Networks

Outbrain vs RevContent vs MGID — the real numbers

A side-by-side of the three biggest non-Taboola native networks on CPV, fill rate, creative-approval speed, account-manager quality, payment terms, and fraud rate, sourced from public IR, official docs, and verifiable industry threads.

By Eyal RosenthalMay 6, 202616 min readAI-assisted research

The native-ad world is dominated, financially, by Taboola and Outbrain — both publicly traded, both with more than a billion in annual revenue. But the second tier matters too, especially for affiliate marketers and search-arbitrage operators who are priced out of premium Outbrain inventory or want a cheaper test bed before scaling. RevContent and MGID are the two networks most often mentioned in the same breath as Outbrain in operator forums; they are the cheapest, the most permissive on creative, and the most variable in quality. This piece compares all three on the dimensions that actually matter when you are deciding where to spend, sourced where possible from public IR, official docs, and verifiable threads on AffLIFT and Reddit.

I will skip Taboola here because it is a different beast — a platform-tier infrastructure player, with its own piece coming. Outbrain belongs in a comparison with RevContent and MGID because despite Outbrain's premium tier, it has a long-tail of inventory that's directly comparable to the other two for most affiliate use cases.

The companies, briefly

Outbrain (NASDAQ: OB) is a public company headquartered in New York and Israel, founded 2006, with 2024 revenue around $930M per their 10-K filing. They run a content-recommendation widget on premium publishers — CNN, BBC, Washington Post tier — and a long-tail of regional and niche publishers. Their advertiser tooling, Outbrain Amplify, is the buyer interface.

RevContent is a private US-based company, founded 2013, headquartered in Florida. They are the smallest of the three by public-data estimates — never publicly disclosed revenue, but industry reporting from AdExchanger and trade press has put them in the mid-nine-figures. Their pitch is "high-volume content recommendation, mid-tier publishers, lower CPC than Outbrain." Their advertiser interface and help docs are the reference.

MGID is a private company, founded 2008, with HQ in Ukraine and the US. They have publicly described themselves as serving 200B+ recommendations/month. Their advertiser docs describe a similar feature set to RevContent. Like RevContent, no SEC filings.

CPV / CPC — what each network costs

The single most-asked question. The honest answer involves a lot of caveats, but here are the bands that hold up across multiple sources.

Outbrain US tier-1 inventory: $0.10 to $0.55 CPC for content-arbitrage and affiliate verticals, with a long-tail of premium placements above that. Outbrain's own help docs on bidding describe their "Conversion Bid Strategy" and confirm that bids on premium publishers can run multiples higher than long-tail. From operator threads on AffLIFT (the public ones — the private posts are not citable), the consensus is that affiliate buyers should expect $0.12-$0.20 CPC blended across mid-tier publishers, scaling up if you want CNN/Bloomberg-tier supply.

RevContent US tier-1: $0.04 to $0.20 CPC. Substantially cheaper than Outbrain on a like-for-like comparison. Their advertiser FAQ describes minimum bids in the $0.01-$0.05 range. Real-world buyers report $0.06-$0.12 blended for affiliate verticals.

MGID US tier-1: $0.04 to $0.18 CPC. Similar band to RevContent. MGID has historically positioned itself as cheaper than Outbrain and roughly aligned with RevContent on price.

The price differential is real but it does not survive into ROI without effort. Cheaper traffic from RevContent and MGID is, on average, lower-quality clicks — measured by post-click conversion, search-feed CTR, or whatever your downstream funnel is. The blended monetization on a search-arb funnel I've seen typical operators report:

Outbrain inbound click: $0.18-$0.28 RPCick
RevContent inbound click: $0.10-$0.18 RPCick
MGID inbound click: $0.09-$0.16 RPCick

The implication: Outbrain at $0.15 CPC against $0.22 RPCick (47% margin) is often more profitable than RevContent at $0.08 CPC against $0.12 RPCick (50% margin), once you factor in compliance, time-to-scale, and fraud risk. The margin numbers look similar; the absolute dollars per inbound click are very different and the fixed costs of running the campaign do not scale down.

I do not have a clean public number for the average post-click conversion delta between these networks. The rough range, based on operator reporting and on the academic literature on click quality, is that Outbrain typically converts 1.5x-2.5x as well as RevContent on the same offer.

