85 updates across 6 releases published during August 2026. Newest first.
Amazon, Shopify, WooCommerce, eBay, Etsy, Walmart Marketplace, TikTok Shop, Mirakl, Mercado Libre, Allegro and Kaufland now join Trendyol, Hepsiburada, n11, Çiçeksepeti and Pazarama. Orders are imported into Rocketly as draft invoices, with product, tax and warehouse mapping checked separately for each provider.
Connections now have one home where you can see their owner, health, items needing attention and service-level usage. Manage a Google Calendar connection here and preview cross-channel campaign impact in the same context.
Save a report layout you use often for yourself or your team, then open it again when needed. See alerts when a threshold is crossed and open an authorized shared report view in a print-friendly format.
For supported CRM and reporting questions, you can see a visual summary alongside the answer. Drill into the relevant record or report breakdown; answers are generated only from data and permissions available to you.
Share the way you are working in the media library. The link preserves the selected folder, filters, sorting and view preferences, so the recipient opens the same context without rebuilding the full media workspace.
Readers can search for a relevant post within a category and copy a direct link to a section of an article. Finding and sharing the exact part of a long article is faster.
You can see where your competitors stand on the keywords you track. Ranking data now comes from the search results page, not only from your own click data, and visibility in AI search is brought together with search engine visibility in one place. Content intelligence shows which page is thin on which topic.
Topics the AI could not answer now turn into a draft: an FAQ suggestion filtered by subject, stripped of personal data and free of links. You can also see which conversations a single answer affects — every chat that draws on the same piece of knowledge, in one list.
A coaching profile you share can now be protected with a password and capped to a number of views. You can also see whether the link was opened and when. On top of that, items in an improvement plan turn into real tasks with one click.
Giving one person an extra permission no longer means cloning a role. The permission is granted directly to the user, so your role list stops filling up with near-identical copies. There is also a report that finds and groups existing cloned roles — it shows you what exists first, and applying the change is a separate step.
Granting critical permissions and assigning sensitive roles can no longer be completed by one person; a second authorised user must approve. Whoever declines must give a reason, so the rationale stays on the record. Pending permission changes follow the same path.
Call recordings that need reviewing collect in their own queue, so the quality team picks up the next one instead of scanning the whole list.
Customer satisfaction scores were already aggregated on the dashboard, but you could not act on them. Now you can ask for calls at or below a score — one filter to find the conversations dragging the average down.
Live agent status and queue wait time on one screen, in type large enough to read from across the room. The quality leaderboard sits on the same display. Each block runs inside its own error boundary, so if one cannot load its data the rest keep working, and the clock updates in step.
The score from the analysis model is now compared with the score a reviewer gave. Three numbers are shown together: the direction of the gap (do reviewers score more generously or more harshly than the model), the typical size of the gap, and the share of reviews where the two scores are within 10 points. If the average gap is near zero but the typical gap is large, the model is right on average and wrong case by case — you can only tell those apart by looking at both. How many reviews could actually be compared is stated separately.
Define a review form of weighted criteria and score calls straight from the call detail. A criterion you leave blank is not counted as zero; the score is calculated only from what you filled in, and the screen shows how many criteria were counted. Agents can dispute their own review — a dispute never erases the score, it is recorded alongside it. Two reviewers can score the same call, so you can measure consistency between reviewers.
Upload a target list, set the calling hours and the retry rule, and let agents work through it. Hours apply in your organisation's time zone, so customers are never called at the wrong hour. The do-not-call list is re-checked at dial time — including numbers added after the list was uploaded. You are told how many uploaded numbers are suppressed rather than having them silently dropped. Preview and power dialing are supported because they never abandon a call; predictive dialing stays with the phone system.
Did the person who filled in your ad form actually reach the CRM? You no longer have to guess. The new centre at the top of the Lead Forms tab measures every step of delivery: open issues, per-form delivery performance, provider readiness and the queue of records that could not be delivered. It covers Meta, TikTok, LinkedIn and Snapchat, and the panel never shows personal data.
You can pull back the historical records lost when a connection dropped or a permission lapsed. Before anything runs you get a preview of how many records will be scanned, the limits are written on screen, and the job can be stopped while it is running. Available for Meta and LinkedIn forms.
When your ad reports do not show the day you expect, one dialog tells you why: is the connection up, what is the latest day of data and how old is it, when did the background worker last finish successfully, when was account metadata refreshed, and is the access token still valid. The token value is never printed on screen — only its state.
For TikTok, LinkedIn and Snapchat forms, a provider's display name or a free-form UTM label was never a campaign id. Now you map the form — and optionally a specific UTM campaign — to the real campaign id yourself, and the lead → deal → revenue report reads from that audited mapping.
