PWA.Market Review 2026
PWA builder for affiliate teams · Gambling, Betting, Dating, Nutra, Crypto · metered per app, per month
PWA.Market builds Progressive Web Apps dressed as Google Play listings and sells them to media buyers by the app: nine plan cards on the homepage, seven of them priced, from $25 for one PWA a month to $1,380 for two hundred. The pricing is unusually specific for this corner of the market, the six-step builder is documented on the homepage, and the feature list is openly what it is: parsed Play Store designs, hand-set star ratings, AI-generated reviews, cloak, white pages, scheduled pushes and traffic optimisation on installs, registrations and deposits. What is missing is everything on the other side of the invoice. There is no legal entity, no address, no terms of service, no privacy policy and no refund policy anywhere on the domain, the "trial" costs $25, and the same Business plan is priced $1,380 in Russian and $1,318 in English.
By the numbers
Plan names, prices and quotas were read off the homepage pricing blocks in all three locales at time of writing, reconstructed from the Tilda zero-block element coordinates because the page ships no structured data. On the RU page those blocks are rec599288790 for the B2C cards and rec599326062 for the B2B and special-offer cards; the EN and UA pages carry the same two blocks under different record ids inside the same Tilda project. Cost per PWA is our own arithmetic on the vendor's published figures. Policy-page counts and the two locale price gaps are HTTP status checks and side-by-side reads, listed in full below.
Every plan, and what an app actually costs
Everything is metered the same way: how many PWAs you are allowed to create in a month. Nine cards, seven of them priced, spread over a B2C block, a B2B block and a one-card special-offers strip. The fourth column is ours, and it is the column that makes the range readable.
| Plan | Per month | PWAs | Cost per PWA | What it means |
|---|---|---|---|---|
| Trial | $25 | 1 | $25.00 | Labelled "trial version" and charged in full. There is no free tier. |
| Lite | $110 | 5 | $22.00 | Cheapest plan that is not sold as a trial. |
| Lite+ | $150 | 10 | $15.00 | Doubles the quota for 36% more money. |
| Standard | $310 | 30 | $10.33 | The last tier where a single app still costs more than $10. |
| Premium | $660 | 80 | $8.25 | Best unit price below the top plan. |
| Premium Plus | $935 | 100 | $9.35 | Sits in its own "special offers" strip and is labelled for high ROI: adds unlimited iOS and PWA installs plus unlimited team access. Unit cost goes up, not down. |
| Business | $1,380 / $1,318 | 200 | $6.90 / $6.59 | $1,380 on the RU and UA pages, $1,318 on the EN page, for the identical 200-PWA plan. |
| Custom | Not published | Not published | n/a | Fourth card in the B2C block. No price, no quota, quoted only inside a Telegram chat with a manager. |
| Individual terms of cooperation | Not published | Not published | n/a | Fourth card in the B2B block, and a second unpriced card rather than a repeat of the first. Also quoted only by a manager. |
Two things fall out of that table. First, the volume curve is real but it breaks once: unit cost falls steadily from $25 to $8.25 between Trial and Premium, then rises to $9.35 on Premium Plus before dropping to $6.90 on Business. Premium Plus is not a volume tier at all, it sits in its own special-offers strip and sells unlimited iOS and PWA installs and unlimited team access, so you pay more per app for capabilities rather than capacity. Second, the add-ons are stated in a footnote under the cards, and they are the site's second price gap: $5 for a domain bought from the vendor and $5 for the mirror function on the RU and EN pages, $3 for each of them on the UA page, and free if you point your own domains at it.
Read this before you send $25
PWA.Market takes subscriptions of up to $1,380 a month and publishes no terms of service, no privacy policy and no refund policy. We checked /terms, /privacy, /docs, /api and /faq: at the root every one returns 404. Under /en/ and /ua/ those paths answer 200, but only because any path in those directories serves the locale homepage, so no legal document exists in any locale. There is also no legal entity, company number or registered address anywhere on the site, and the registration form has no terms checkbox to tick, because there is nothing to tick it against.
