketing Service Providers Q1 2026 (company page)Zeta's placement; the report itself is paywalled
Zeta is proud to be named a Leader in The Forrester Wave™: Email Marketing Service Providers Q1 2026.
backsC27 - S14secondarycompany release2026-03-23a rival's account of the same Wave: three Leaders of 12
The report places Cordial among three Leaders out of the 12 vendors evaluated.
backsC27 - S15secondaryvendor data2026-09-27IBKR weekly price history, ZETA, September 2024 to September 2026IBKR get_price_history, contract 495763832weekly closes around the November 2024 short report and the April 2025 lowbacksC15
- S16primarycompany filing2026-02-20the Omni platform including Acxiom, acquired with IPGbacksC31
- S17primarycompany filing2026-09-27two discretionary director sales, no purchases, gifts by the CEO, CFO and presidentbacksC36
- S18primarycompany filing2024-11-18the company's response to the short report: a webinar on accounting controls, data collection and privacybacksC15
- S19primarycompany filing2026-04-24executive officers and their tenure, beneficial ownership and voting power at the April 17, 2026 record date, the 2025 bonus metrics and payout, the 2024 PSUs' share-price test, the Marigold deal bonus and the December 2025 vesting changebacksC46C47C48
- S20primarycompany filing2024-02-28the 2022 and 2023 income statements: cost of revenue, general and administrative, selling and marketing, and research and developmentbacksC44