The list of pages losing traffic can now be reviewed by team members without task-writing permission, and every row jumps straight to that URL's lifecycle. When an internal link draft is being prepared you can see which row is working, audit results download in three formats, and rank tracking can measure the same keywords on another device or country and show you the gap.
Guardrail thresholds now carry the currency of the ad account they belong to: in an organisation whose base is one currency and whose account is another, "500" meant two different real thresholds. When a button is disabled the reason is written on screen, and an empty list now tells you what to do next.
Agents switch their status from the softphone header in one click. In the supervisor panel the idle team is no longer one pile: available, on break and wrapping up after a call each get their own heading, with how long they have been there next to them. Nobody can put someone else on break — only agents change their own status.
Convert a recording from the voicemail inbox and the title fills in with the number, the description with a transcript preview; the task opens at high priority and is due four hours later. If the number matches a lead or a customer, the task is linked to that record. The same voicemail cannot become a second task — the existing one opens instead.
You set how many seconds and what share of calls should be answered within them; when the measure slips, users with supervisor permission get a notification. An abandoned call always counts as a breach, no alert is raised while the sample is too small, and you decide how often the same alert may repeat.
Search the text of analysed calls: a refund, the word "expensive", a competitor's name, a campaign name. Results show only the passage the match sits in, with the term highlighted, rather than the whole transcript, and each row jumps straight to that call. Turkish dotted and dotless i are matched correctly.
Changing a call's outcome no longer means opening the record — you pick it from the list. You decide which boxes stay in the indicator strip at the top, and that choice is remembered on the device. Expanding an agent's row in the leaderboard draws their quality curve over 7, 30 or 90 days.
One card sits above the inbox list: the conversation that genuinely needs you right now. The order comes from delivery problems, breached response times and conversations waiting for approval, and the card says how long it has been waiting. You can assign an owner or open a follow-up task without leaving it.
The "All history" panel in a contact's detail brings that person's WhatsApp, e-mail and other channel messages, internal team notes and CRM events into a single timeline. Events such as lead created, converted to customer or record merged appear as blue rows, and team notes are marked in yellow so they are never mistaken for something the customer saw.
When a reply is written together, nobody's work is lost: if two people edit the same draft the second sees "the team draft has changed, load the current version" and their colleague's text is never overwritten. Team notes that never reach the customer can be left under the draft. Admins can route by channel, language, product or segment.
On e-mail channels that send through Resend you can see what actually happened to a message: the provider accepted it, reported a delay, delivered it, could not deliver it, or failed permanently. "Accepted" and "delivered" are deliberately kept apart — handing a message to the provider is not the same as handing it to the recipient.
At the bottom of softphone settings, a table lists row by row what AloTech and Bulutfon provide: agent attribution, queue breakdown, voicemail box, queue wait time, recording and analysis, live call state, transfer legs. Every closed cell carries its reason — either the provider never sends that data, or it simply has not been built yet.
Automated follow-up sequences no longer wake people at night or bury them in messages. If a message comes due during quiet hours it is not skipped — it is sent when the window ends. You can set a per-person daily cap, and that cap counts every automation message reaching them, not just this one sequence.
Search now looks at the body as well as the name and subject — remember a sentence and you will find the template. The list can be ordered by default first, by name, by most used or by last updated. Before you send a saved template to anyone, you can send yourself a test copy.
At the top of the tasks page sit at most three cards: the work to do now. They are labelled overdue, due today, high priority and next up. Debt older than thirty days no longer clogs the queue — one representative card is shown and the rest becomes a count in the heading.
A lead who called five times and hung up angry every time no longer scores the same as one who spoke warmly. Scoring takes into account the customer's sentiment in analysed calls and the objections they raised. Organisations that do not use call analysis are unaffected — nothing is deducted, the signal is simply not counted.
The copy button on an opportunity card gives you a link that opens the same card for whoever receives it. For cards that are done or dismissed the view is carried in the link too, otherwise the other person would land on an empty list. And when a run could not be measured, the verdict card now takes you to the measurement history with your period and engine filters intact.
When a total comes from more than one currency, or contains lines that were never converted, a warning mark appears next to the figure. The number is not hidden — you are simply told it cannot be labelled with a single currency symbol. It shows up in the collections list, expense payment reports, statement risk, exchange differences and cash flow projections.
The FAQ page went from 6 sections and 18 answers to 13 sections and 100 answers, in Turkish, English and Russian. Results filter as you type, the 13 topic shortcuts at the top jump straight to a section, and a live counter tells you how many answers are showing.