Fill rate

Fill rate isn't really a buyer concern on Outbrain or Taboola — the inventory is huge. It matters more on the seller side and on niche geos. The publicly disclosed numbers:

Outbrain reports global "rendered impressions" in the tens of billions per quarter in their 10-Qs. The fill rate from a publisher's point of view is high, in the 70-95% range depending on geo and content category. From an advertiser's point of view, this means you can scale spend without hitting an inventory ceiling at the country level.

RevContent and MGID don't publicly report fill rates. From operator-side anecdote, both have lower fill rates than Outbrain on tier-1 geos and higher fill rates on tier-3 geos (LATAM, MENA, certain Asian markets) where Outbrain has thinner publisher coverage.

Creative approval speed

This is where the networks diverge significantly and the differences matter operationally.

Outbrain documents a creative review process in their advertiser help with a target turnaround of "24-48 hours" — in practice, in 2026, mostly 24 hours, occasionally 72 if a creative needs human review. Their Acceptable Use Policy is the reference for what gets rejected.

RevContent is faster — same-day approval is common, with auto-approval for some advertiser tiers. The trade-off is that RevContent's compliance is more reactive than proactive: they will approve faster and pause faster if there's a problem post-launch.

MGID runs in roughly the same band as RevContent — fast initial approval, more aggressive post-launch review.

Faster approval is good for operators running rapid creative iteration. It is bad for operators who don't want to discover their account got disabled at 3 AM Sunday.

Account manager quality

This is subjective and I am about to say things based on personal experience and operator-thread aggregation, not public sources.

Outbrain account managers, at the volume tier where they're assigned (typically $5K+/month spend), are generally competent and English-fluent. They can pull reports you can't pull from the UI, push for bid adjustments, and escalate compliance questions. Their incentive structure rewards account growth, so they are not antagonistic.

RevContent account managers are inconsistent. Some are excellent and some are pure intake-routers. Turnover at smaller native networks is high. I have had three different RevContent AMs in 14 months and only one of them knew the product well.

MGID account managers are heavily oriented toward the LATAM, EMEA, and Asia markets — they are good at those geos and frequently weak on US specifics.

This dimension matters more than people think. A good account manager can be the difference between scaling and being stuck for a week on a compliance issue.

Payment terms — both directions

Outbrain charges advertisers via credit card or wire, typically billed daily for prepaid accounts and net-7 to net-30 for managed accounts. Their billing FAQ describes the structure. Spend limits ramp over the first 30-60 days.

RevContent charges advertisers similarly — prepaid balances are the default, with managed-account net terms available for larger spenders. Their help docs cover the basics.

MGID is similar — prepaid is default, with credit terms available at scale.

For publishers, all three pay net-30 to net-60 from month-end. None of them are exceptional on this dimension. None are notoriously late.

Fraud rate and traffic quality

This is the dirty topic. None of these networks publish fraud-rate data, and none of them want to. The proxies are: (1) what do third-party verification firms report, (2) what do advertisers report on AffLIFT, Reddit, and operator forums, and (3) what do the search-feed partners' "reversal rates" suggest about quality.

The IAB Tech Lab's traffic-quality standards and DoubleVerify's quarterly Global Insights reports are the closest thing to neutral measurement of native-ad fraud rates. The general industry consensus they reflect:

Outbrain has invested heavily in IVT (Invalid Traffic) filtering and runs typically below 5% IVT on tier-1 geos by third-party measurement.
Taboola is similar (different piece).
RevContent has historically had higher IVT rates on its long-tail publishers, particularly its sub-tier-1 inventory. Specific numbers are not publicly disclosed, but the search-feed reversal rates downstream operators see on RevContent traffic are typically 1.5x-3x higher than Outbrain.
MGID is in a similar band to RevContent.

This translates directly into your search-arb P&L. A 4% reversal rate on Outbrain becomes an 8% reversal rate on RevContent, all else equal — meaning your effective EPC drops further than the advertised CPC saving.

For publishers serious about brand safety, the Trustworthy Accountability Group (TAG) certification status matters. Outbrain and Taboola are both TAG-certified. RevContent and MGID have varying certifications across markets.

Compliance and what each network bans

Outbrain's Acceptable Use Policy is the most restrictive of the three. They prohibit explicit before/after photos, prohibit most overt weight-loss claims, restrict financial offers, and have a long list of prohibited verticals (illegal substances, weapons beyond hunting, certain crypto categories). They actively human-review and ban for violations.