That matters most where the money is ambiguous, and the money is ambiguous twice. The Business plan is $1,380 on the Russian and Ukrainian pages and $1,318 on the English page for the identical 200-PWA product, and the add-on footnote prices a domain and the mirror function at $5 each on RU and EN and $3 each on UA. With no published terms, nothing on the site tells you which figure is the contract, what happens to your unused quota, or what a refund would even mean. Budget as if the money is unrecoverable once sent, and ask the Telegram manager to state the price, the billing period and the refund position in writing before you pay.
What you actually get
A 6-step builder, documented on the homepage
Link a domain, choose or upload a design including video and a split mode, set the app name, author, rating and category, generate the comments, schedule the pushes, then pick GEO, cloak, prelanding and postbacks. Six screens from empty to live.
Google Play designs, parsed
Every priced plan card lists "parsing designs from Google Play", so the output can be built to look like a store page. The apps are explicitly never uploaded to Play, which is the entire pitch: no store moderation to pass. Step 2 of the builder is the design step itself, where you upload or pick a template, set landing and button colours, add video and enable PWA split mode.
The rating and the reviews are inputs, not outcomes
Step 3 has you add the name, description, author, rating and category. Step 4 generates the review wall with AI or takes comments you paste in. Both numbers a user reads as social proof are fields in a form.
Cloak and white pages across the priced tiers
Cloaking, white pages and interactive prelandings are not an upsell here. Six of the seven priced cards list them verbatim, from the $25 trial upward, alongside Gambling and Betting design templates. The seventh, Premium Plus, is written as a "+" on top of that list and shows only its own extras.
Scheduled and event-based push notifications
Step 5 of the builder: "set up push notifications based on schedule and events". Push is the reason the format exists for affiliates: the install survives, so the retargeting channel survives with it.
Traffic optimisation on installs, registrations and deposits
Step 6 is where you choose GEO, cloak and prelanding and set postbacks and a traffic-optimisation system. The three events are named on the plan cards themselves, as "traffic optimization for installations, registrations and deposits". That is the one part of the stack aimed at measurement rather than appearance.
Domains and mirrors at $5, or $3, your own for free
Priced in a footnote under the plan cards, and the footnote does not agree with itself. On the RU and EN pages a vendor domain costs $5 and the mirror function costs $5. On the UA page both read $3. Linking domains you already control costs nothing in every locale.
Telegram is the entire support surface
Signup demands a Telegram handle next to your email. Support is @pwa_market_support, there is a manager handle, a channel and a public chat, and the custom plans are quoted only in a manager DM. The contacts block has an EMAIL label with no address behind it. No ticketing, no SLA, no docs.
The 404 audit
We requested the obvious paths with a desktop browser user agent. This is the whole result, pass and fail together.
| Path | Status | What is there |
|---|---|---|
pwa.market/ (RU) | 200 | Static Tilda export, project 7475696, roughly 1.5 MB, Cloudflare in front |
pwa.market/en/ | 200 | Machine-translated locale that carries a different Business price |
pwa.market/ua/ | 200 | Third locale. Plan prices match RU, but the domain and mirror add-ons are priced $3 here against $5 on RU and EN |
pwa.market/registration | 200 | Self-hosted Bootstrap dashboard shell: email, password, Telegram handle, account type and your source and vertical. reCAPTCHA, no terms checkbox |
pwa.market/login | 200 | Same shell, Google reCAPTCHA v3 |
pwa.market/terms | 404 | No terms of service exist on the domain |
pwa.market/privacy | 404 | No privacy policy, on a site that collects email plus a Telegram handle |
pwa.market/docs | 404 | No product documentation |
pwa.market/api | 404 | No API surface published |
pwa.market/faq | 404 | No FAQ |
pwa.market/robots.txt | 404 | No crawler directives |
pwa.market/sitemap.xml | 404 | No sitemap |
pwa.market/en/terms | 200 | Soft 404: any path under /en/ or /ua/ answers 200 and serves that locale homepage, so there is still no legal document |
pwa.market/ru.html | 404 | The RU main nav points its Blog item at this relative file, and it does not exist |
blog.pwa.market/blog_en | 404 | Linked from the EN main nav, dead |
blog.pwa.market/blog_ua | 404 | Linked from the UA main nav, dead |
Three smaller signals sit alongside that. Plain http:// serves 200 with no redirect to HTTPS. www.pwa.market does not resolve at all. And HEAD requests are answered with a Cloudflare 403 while a normal GET returns 200, which is a Cloudflare setting rather than a fault, but it does mean automated link checkers see a broken site. Across the 178 Tilda CDN assets the three locales pull in, the newest Last-Modified is 19 February 2025 and the oldest is June 2023. The vendor publishes no build or release date anywhere, so that is the only dating evidence there is.
