This is the biggest WhatsApp pricing change we have ever made. Starting , HighLevel is removing the 5% processing fee on every WhatsApp message you send. Marketing, utility, service, all of it. From that day you pay Meta's published rate to the cent

> 💥 Open Meta's rate card. Open ours. The numbers will match exactly. That is the whole announcement.

Per marketing message, what you pay today and what you pay from October 1:

This is Meta's published rate card and it is exactly what you will be billed. Service is priced the same as utility in every market.

Rebilling is applied as a multiplier on cost, so your client-facing price moves with the rate. Review your multiplier under and set it where you want it ahead of October 1.

Two categories of messages that are currently free become billable from 1 October 2026. Both changes apply on a per-message basis, consistent with how template messages are already charged.

> 💚 Conversations that start from a Click to WhatsApp ad or a Facebook Page call-to-action button remain free for a full 72 hours, for every message type. Marketing, utility and service messages are all at no charge inside that window. This is the single biggest lever you have to keep your WhatsApp costs down after October 1.

the customer service window is independent of the FEP window. If the 24-hour customer service window closes, you can only send template messages, even though the FEP window is still open and those templates are still free.

route more first-touch traffic through Click to WhatsApp ads and Page CTA buttons. Every conversation that starts there gives you three days of completely free messaging, which the new service-message charge does not touch.

Per-message rates change in the following markets from 1 October 2026. Markets marked * are also moving out of their regional "Rest Of" pricing bucket, see Change 4.

*Market is being moved out of its respective "Rest Of" pricing region to be standalone on the rate card.

The nine markets with utility rate updates were given four months' notice rather than the usual one month.

In the markets marked as an increase above, Meta's increase is larger than the 5% we are removing, so your rate is higher than what you pay today. Everywhere else your rate goes down. The full card is above.

Six markets moved out of their respective "Rest Of" pricing region on 1 July 2026: Hong Kong, Singapore, Hungary, Poland, Romania and Qatar. From 1 October 2026 a further nine markets become standalone on the rate card.

Bangladesh, Iraq, Kazakhstan, Kuwait, Morocco, Nepal, Oman, Sri Lanka and Ukraine.

Until 30 September 2026 a message to one of these countries is priced at its regional "Rest Of" rate. From 1 October 2026 the country has its own line on the rate card, with its own per-message rate.

Many WhatsApp platforms add anywhere from 6% to 25% on top of Meta's rate and never tell you.

Publishing a workflow just got a lot safer. If your workflow has unresolved configuration issues, we now stop the publish, show you exactly what is broken, and help you fix it on the spot. Broken workflows no longer slip silently into production.

This sits on top of a full year of validation coverage built across triggers and actions. Users no longer guess what is misconfigured. The canvas tells them, the sidebar fixes it, and the publish gate makes sure only working workflows go live.

You can now manually retry failed subscription payments directly from the Subscription Details page for NMI, Authorize.Net, Square, and Mercado Pago.

This gives businesses another way to recover subscription revenue after automatic payment retries have been exhausted or unavailable, along with an option to update the card for the subscription and collect payment from a new payment method.

Go to and open any subscription with an '' or '' payment status.

Once automatic retries are exhausted, you can:

- https://help.leadconnectorhq.com/support/solutions/articles/155000008665-how-to-manually-retry-a-failed-subscription-payment

We will continue improving subscription payment recovery across payment providers and making failed payment management more seamless.

Community members can now record and send voice notes directly in posts and comments - a faster, more expressive way to communicate without typing long replies.

A new mic button in the post and comment composer lets members record voice notes on the spot. Record, pause, resume, or cancel before sending - all inline, without leaving the editor.

In addition to recording, members can upload audio files alongside images, videos, and other attachments. Multiple audio files can be attached to a single post.

Voice notes and audio files play directly in the feed - in both posts and comments. Recorded notes display a waveform view, while uploaded audio shows a timeline view. Playback speed can be toggled between 1x, 1.5x, and 2x, and audio autoplays to the next track for a continuous listening experience.

Voice notes give members a faster, more personal way to share updates, give feedback, or reply to discussions, without needing to type long responses or switch to a separate chat tool. Also, voice messages are a better form of expression, as sometimes the text messages are just not enough!

We’ve integrated the HighLevel App Marketplace with Ask AI, making it easier for users to discover the right apps without having to manually browse or search through the Marketplace.

