Skip to content

Records

Overview

The Conversations page (also called “Records”) is where all your uploaded calls and chats live. From here you can upload new recordings, search and filter existing ones, view detailed analyses, and export data.

Records List

Owner Admin Supervisor Rep

The page has two views, switched by the List and Table buttons in the page header. List (the default) shows the records as rows with the open record beside them. Table is a data grid with a checkbox on every row (see Bulk Actions below) and these columns:

ColumnDescription
QAMean of the six conversation metrics — weighted, if the project sets its own criteria weights (0–100, one decimal), color-coded — the same score the dashboard averages. A record whose analysis rates none of the metrics has no score
Sale prob.Probability of a sale from the latest analysis (calls only)
SentimentPositive, Neutral or Negative, derived from the quality score
TypeAn icon: a chat, or a call marked incoming, outgoing or direction unknown
RepName of the rep; when there is no name, the extension that handled the call
ParticipantName of the contact, with the phone number under it
CRMLink to the CRM entity the record is linked to
DurationLength of the conversation; for a chat with no duration, the number of messages (e.g. “18 msg”)
CategoryMissed, Short, or the category assigned by analysis (the same list as the Category filter below)
#Position of the call in its call series, e.g. “#2/5”
ProjectWhich data source/project the record belongs to
★Rating of the analysis accuracy (1–5 stars), if someone has rated it; a dash otherwise
Conversation timeWhen the conversation happened, as reported by the source. A dash means the source reported no conversation time (file upload, some API submissions)
ReceivedWhen Aelo received the record. Hidden by default — enable it from the column gear. A gap between it and the conversation time shows delivery delay on the source side, not analysis time
A: Greeting … F: ClarityThe six component scores: greeting, needs, objections, closing, politeness, clarity. Hidden by default — enable them from the column gear

Sorting: Click a column header to sort ascending/descending. QA, Sale prob., Sentiment, Rep, Participant, Category, Project, ★, Conversation time, Received and the component scores can be sorted; Type, CRM, Duration and # cannot. Sorting orders the whole list, not only the records already loaded: rows without a value (for example, an unscored record) go last in either direction. A third click removes the sorting. While a search is active, the list keeps the search’s own order — newest first for a text search, closest match first for a semantic one — and column headers do not sort. A record opened from the table appears in the List view in the same order, so previous/next follow the sorting.

Loading more: The list loads 50 records at a time and fetches the next 50 as you scroll down; there are no page numbers and no page-size setting.

Record count: Above 5,000 records the list shows “5,000+” instead of the exact number. All records stay available in the list.

Status in the list: There is no separate Status column. While a record is being processed, its QA cell shows an icon of the current stage instead of the score (pulsing while work is running), a static icon for a record waiting for a start or a top-up, and a warning triangle for a failed record; the score appears when processing ends. To narrow the list by status, use the Status filter.

Records table

The column gear: tick the columns to show, including Received and the six component scores

Records in processing: an icon of the current stage in the QA cell instead of the score

Failed records: a warning icon in the QA cell

Record Statuses

Records move through a pipeline after upload. The names below are the ones the product shows — for example, in the hint on the stage icon in the list:

StatusMeaning
PendingRecord received, waiting to be processed
Awaiting MediaA call has arrived, but its recording has not been fetched from the source yet
Waiting for ChatThe chat is not closed in the CRM yet; Aelo waits for it to finish
Awaiting startThe record waits for you to start it (manual mode, or Auto-Analysis is off in its project); no minutes are spent until then
Awaiting top-upThe organization is out of funds, so analysis did not run; top up the balance, then start the record
TranscribingAudio is being converted to text
TranscribedThe transcript is ready, analysis is next
Personal data removedPersonal data has been removed from the text before analysis
AnalyzingAI is evaluating the transcript
AnalyzedThe analysis is ready, delivery is next
DeliveringResults being sent to CRM or other channels
DeliveredResults reached all the channels
Partially deliveredResults reached some of the channels but not all
CompletedAll processing finished successfully
No FileThere is nothing to transcribe or analyze — for example a missed call or a call with no recording; the record is not processed further
Processing objectedThe participant objected to processing; the record is not analyzed
FailedAn error occurred during processing
Transcription FailedThe record stopped at transcription
Analysis FailedThe record stopped at analysis
Delivery FailedProcessing finished, but an error occurred while delivering the result — it may not have reached the CRM, email or webhook

How to start records that are waiting for a start, and how manual mode differs from automatic, is described in Automatic and Manual Analysis.

