Publishers and media
Your next subscribers already read you. The meter is the only message they get.
Pageviews, registrations and your subscription platform record how often someone reads and when they cancelled. They can’t say what the reading is for, or what stops a regular from paying. Moonbrush reads the purpose behind the reading, drawing on 7,015 active consumer topics across 25 categories, so offers, win-backs and the audiences you sell all carry a reason.
The decisionWhether a reader pays, and whether an advertiser does.
01Pressures
What publishers are up against
Less traffic means fewer regulars to convert and fewer impressions to sell, and the readers who do pay cost more to keep.
-
Search sends fewer readers
Google search traffic to publishers has fallen hard, and AI answers now cover recipe and how-to queries more than hard news. So fewer casual visits turn into a paying habit.
−38%
Google organic search traffic to U.S. publisher sites tracked by Chartbeat.Source: Reuters Institute, Journalism, Media and Technology Trends and Predictions 2026 (Chartbeat data). Year to November 2025. -
Fewer payers, deeper discounts
The share of Americans paying for online news has slipped. The readers you win often start on a deep intro offer, then leave at the step-up or stay at a price that drags ARPU down.
−4 pts
Change in the share of Americans who pay for online news, year over year.Source: Reuters Institute, Digital News Report 2026. 2026 survey. -
Advertisers buy outcomes, not pageviews
Buyers keep shifting budget to retail media and platforms that tie ads to sales. With fewer impressions to sell, open-exchange CPMs won’t close the gap on their own.
02Blind spot
You know how often they read. Not what the reading is for.
What your data shows
Your analytics
Pageviews, frequency and recency.
Your subscription platform
Starts, step-ups and cancellations.
Your ad server
Impressions, and the segments you already sell.
What it can’t show
The purpose behind the reading
A dinner to cook, a project underway, a trip in planning.
What stops a regular from paying
The meter, one title, the price after the intro or fatigue.
Why a subscriber left
A price step-up, a finished project or cutting back.
What Moonbrush adds
atlas reads the purpose behind your readers’ visits and the open episodes in their lives, so each offer, win-back and advertiser segment carries a reason. The site tag reads your consent manager first.
03Decision
From first visit to paying reader, and back
A reader’s path to paying runs through five moments. Most publishers only speak at one of them.
-
01
A purpose
A recipe, a renovation, a move, a team.
We read
The topics and episodes behind the visit.
You can
Recognize what they came for.
-
02
A habit
They come back, and they register.
We read
Engaged non-subscribers, by purpose.
You can
Show them more of what they came for.
-
03
The wall
The meter runs out. Pay, or leave.
We read
What stops each regular from paying.
You can
Build the offer around their purpose.
-
04
The step-up
The intro price ends.
We read
Who is at risk, and why.
You can
Keep them before the renewal, not after it.
-
05
The win-back
They left for a reason.
We read
Your cancelled file, split by why they left.
You can
Make a return offer for each reason.
04Calendar
When publishing decisions get made
Offers, ad commitments and renewals bunch into a few windows each year, and these are the ones that move the numbers.
January to March
Q1 reset
Ad budgets reset and rates are at their softest, so subscriptions carry the quarter. New-year resolutions bring cooks, savers and the fitness-minded to your pages.
With MoonbrushFind the readers behind January’s habit topics and offer a plan sized to the habit before it fades.
March to June
NewFronts and upfronts
Brands and agencies commit a large share of next season’s video and premium budgets. IAB moved its NewFronts to March in 2026, ahead of May’s TV upfronts.
With MoonbrushBring segments that say why readers are in-market and when, so the pitch carries a reason buyers can’t get from reach alone.
April to September
Project season
Renovations, moves and weddings cluster in spring and summer, and project readers arrive with a deadline. They pay for the project, then leave when it ends.
With MoonbrushOffer project-length plans when project topics appear, and time the win-back for when the project is likely over.
September to November
Annual planning
Advertisers set next year’s budgets and partner lists. Publishers set subscription targets, price plans and the ad packages they’ll take to market.
With MoonbrushStart the 60-day proof in late summer so holdout results are in hand before budgets lock.
October to December
Holiday ad peak
Retail budgets lift CPMs to their high for the year, and advertisers pay most for shoppers in-market for specific gifts and purchases.
With MoonbrushPackage reader segments in-market for a purchase, with timing attached, and deliver them hashed or inside a clean room.
November
Black Friday sales
The deepest offer of the year often brings the biggest wave of starts, while last year’s annual-deal cohort comes up for renewal at full price.
With MoonbrushSplit last year’s cohort by why each might leave before the renewal notice, and aim this year’s offer at regulars who only need a price.
05Cohorts
Three readers, same frequency. Three reasons to pay.
Engaged non-subscribers look identical in your analytics. Read by purpose, they need different offers.
-
The project reader
Here for a renovation, a wedding, a move.
- What holds them back
- The project will end.
- What they answer to
- A short, project-sized offer.
-
The single-title regular
Comes for one title, every day.
- What holds them back
- Won’t pay for the whole bundle.
- What they answer to
- That one title, priced on its own.
-
The subscription-weary
Reads widely, pays for nothing new.
- What holds them back
- One more monthly charge.
- What they answer to
- An annual price, or a bundle with what they already pay for.
Illustrative cohorts, shaped like the splits we return.
06Lessons
What we’ve learned in this market.
-
Paying is still a minority habit.
Most people who read news online pay for none of it. Growth comes from converting the regulars you already have, and keeping them past the step-up.