Users can now simply describe what they are looking for in Ask AI—such as “Is there any IDX integration or real estate integration available?”—and Ask AI will understand the requirement and recommend relevant Marketplace apps directly within the conversation.

These controls apply to sub-account admins only; agency admins continue to have access to Marketplace discovery.

Ad Manager now brings , helping users identify and act on optimization opportunities without switching to Meta Ads Manager.

Once a Meta campaign is published, users can now:

When you enable SaaS for a sub-account and choose to attach a snapshot, we now check whether the snapshot would overwrite anything you've already customized in that sub-account. If it would, you'll see a warning before any changes are made, and you can choose to either keep your existing assets or overwrite them with the snapshot's version. If there's no conflict, the snapshot applies just like before.

Previously, attaching a snapshot could silently overwrite funnels, websites, or other assets you'd already updated in a sub-account — with no warning and no way to undo it. Now you're always in control: you'll see exactly when a conflict exists and get to decide what happens to your work.

Custom forms attached to your bookings now have complete conditional logic support across Services and Rentals. Show/hide field conditions have always applied automatically during booking — and now, post-booking actions like redirects, custom messages, and lead disqualification or confirmation page configured in the form can also be activated through a new global setting. This gives businesses full control over the booking experience from start to finish, powered entirely by their custom form.

Custom form conditions that control the confirmation page — including Redirect to URL, Display custom message, and Disqualify lead — now fully apply after a booking is submitted, when enabled in Global Settings.

When the global setting is enabled, the custom form's On Submit action also controls the confirmation page experience, replacing the calendar-level confirmation settings.

A new checkbox — "Let a custom form override this with its own rules" — in Global Settings > Form & confirmation gives you one place to activate full form control across all listings and services. A tooltip clarifies exactly what enabling it does.

If the attached custom form has no applicable settings or rules, the confirmation page automatically falls back to whatever is configured in Global Settings — so nothing breaks if a form is attached without applicable post-booking rules (such as On Submit action set to Order Confirmation).

When selecting a custom form at the individual listing or service level, a new inline banner tells you if and exactly how the form's rules will behave.

GPT Image 2.5 is now live in the AI Image Generation action. Two variants ship together, Flare and Sunburst. Open the model dropdown, pick the one that fits the job, and generate.

The default pick for most use cases. Flare produces higher quality output than GPT Image 2 at up to 50% lower latency, which makes it the right fit for creator and social content, product visuals, visual search, quick prototyping, and high volume generation.

Built for premium visual work that needs tighter control across edits. Use it for campaign creative and polished product imagery.

Reference photos carry much further than before. Subjects stay recognizable across new settings, styles, and compositions, lighting and textures look natural, and distinctive features carry through instead of getting lost. Variations stay anchored to the original source, so a series holds together rather than drifting image to image.

You can also swap a single element, such as a product, a background, or a piece of copy, while the subject, composition, and brand treatment stay intact. That turns image generation into something you can re-version and localize inside a workflow instead of rebuilding each asset from scratch.

The AI Image Generation action now accepts reference images. Attach up to five images as visual context alongside your prompt, and control the setting, the subject, and your branding directly instead of trying to describe them in words.

– Attach up to five images to a single AI Image Generation action. Each is passed to the model as visual context along with your prompt.

– Upload from your system, pick from the Media Library, or point at a URL.

– The URL field supports custom values, so a reference can come from a webhook payload, a contact field, or a previous action. One workflow, a different reference image per contact.

– Images are passed in the order you add them, so your prompt can address each one by position.

The goal is a brand lifestyle image of a model at a seaside restaurant, with the brand logo placed in the corner. No photographer, no designer, no studio.

Three reference images are uploaded, in this exact order:

The prompt then assigns a job to each one by position:

"A realistic image of a lady wearing a beautiful summer dress with light colors like blue and white, sitting in the location shown in the first reference image, looking directly into the camera with a smile. The scene captures her clearly and naturally within that setting. The logo from the third reference image is placed visibly in either the bottom left or bottom right corner of the image, fitting suitably with the overall composition."

The output places the subject from the second reference inside the setting from the first, with the logo from the third rendered cleanly in the bottom right corner.

A face, a room, or an exact logo never survives a text description. Reference images move image generation from describing to showing, which is the difference between an approximation and the asset you wanted.

Brand assets stay exact rather than approximated, so generated visuals can carry real logos and real product shots instead of near-misses.

Dynamic reference URLs make personalised visuals possible at volume. A single action can produce a different image for every contact who passes through it, driven by whatever the workflow already knows about them.