Two versions of the same fact
A static site exported three times drifts. Here is every place we found the site disagreeing with itself.
| Item | One version says | The other says |
|---|---|---|
| Business plan, 200 PWA a month | $1,380 on the RU and UA pages | $1,318 on the EN page |
| Domains and the mirror function | $5 each in the RU and EN footnote | $3 each in the UA footnote |
| Who Premium and Business are for | RU: "Для небольшой команды", for a small team, on all three B2B cards | EN and UA: "for a large team" on the same two cards |
| The language switcher | EN and UA headers offer all three locales | The RU header offers only EN and RU, and the secondary switcher on each page links its alternate languages back to the page you are already on |
| The current year | Homepage footer: © 2021-2025 | Registration footer: © 2021-2026 |
| Ban risk | PWAs "don't get tagged or banned by advertising sources" | Cloak, white pages and prelandings across the priced tiers |
| The blog | In the main nav of all three locales | blog_en, blog_ua and the relative ru.html all 404 |
| The $25 entry plan | Labelled "пробная версия", "Пробна версія" and "trial version" | Charged in full, no free tier exists |
| Where the pages live | og:url points at blog.pwa.market addresses | Those 404, and canonicals name Tilda files such as pwamarket_en.html |
Two entries have money attached: the Business card is $62 wide and the add-on footnote is $2 wide per domain. The rest add up to a picture of a marketing site that is partially maintained: the product may well be current, but the shop window has two different current years in it and a nav item pointing at nothing. The EN locale is also visibly machine translated, with strings such as "You can purchase domains are provided by PWA.Market also" and "PWA Apps has been created to 2025 year". If you are not a Russian speaker, read the RU page through a translator rather than trusting the EN one, because the EN one is where the price is different.
Headline claims, and what backs them
These are the numbers the homepage leads with. We are reproducing them as claims, because that is all they are.
| Claim | Where it appears | Our reading |
|---|---|---|
| 60,000+ PWA apps created "by 2025" | Homepage counter block | Vendor claim, no methodology, frozen at 2025 |
| 10,000+ service users | Homepage counter block | Vendor claim, not independently checkable |
| 80% average ROI | Homepage, attributed to "partner reviews" | Vendor claim, no sample, no time window, no source |
| "Don't get tagged or banned by advertising sources" | Homepage "Why PWA app?" block | Contradicted by the cloak and white pages sold on the same page |
| "Unlimited installs on iOS and PWA" | Premium Plus card, $935 | Mechanism for iOS installs is never explained anywhere on the site |
| "Sensitive support by a manager" | Every priced tier | Telegram only, no ticketing, no SLA published |
Who it is aimed at
Solo buyers testing Tier-3 volume
The positioning is explicit: Tier-3 geos, Gambling and Betting first, then Dating, Nutra and Crypto. At $110 for 5 apps a month, the Lite plan is priced like a testing budget rather than an infrastructure one.
Teams that need many near-identical funnels
The marketing point of a PWA here is that it can be duplicated without limit and never faces store moderation. Standard at $310 for 30 apps, or Business at $1,380 for 200, is the shape of a team churning creatives.
Buyers moving off webview Android builds
The product pitches itself as the alternative to wrapping an offer in a webview APK. No store review, no signing, no takedown of a listing you do not own.