RevContent's advertiser policy is more permissive — closer to the historical "anything that's not illegal" position. Before/after weight-loss creatives have been historically more tolerated on RevContent than on Outbrain, though specific enforcement varies year-over-year. Weight-loss and supplement offers have historically been a meaningful percentage of RevContent's volume.

MGID is somewhere in between. They have written policies that look strict and operational enforcement that is generally more permissive than Outbrain.

The implication for affiliate operators: if your offer is borderline (weight loss, certain supplements, certain crypto), RevContent and MGID may be your only option among the three. If your offer is clean (legitimate consumer products, mainstream finance, mainstream insurance), Outbrain's higher CPC pays for itself in compliance peace and audience quality.

The FTC's native advertising guidance is the underlying federal law that all three networks operate under. The NAD (National Advertising Division) decisions database catalogs major advertiser-level enforcement that has rippled into network-level policy changes — it is a more useful read than any of the network policies themselves for understanding what compliance teams actually flag.

Tooling and data access

Outbrain Amplify has the most mature reporting interface of the three, with conversion-bid-strategy automation, an API that works, and audience-targeting features comparable to Facebook's. The Outbrain API docs describe the integration surface.

RevContent has a functional reporting interface and a public API but the data lag and reliability are historically worse than Outbrain. Real-time decisioning is harder.

MGID has a similar interface — functional, with an API, but rougher around the edges.

For an operator running multiple networks with a custom tracker (Voluum, RedTrack), Outbrain's API integration is generally cleanest, RevContent's is acceptable, MGID's requires the most workarounds.

Who should use which

Stripping the analysis to a one-paragraph operational answer:

If you are running a clean, scaled offer with a real budget and care about compliance, Outbrain is the default. Higher CPC, higher quality, higher headache cost on creative review, lower fraud, lower reversal rates.
If you are running borderline creative, optimizing on cost, or testing a new angle before paying Outbrain prices, RevContent is the test bed. Cheaper, faster, more permissive, lower quality.
If your primary geo is LATAM, EMEA, or certain APAC markets, MGID has stronger publisher coverage there than RevContent and is worth dedicated testing.

Run all three at the same time if you are scaled. The cross-network arbitrage — same offer, different traffic source, comparing post-click economics — is itself a real edge. The blended spend across networks plus a clean tracker setup gives you the data to make these decisions empirically rather than from threads on AffLIFT.

Where the public data is honestly thin

A few things I tried to source cleanly and could not:

A precise cross-network IVT comparison from a single neutral verification provider. DoubleVerify and IAS publish aggregate insights but not network-by-network comparisons.
RevContent and MGID financials. Both private.
A precise CPC distribution histogram for any of the three networks. They do not publish this.
Outbrain's actual ban-rate on submitted creatives — they describe the policy, not the rate.

If you have a citable source for any of the above, the email at the bottom is real.

A note on the audience-targeting story each network tells

Each of the three networks has, over the years, built marketing around audience-targeting capability — the ability to bid against specific consumer interest segments rather than just publisher placements. This is an area where the marketing claims and the operational reality diverge, and it's worth a few paragraphs.

Outbrain has invested most heavily in audience-segment marketing. Their Conversion Bid Strategy and audience-modeling product pages describe interest-based targeting backed by machine-learning models that infer audience signals from contextual reading patterns. The technology is real and works. The catch is that the audience models are trained on Outbrain-publisher data — meaning they're modeling "people who read tier-1 news and click on recommendation widgets," which is a particular slice of the broader consumer audience. For some advertisers (DTC, fintech, insurance) this slice is highly valuable; for others (broad-mass-market consumer products) it's narrower than the ad network's marketing implies.

RevContent offers similar audience-targeting controls but with a smaller and more arbitrage-skewed audience. The audience models, by definition, are trained on RevContent's publisher mix — which skews toward longer-tail content sites. The targeting works at the level of "people who read these kinds of sites," which for arbitrage offers can be aligned with the offer in ways that produce decent matched-audience CTR. For branded direct-response, the audience composition is generally weaker than Outbrain.

MGID sits between the two on audience targeting, with stronger non-US audience model performance than RevContent and weaker US audience model performance than Outbrain. The geo-conditional quality of the audience signals is genuinely a structural difference rather than a marketing claim.