All 25 product pages were audited against the product's own code: every feature card describes behaviour that genuinely exists, and claims with nothing behind them were removed. Card count went from 348 to 498 per language. The 69 sector guides gained a section navigation bar, and the catalogue of 27 integrations gained search and category counts.
The integration centre no longer waits for everything to load — cards fill in as they arrive. Rarely used dialogs are downloaded only when needed, and document template bodies are fetched only when you open one to edit. Discovery lists moved to cursor-based paging, and the Kanban view no longer fetches a list you cannot see.
A screen on your phone that answers one question: who do I call now. The queue arrives in the order the server decides, and each row carries a coloured mark for the overdue, today, new and upcoming buckets, the person's name, their company and a callback time if there is one. Calls go through the app's own dialling flow, so conversations made from a phone are recorded too.
When the contracts screen is empty it now offers two paths instead of a single “New Contract” button: start by picking an accepted quote, or create the contract manually. Picking a quote lets the contract inherit its details, and nothing is created until you confirm. On contract analytics, the filter button now carries a badge showing how many filters are active.
Follow your competitors' ads, keywords and search visibility in one place. Build your brand list, run a scan and see who is winning where, backed by your own data.
Scans now build up into a timeline, so you can put two moments side by side and see exactly what changed. Saved comparison sets, tags and filters, CSV/PDF export, scheduled automatic scans and an opt-in weekly email digest are all in. Turn a keyword gap into an AI content brief in one click, link a finding to a deal, and compare yourself with your sector without exposing anyone's identity. Multiple brands and multiple branches are supported.
Score banners, videos and landing pages the moment you upload them, and see where you stand against the sector average line. There is a dedicated video tab, side-by-side creative comparison and landing page review. The price is shown on screen before you spend anything, and your credit is reserved when the job starts — no surprises.
The menu was rebuilt around four questions — where do I stand, who is ahead, what should I do, what does the evidence say — with everything else gathered under Advanced. A Simple / Advanced view mode lets you strip the screen back to what you need. Every number on the dashboard is a door into its own detail, and “not measured” no longer looks like “zero” in the tables.
Accounting now shows your CRM customers and companies as current-account cards without copying them into a second table. One card carries the roles, the document timeline and the movements together, and the column manager lets you shape the list your way. The card appears only when the accounting module is on and the person's CRM data scope allows it.
The cash ledger, reconciliation, statement preview, cheque form and the invoice and recurring invoice/expense forms all moved into a wider workspace, and cash movements can now open full screen. Duplicate an invoice design safely, and see the outcome of a statement delivery right on screen.
The customer list surfaces the next action by priority, response risk on leads turns into a clickable piece of work, and companies that need intervention gather in their own focus area. You can look at the list and answer “who should I touch right now”.
Save your own quick views on the customer and lead lists and switch between them in one click. If you like a view your team shared, you can copy it to yourself and customise it without touching the original.
The calendar flow was simplified, you can start a follow-up straight from an appointment's detail, and reminders now have a quick-start setup. On import, a preflight screen shows what will happen before the file is processed, and calendar export and empty-metric states are clearer too.
Notes are part of global search, with the visibility rule written into the query itself, so a note you are not allowed to see never appears in the results. Tab badge counts show the real number of matches, and picking a tab narrows the search to that area only. A matching index took search from 445 ms down to 0.06 ms.
The contract PDF was produced in Turkish only and printed status fields as raw codes. It now follows the customer's language and shows statuses as readable labels. Documents for existing customers who never picked a language are left unchanged.
Campaign results are now filled from Meta's own delivery receipts: how many messages were delivered, how many were read, how many failed — measured, not estimated.
You can export the coaching comparison as a CSV that respects your permission boundaries. On the Voice of Customer side, sentiment breakdowns are now clickable: pick a sentiment and drill into the conversations behind it.
List rows, detail actions, filter bars and mobile headers moved to a compact density system. The same rule now applies across customers, leads, tasks, quotes, appointments, accounting and construction: actions get tighter while accessible touch targets stay intact.
With the app open in several tabs, each tab no longer opens its own connection: they share a single live stream, and missed events are replayed after a drop. A change you make in one tab shows up in the others straight away.
The page you are reading is generated from a single source: every update carries its own date and version, and releases line up as a timeline. Narrow it down with area and type filters, and find anything in the history with the search box — in all three languages.
3D model loading is now bounded and the scene stream is isolated from other work, so large models no longer hold up the rest of the page. Scenes opened from a shared link use the same protection, and the workspace controls were unified into a single layout.