Not for anyone who needs paperwork
No legal entity, no invoice-bearing company, no terms, no privacy policy, no SLA, no API docs and not even an email address. If procurement has to approve the vendor, this fails at the first question.
Honest limitations
- The marketing contradicts the product on the same page. The homepage says PWAs "don't get tagged or banned by advertising sources", and then sells cloaking, white pages and prelandings across the priced tiers. Those tools exist precisely because tagging and bans happen. Take the feature list as the honest half and the promise as the sales half.
- There are no terms to contradict, which is worse. No terms of service, no privacy policy, no refund policy:
/terms,/privacy,/docs,/apiand/faqall 404 at the root, and under/en/and/ua/they answer 200 only because those directories serve the locale homepage for any path. Either way there is no document, on a service that collects your email and Telegram handle and bills $25 to $1,380 a month. - No legal entity anywhere. No company name, registration number or registered address across all three locales, the registration page or the login page. Escalation means messaging a Telegram handle.
- The same plan has two prices, and so do the add-ons. Business is $1,380 in Russian and Ukrainian and $1,318 in English, for 200 PWAs a month either way. The domain and mirror footnote reads $5 on RU and EN and $3 on UA. Whichever numbers are intended, one set of customers is reading the wrong ones.
- There is no email address on the site. The contacts block prints an EMAIL label with nothing behind it: no address, no
mailto:link, nowhere on any of the three locales. The manager and PR contacts are Telegram links too, and they point at handles branded for the vendor's sister project rather than at named people. - The "trial" is a paid plan. $25 buys one PWA for one month. There is no free tier and no free look at the dashboard, so evaluation starts with a payment to a vendor with no refund policy.
- Enterprise pricing is two DMs. Two of the nine cards, Custom and Individual terms of cooperation, publish no price and no quota at all and are quoted only inside a Telegram chat, which means the largest deals are the least documented ones.
- Support has no floor. Telegram only. No ticketing system, no published SLA, no status page and no API documentation, while "sensitive support by a manager" is listed as a paid feature on every priced tier.
- The site is only half maintained. A blog in the main nav of all three locales where every link 404s, homepage stats frozen at "to 2025 year", a 2021-2025 footer on the homepage and a 2021-2026 footer on the registration page, a Russian
og:descriptionon the English and Ukrainian pages, an editor note pasted into a livehrefon the Individual card, plus no hreflang, no sitemap, no robots.txt and canonicals pointing at Tilda filenames that only resolve because those directories serve the homepage for anything. - The English locale is machine translated. "You can purchase domains are provided by PWA.Market also" is representative. That is cosmetic until you notice the translated page is also the one with the different price.
- The headline stats cannot be checked. 60,000+ PWAs created, 10,000+ users and an 80% average ROI "according to partner reviews" are vendor counters with no methodology, and the first is explicitly frozen at 2025.
- The platform-terms exposure is entirely yours. Pages styled as Google Play listings, parsed Play Store designs, hand-set ratings, AI-fabricated reviews, cloaking and white pages are moderation-evasion tooling. Google, the ad networks and the app stores take a dim view of all of it, and PWA.Market publishes no compliance guidance at all. We are describing what the product does, not recommending you point it at anything.
PWA.Market + PROXIES.SX: no overlap, one disclosure
PWA.Market does not sell proxies or bandwidth on any of its nine plan cards. There is no competitive overlap to manage, so this section can be short and specific. One thing does need saying up front: the PWA.Market homepage carries a partner block that cross-promotes MangoProxy, a residential and mobile proxy vendor and therefore a direct competitor of ours, complete with a signup promo code. We would rather name that than have you find it yourself and wonder why we left it out.
The real complement is testing what your funnel shows other people. A cloak is a decision made about a visitor, and by definition you cannot audit it from your own connection: you always see the money page. The questions that decide whether a $310 month is wasted are all geo-bound. Does the white page render for a Tier-3 mobile visitor on a real carrier IP the way you configured it? Does the push prompt fire on a genuine mobile network rather than your office fibre? Does the install actually land, and does the postback fire on the register and deposit events you wired in step 6?