A failed record shows a notice with the reason above its tabs; the Retry processing button in that notice processes the record again (see Re-analyze & Retry below for who can use it).

The failure notice above the record's tabs: what went wrong and what happens next

Uploading Records

Owner Admin Rep

  1. Click the Upload record button in the top-right corner of the page. A panel opens on the right.
  2. In the panel:
    • Drag and drop files onto the drop zone, or click to browse.
    • Select the Project the record belongs to (required).
    • Choose the Channel Type: Phone Call or Chat/Messenger.
    • For calls, the Enable Speaker Diarization switch separates the speakers in the audio; it is off by default and may increase processing time and cost.
    • Optionally fill in Participant (phone number or identifier), Rep (shown only to roles that can see the list of users) and Activity Name (CRM activity name or a description of the record).
  3. Click Upload to begin processing.

Supported formats (calls): MP3, WAV, M4A, OGG, WEBM, FLAC

Chats: a .json file.

File limits: An audio file can be up to 25 MB and a chat file up to 10 MB. The drop zone shows the formats and the limit for the selected channel type.

Upload record panel

Batch upload: You can select multiple files at once. Each file becomes a separate record in the selected project.

Project Status

Owner Admin

Each project has a status, set from the project’s own settings page (Overview tab): Active, Paused, or Archived.

The project status list: Active, Paused, Archived

StatusAccepts new recordsAppears in filters & analytics
ActiveYes — from the CRM integration, the public API, and file uploadYes
PausedNo — every source is blocked (CRM integration, public API, file upload)Yes
ArchivedNoDashboards: No · Records: Yes

Changing a project’s status never touches the records already in it — everything uploaded before the change stays in the Records list and in your analytics. The one exception is a conversation still waiting for its recording or for its chat to close when the project is paused or archived: it is not transcribed or analyzed and uses no minutes, and the record shows the reason The project was paused or archived. Resuming the project does not process it. An archived project drops out of the dashboard’s project filter, but keeps showing up in this page’s own Project filter, so you can still narrow the records list down to its history.

The filter bar above the records list holds the search box with its Text / Semantic switch, dropdowns for project, channel, direction, category, rep and CRM entity type, the analysis-accuracy stars, a Missed button and the Filters button (controls that do not fit are hidden). The Filters panel, which opens on the right, holds all the filters below, including those the bar does not show:

FilterTypeDescription
ParticipantTextSearch by name, phone, or email
ProjectDropdownFilter by project
ChannelDropdownAll channels, Calls Only or Chats Only
Date From / Date ToDate pickersStart and end date
QA rangeNumber (0–100) — minimum and maximumFilter by the quality score (the list’s QA column)
DirectionDropdownIncoming, Outgoing, or Direction unknown (the source gave no direction, or one Aelo does not recognize). Only calls have a direction: picking any value takes chats out of the list
CategoryDropdownMissed, Short, or a category assigned by analysis: Product Inquiry, Purchase, Scheduling, Support, Billing, Onboarding, Complaint, Cancellation, Follow-up, Other. In the web app the list hides missed calls by default: picking Missed shows them, and so does a Status group; Direction alone does not. Missed returns missed calls of both directions, and each record’s label says which kind it is — “Missed” (incoming, nobody picked up), “Client No-Answer” (outgoing, the client did not answer), “Missed, direction unknown”
RepDropdownFilter by assigned rep
Analysis accuracyStars (1–5) with ≥ / = / ≤Records whose analysis accuracy rating (the ★ column) is at least, exactly or at most the stars you pick
SentimentDropdownPositive, Neutral or Negative (derived from the quality score)
StatusDropdownProcessing outcome — the same groups as the funnel in the Received and processed card on the dashboard’s Conversations tab: “Analyzed”, “Missed calls”, “Waiting to be started”, “Too short to analyze”, “No conversation recorded”, “In progress”, “Processing error”, “Other statuses”. Picking a group also shows missed calls, because the funnel counts them — in a group of their own, “Missed calls”. “Delivery Failed” finds calls whose analysis is done but delivery to the CRM failed — the funnel counts them as analyzed
Entity TypeDropdownRecords linked to a CRM Deal, Lead, Contact or Company
ActivityTextFilter by CRM activity name
CriteriaUp to three conditionsA criterion and its condition (e.g. triggered / not triggered); the conditions combine with AND. The section appears only when your organization has criteria
Hide short callsSwitch (off by default)Hides the calls Aelo marks as short — below the analysis threshold, which it does not analyze, including Bitrix24 calls under 15 seconds — and chats too short to analyze. The list and its total count only the calls that are left. Off by default, so short calls stay visible; it does not apply while a funnel Status group (such as Too short to analyze) or the Short category is picked — the funnel files short calls under several groups, and the list has to match its number. “Delivery Failed” does not lift it. Found under Filters, and turned off by Clear

Text Search: with semantic search off, the search bar finds records by participant and rep name, phone numbers, line number, external ID, CRM activity name, and the beginning of the transcript. Letter case does not matter for Latin or Cyrillic letters — “oleg” finds “Oleg”, “олег” finds “Олег”. The exceptions are one-character queries and a rare fallback mode, most often for two-character queries: in both, only the basic Latin letters A–Z ignore case. If a short query with other letters (Cyrillic, accented) misses a record, type at least three characters. When a query matches a very large number of records, a note above the results says “At least N matches — the count is not final”: the list keeps loading as you scroll, but the total shown is a lower bound. Refine the query or narrow the period to get an exact count. In the web app, when many other records match first, search keeps going through the matches — “Searching further through the matches…” — and after a few steps offers a Search further button to continue. If part of the results came from the fallback mode, a note above the list says some records may be missing: try again in a minute, or type at least three characters. The same search works on the Records tab of the Bitrix24 app, with the same notes and the same Search further button: results load 20 at a time, and Load More adds the next ones.

Semantic Search: Use the search bar for natural language queries across transcripts (e.g., “customer complained about delivery”). The filters you have selected — status, project, period, rep and the rest — apply to the results too, and the number next to the page title counts what was found: up to the 10 closest matches. If the filters set some of the closest matches aside, a note above the results says there may be more — make the query more specific to find them.

Remembered filters: the filters you set are kept while you work in the app. Go to another section — the dashboard, say — and come back through the menu, and the list opens with the same filters. Project, rep, channel and the like show as chips above the list; status, period, participant, activity and quality are counted on the Filters button. They are kept for your account and organization separately, so another organization never inherits the project or rep you picked in this one, and are forgotten when you close the browser tab. Clear returns the list to its defaults and forgets them. The search text and Criteria conditions are not remembered.

Links with a filter: a link from a dashboard or a report (“calls of this category”, “calls of this rep”) opens the list with exactly the filter of that link — it replaces the remembered filters and never mixes with them. A link from the Script Steps report adds a “Step not completed: …” chip above the list; dismiss it with its cross — a step cannot be picked in the filter panel.

Filters panel

The bar above the list: search, the Text switch, project, Filters and Missed

Record Detail Panel

Click any record in the list to open it beside the list: the list stays on the left and the record’s panel opens on the right (a master-detail layout). In the Table view, a click on a row switches to List and opens the record. In a window narrower than 560 px the two do not fit side by side, so the record opens on its own with a Back button above it.

A record opened beside the list: the header, the tabs and the transcript

At the top is a header summarising the record: quality score, conversation type and duration, line number, category, deal outcome, CRM entity link, position in the call series, project, conversation time and the time Aelo received it, sale probability (Sale N%).