Review your AI agents' real conversations from one screen, rate them and mark them for training. Close the loop with review progress, pending review and average rating; move fast even at high volume with search, filters and saved views.
Present your project's real 3D model, floor plans, 360° panorama and in-unit photo gallery through a public link. With presentation/kiosk mode, a clickable site plan and spam protection on the showcase form — built for construction and real-estate sales.
A new module that measures your brand's visibility in generative-AI answers. Define the set of prompts to track, watch each engine's answer and the sources it cites; whether you were mentioned, with what wording and from where — all on one dashboard.
Incoming conversations route to the most suitable AI agent, and a handoff map finally answers “why did it go to this agent?”. With load balancing, threshold simulation and handoff to a human when needed — your rules, a traceable decision.
Agent performance heatmaps, persistent improvement plans, manager coaching notes and you-vs-team benchmarking. Plus an opt-in weekly coaching digest email, and drill straight down to the real conversations behind every metric.
A business-hours- and priority-aware SLA engine with breach escalation, auto-assign on incoming messages, recurring scheduled messages, per-agent signatures, and template rendering in scheduled messages with “skip if the customer already replied”. Apply AI tag suggestions with one click, too.
Service level and ASA, shift planning from the intraday volume distribution, a sortable and exportable supervisor report, a two-axis disposition taxonomy and call-back from voicemail. Plus a persistent playback bar with scrubbing, speed and mute.
SLA tracking, ticket merging, canned responses, inbound email that turns straight into a ticket, and CSAT satisfaction surveys. Environment info is captured automatically and new replies are badged — fully working on mobile too.
Kick off outreach automatically when a signal you track fires (opt-in, off by default). With a ProspectMap heatmap view, prospect notes and tags, a “My Templates” outreach library, and a churn alert when a tracked prospect closes.
Bulk pause, resume and update budgets for TikTok, LinkedIn and Google Ads campaigns without leaving Rocketly. LinkedIn and Yandex now run in embedded mode, and the marketing hub is tidier with tabs.
See your sites' Core Web Vitals scorecard on one screen; catch performance regressions early with 40-week trend charts from CrUX history and regression signals. With automatic weekly refresh in the background.
Inspect incoming webhook payloads line by line, search by source and see each request's status. Plus an IP allowlist for security, the missing tracking-code snippet on the source card, and “call.*” events now wired into the pipeline.
Review who can access what from one screen, see your team tree and import invites in bulk via CSV. Ready-made persona roles tied to add-ons get new teammates started with the right permissions in minutes.
Enter your billing details (tax ID and address) and see the VAT note clearly; forecast where your AI-credit usage is heading before it runs out. With contextual upgrade nudges and cancel-reason capture.
The quote PDF — the one document your customer actually reads — is no longer Turkish-only; it's generated in the recipient's language. Your quotes reach international customers in the right language.
Site needs now flow end to end: from requisition items to quote collection (RFQ: save, compare, pick the winner) to a purchase order with a supplier picker. A new Cost & Profit panel shows project profitability, margin, cost per m² and BOQ budget usage on one screen, and every pending approval for a project is gathered in a single “Approvals” tab.
Supervisors can whisper-coach an agent on a live call, barge into a three-way conversation, or drop back to silent monitoring. During sensitive moments like payments, the agent can pause recording temporarily and resume it with a persistent badge — for PCI-DSS compliance.
Version your form changes and roll back to an earlier version whenever you need. Track submissions through a review flow with analytics, and see how each form fills in and converts.
With the new Rocketly browser extension (Manifest V3), select a phone number on any web page and call it in a single click. The call goes through Rocketly, applying your session’s Do-Not-Call check and softphone flow.
When one of your saved views crosses a threshold you set (e.g. the number of conversations awaiting a reply), its assigned owner gets alerted — catch problems before they grow. Conversations are handed over between agents in a controlled way.
Long lists like the customer timeline, contact directory, call history and call queue now render only what’s on screen through virtualization; scrolling stays fast and smooth even with hundreds of records.
On the call-center side, a Bulutfon integration (with connect, configure and call-history screens); on the messaging side, two new SMS providers — Posta Güvercini and Vatan SMS.
TikTok Ads, with campaign, spend and demographic data, in its own tab in the Marketing Center. TikTok now joins Meta, Google and Yandex.
Track supplier performance and operational queues from a single workspace, and see at a glance how each supplier performs on delivery and cost.
No updates match these filters.
85 results
Sports & Wellness
Personal Care
Real Estate
Automotive
E-Commerce & Retail
Food & Beverage
Tourism & Events
Finance & Insurance
Professional Services
Technology
Media & Advertising
Home Services
Discover the CRM built for your industry.