That is a job for PROXIES.SX: our own 4G/5G mobile and residential network, billed per GB from $4 down to $2.40 at volume, with GB that never expire and endpoints included free. Per-GB matters here because checking a funnel is a handful of megabytes, not a month of traffic, so you are not renting a port between campaigns. Non-expiring GB matter because a testing budget bought in January is still there in March. And free endpoints mean you can hold one per geo you buy traffic in without a per-port bill.
PWA.Market
Funnel layer, the app page, the prelanding, the pushes and the postbacks.
PROXIES.SX
Network layer, a real carrier IP in each target geo, per GB, endpoints free.
Verification layer
Check what the visitor really got with our proxy tester and DNS leak test.
Verdict
The product looks real and reasonably mature. Its copyright line runs from 2021, it ships three locales and a self-hosted dashboard, seven of its nine plan cards are fully public with exact quotas from 1 to 200 PWAs a month, its add-ons are priced in the open in a footnote, and the six-step build flow is laid out where anyone can read it. In a segment where most vendors publish a Telegram handle and nothing else, that is genuinely above average. What holds it to 3.2 is everything the money side is missing. A service billing up to $1,380 a month has no legal entity, no address, no terms, no privacy policy, no refund policy and not one email address. Its "trial" costs $25. Its headline plan is $62 cheaper if you read English, and its add-ons are $2 cheaper if you read Ukrainian. Its blog is in the nav of all three locales and 404s in all three. Its counters are unverifiable and frozen at 2025, which is also where its newest asset timestamp sits. And the product itself is moderation-evasion tooling, fake store listings, cloaking, fabricated ratings and AI reviews, sold with not one line of guidance about the platform risk it hands you. 3.2 out of 5: competent product, absent accountability. If you buy, buy the smallest tier, get the price and refund position in writing from the manager first, and never let the invoice run ahead of what you can afford to lose.
Related reviews
Sources and method. We read the public marketing site at pwa.market in all three locales (RU at the root, EN at /en/, UA at /ua/), the registration and login pages, and the footer and partner blocks, then requested the common legal and technical paths listed in the 404 audit above with a desktop Chrome user agent. The site is a static Tilda export (project 7475696) with no __NEXT_DATA__, no __NUXT__ and no initial-state payload, so the plan-to-price mapping was reconstructed from the Tilda T396 zero-block element coordinates. On the RU page those are blocks rec599288790 and rec599326062; the EN and UA pages carry the same two blocks under different record ids within the same project, which is how the $1,380 versus $1,318 Business discrepancy was confirmed as being the same card at the same coordinates on different locales. The nine-card count is a count of plan cards in those two blocks, seven of which carry a price. The 19 February 2025 date is the newest Last-Modified header across the 178 Tilda CDN assets the three pages reference. Cost per PWA in the pricing table is our own arithmetic on the vendor's published prices and quotas, not a vendor figure.The 60,000+ PWAs created, 10,000+ users, 80% average ROI, "don't get tagged or banned by advertising sources", "perfect for webmasters, who work with Tier-3 countries" and "unlimited installs on iOS and PWA" statements are the vendor's own marketing claims, reproduced as claims and not as findings. We could not verify any launch or last-update date for the platform itself: none is published, so we do not assert one.We did not purchase any plan, did not create an account and did not enter the dashboard, so build quality, delivery, support responsiveness, refund behaviour and the actual performance of the cloak are unverified by us. Prices, plan names and page contents are a snapshot at time of writing and the vendor can change them without notice, particularly given the site publishes no terms. PWA.Market sells no proxies or bandwidth, so there is no competitive overlap with PROXIES.SX, but its homepage cross-promotes a competing proxy vendor and we have disclosed that above rather than omitting it. Nothing here endorses cloaking, imitation store listings or fabricated ratings and reviews: those breach the terms of Google and of most advertising networks, and the exposure falls on the operator. Outbound links use rel="nofollow sponsored noopener noreferrer" and carry our partner reference.