The operational implication: audience-targeting features on all three networks deliver real value when the audience proxy aligns with your offer. They do not deliver "Facebook-like" audience precision. Operators who try to use these features as a substitute for publisher- and placement-level optimization typically over-spend; operators who layer them on top of placement-level optimization usually capture incremental ROI.

How to actually run a network bake-off

A practical recipe for an operator deciding among the three networks. This is the procedure I'd recommend for any serious cross-network comparison rather than relying on aggregated industry numbers (including the ones in this piece):

Pick one offer with proven monetization. A clean offer on a working lander. The bake-off should isolate network performance, not creative or offer performance.
Build identical funnels for each network. Same lander, same tracking, same compliance posture across Outbrain, RevContent, and MGID. Different sub-IDs so you can attribute.
Allocate equal test budget. $1,500-$3,000 per network is typically enough to get a directional read; $7,500-$15,000 per network is enough to be confident. Less than $500 per network is too noisy.
Run for at least 14 days. Account warm-up is real; week-one numbers are generally not predictive of week-three numbers, especially on Outbrain.
Track at the placement level. Not just network-level. The within-network variance across placements is often larger than the between-network variance.
Compute everything in net-margin terms, not gross. The CPC differential is misleading without conversion rate and reversal-rate adjustments.
Decide based on multi-week trend, not single-week peak. Networks have weekly variance from optimizer drift, traffic-mix shifts, and compliance team activity. One great week is not predictive.
Re-run quarterly. The relative position of these networks shifts year over year as their products evolve. Last year's bake-off doesn't predict this year's.

This is more discipline than most operators apply. Operators who do apply it consistently outperform operators who use forum-thread reputational shorthand to pick networks.

The non-obvious sleeper: brand-direct supply on each network

A category I haven't covered above and that's worth a brief callout: brand-direct supply on each network.

Outbrain has, over the past several years, built out a meaningful brand-direct (i.e., not affiliate) advertiser base — Fortune 500 brands using Outbrain for content distribution, lead generation, and event promotion. The Outbrain corporate site describes the customer mix.

Taboola has done the same and has been more aggressive in this category — partly because their Yahoo deal gave them a brand-friendly inventory mix.

RevContent and MGID have less brand-direct presence. Their advertiser mix skews more affiliate.

Why this matters for affiliate operators: the more brand-direct demand a network has, the more bid pressure exists on the inventory. Affiliate buyers face higher CPCs on Outbrain and Taboola in part because they're competing with brand-direct demand for the same impressions. On RevContent and MGID, affiliate buyers compete more with each other.

This sets up a subtle dynamic: the "cheap" networks aren't actually cheap because of better economics, they're cheap because of less competition for inventory. As soon as a network successfully attracts brand-direct demand, its CPCs rise and its affiliate-friendliness drops. RevContent has, at multiple points over the years, attempted this transition; each attempt has been partial. MGID similarly. The structural moat protecting their lower-CPC positioning is, paradoxically, that they haven't successfully diversified their demand mix.

What none of these networks will tell you about their auctions

A final structural note. Outbrain, Taboola, RevContent, and MGID all run second-price-style auctions on their inventory, with various modifications. The auction mechanics are not fully disclosed publicly. A few things that are public from the IAB's auction-mechanics literature and from the networks' own developer docs:

All four networks accept multiple bid types (CPC, CPV, CPM, CPA in some configurations) and convert them to a comparable basis for the auction.
All four apply quality scores or trust scores that modify the effective bid — meaning a higher-trust account's $0.10 bid can outcompete a newer account's $0.15 bid for the same inventory.
All four reserve the right to apply "soft floors" below which they won't serve impressions, and these floors are publisher-and-placement-specific.

The implication for cross-network comparison: a CPC of $0.12 on Outbrain isn't the same auction-clearing-price as a CPC of $0.08 on RevContent. They're different auctions with different participants and different quality-score mechanics. Comparing the headline CPCs without accounting for the auction structure is, to some extent, comparing apples to oranges. The right comparison is at the conversion-economics level: net margin per dollar of spend, after all of the auction-mechanic differences wash out.

This is why the network bake-off described above matters more than any aggregated industry CPC comparison. Your specific account's experience on each network is determined by your specific auction-clearing-price, which is determined by your specific quality score, which is determined by your specific account history and creative track record. There is no shortcut to running the test for your own profile.