17%
Of people across 20 richer countries pay for online news.Source: Reuters Institute, Digital News Report 2026. -
People pay for a purpose, not a publication.
The reason someone reads predicts which offer converts: a cook pays for recipes, a renovator for three months of a project. The meter treats them all the same.
-
Advertisers pay for the reason too.
A segment that says why readers are in-market, and when, tells an advertiser more than a demographic does. Enriched first-party segments can be activated hashed or inside a clean room.
07Plays
Three places to prove it first.
Each is scoped to one decision, measured against a holdout you choose and read out on day 60.
Usually through
-
Convert readers who already return
Split engaged non-subscribers by what stops them from paying: the meter, a single title, the price after the intro or subscription fatigue.
What you getEngaged non-subscribers split by what stops them from paying, each with its offer.
-
Win back by why they left
Split your cancelled-subscriber file by why they left: price step-up, finished project or cutting back, each with its own return offer.
What you getYour cancelled file, split by reason, with a return offer for each.
-
Sell audiences with a reason attached
Enrich your first-party reader segments with motivation and timing, then activate them hashed or inside a clean room for advertisers.
What you getAdvertiser segments with motivation and timing, activated hashed or in a clean room.
In the workspace.
Find the regulars before the meter doesDescribe the readers you want in plain English. stage sizes the pool against the graph and shows what sets them apart from the readers who already pay.
A stage screen with demo values.
08Metrics
The numbers publishers report upward
What consumer revenue and ad sales teams take to the board, and what Moonbrush changes about each.
-
Registrations and logged-in rate
Readers who give you an email, and the share of traffic signed in. Known readers feed conversion, newsletters and the first-party segments you sell.
With MoonbrushRegulars the site tag resolves come split by what they read for, so the registration wall can offer the matching newsletter, not a generic sign-up.
-
Starts and conversion rate
New paid subscriptions, and the share of engaged or registered readers who start one.
With MoonbrushEach offer goes to the readers whose barrier it answers, whether the meter, one title or the price, and starts are counted against a holdout you choose.
-
Digital ARPU
Average monthly revenue per digital subscriber, after intro discounts and bundles. Deep discounts lift starts and pull it down.
With MoonbrushMoonbrush separates readers who need a discount to start from those who would pay closer to full price, so the deepest offers go where they’re needed.
-
Churn and step-up retention
The share of subscribers who stop each month, by choice or by failed payment, and how many stay once the intro price ends.
With MoonbrushSubscribers come scored for step-up risk with the likely reason (price, a finished project or cutting back) before the renewal notice goes out.
-
Yield on audience deals
What advertisers pay per thousand impressions on deals built from your first-party segments, set against open-exchange rates.
With MoonbrushSegments carry motivation and timing, reach buyers hashed or inside a clean room, and campaign results can be read against a holdout.
09Proof
How you’ll know it worked.
What we measurePaid starts, step-up retention and win-backs against a holdout you choose, and CPMs on enriched advertiser segments.
The first 60 days
- Week 0
Agree on the decision, the data and the success criteria. Paperwork includes the data use agreement and the acceptance criteria.
- Weeks 1–2
Match keys and attach scores in the clean room or your warehouse. Your match rate is reported before anything is built.
- Weeks 3–8
Activate in your channels against a holdout you choose. Outcomes return as they happen.
- Day 60
Read-out on lift and impact, with the validation report. Then an annual program, if the numbers earn it.
10Questions
What publishers ask us first
-
We already run a dynamic paywall with a propensity score. What does this add?
Your paywall scores likelihood to convert from on-site behavior. Moonbrush adds the purpose behind the reading, drawn from what readers do off your site, so the offer changes, not only the timing of the wall.
-
Most of our audience is anonymous. Does this only work on logged-in readers?
It starts with your subscriber, registered and cancelled files, matched on hashed keys. The site tag reads your consent manager first, then resolves part of your anonymous traffic to individuals and households.
-
Our subscription numbers swing with the news cycle. How would we know it was you?
Every test runs against a holdout you choose, drawn from the same readers in the same weeks, so a big news week lifts both sides. Paid starts and win-backs are read on that difference.
-
How do we know our audience won’t leak into someone else’s segments?
Your data stays in your own instance, is never pooled with another client’s and never trains models shared with other clients. Segments you sell go out hashed or inside a clean room.
11Rules
The rules that shape reader data and ad sales
Privacy, tracking, auto-renewal and children’s privacy laws decide what you can collect, sell and charge, and Moonbrush works inside each.
CCPA/CPRA and state privacy laws
Many state privacy laws let readers use a browser signal to opt out of sale or sharing for ads; California makes browsers offer one from 2027. The site tag sends nothing without consent.
VPPA and wiretap claims over pixels
Class actions target pixels that pass what a reader watched or searched to third parties, under the federal VPPA and California’s CIPA. Audiences go out as hashed identifiers, with no reading history attached.
ROSCA and auto-renewal laws
ROSCA and state auto-renewal laws require clear terms, express consent and easy cancellation; California requires notice before an intro price ends. Moonbrush flags step-up risk in time to pair a retention offer with that notice.
COPPA and children’s data
COPPA requires parental consent before collecting children’s data, and the FTC’s amended rule, binding since April 2026, adds separate consent for sharing it for ads. Moonbrush holds no data about minors.
The lines we don’t cross here
The site tag reads your consent manager and sends nothing without consent.
Reader audiences leave only as hashed identifiers, never as reading histories.
Modeled ethnicity and political ideology are never selectable, in any segment you sell.