Below it the panel is split into tabs: Transcript, Notes and — only in a narrow panel (under 900 px) — Score. In a wide panel the analysis is always visible to the right of the transcript, so it needs no tab of its own.

At the bottom of the panel is an action bar, from left to right:

  • Task in B24 — in an organization connected to Bitrix24: copies the task text for this record to the clipboard and opens the new-task form in Bitrix24;
  • Create incident — every role except Rep: for an analyzed record opens the incident form (title, description, severity), for a record with no analysis shows an error; incidents are collected on the Incidents tab of the Alerts page;
  • Ask the assistant — see “AI assistant” below;
  • Note — see “Notes on the text” below;
  • Re-analyze — it reads Analyze while the record has no analysis yet; it appears only for a record that has finished processing or failed, and a Rep does not get it;
  • Export — saves the analysis as a PDF report.

When the panel is under 1010 px wide, the buttons sit in two rows.

The action bar at the bottom of the record panel

AI assistant

On the Pro plan and above, the action bar at the bottom of the record panel has an Ask the assistant button (in the same bar as Create incident and Note, when you have them), and the bottom of the app sidebar has an assistant icon (tooltip Assistant) that does the same while a record is open. The AI assistant menu item opens a separate page for questions about all your conversations (see below); the Ask about the deal button in the record header is covered under “Asking about a deal…” below. Ask the assistant opens a side panel where you ask questions about this record — pick one of six suggested prompts or type your own (up to 2,000 characters). The answer is grounded in the record’s transcript and analysis: a thesis carries the quote it rests on whenever the assistant found one in this record, and a quote it matched to one place in the transcript shows a ▶ mm:ss timecode — clicking it plays that moment (for chats, it scrolls the transcript there). A quote it could not match is marked “Not confirmed in the transcript”, and one that turns up in more than one place is marked “Found in several places”; neither carries a timecode, because there is no single moment to play. A thesis the assistant could not back with this record stands without a quote. The panel always names the record it is answering about, right under its title. You can keep browsing with it open — it is not a modal window — and it moves to whichever record you open next, together with that record’s own conversation and its own unsent draft. While an answer is still being prepared it stays where it is, so nothing you have paid for is lost; it then says the record is “not on screen” and offers to move over when you want to. On a screen without a record — a dashboard, say — it keeps the last one and says so rather than going blank. Each organization has a daily question limit; the panel shows how many questions are left today. Conversations stay in the panel for later reading and are deleted when the record is deleted — usually together with it, otherwise by a routine cleanup.

The assistant panel of a record: the scope and six suggested prompts

On lower plans. The AI assistant menu item is shown with a lock and the plan name, and in Bitrix24 the Assistant button is shown with a lock. Either one opens a screen with a short preview of what the assistant gives and a See plans button, or a note to ask the account owner (for roles that cannot open billing, and always in Bitrix24, which has no plans link).

Asking about a deal, lead, contact or company. When the open record is linked to a deal, lead, contact or company in your CRM, the record header shows Ask about the deal (or the lead, contact or company), and the panel gets a scope switcher: This record or This deal. A question about the deal is answered from all of its calls and chats that you are allowed to see — conversations you cannot open are left out — and a quote from another call or chat is marked as such and opens that conversation when clicked. For a long history the assistant reads the conversations most relevant to your question in full and the rest as a one-line summary each, and says so when it could not read everything. A conversation whose text could not be reliably cleared of personal data is not read in full: the answer says how many conversations of the history were left out this way. Conversations about a deal are kept per deal, the same way as for a record. A question and its answer are deleted when any record the answer was built from is deleted — usually together with it, otherwise by a routine cleanup — and in any case at the end of your organization’s retention period. The daily question limit is shared by both scopes.

A record linked to a deal: the scope switcher This record or This deal

Asking about all your conversations. The AI assistant item in the app menu opens a page for questions about every call and chat you are allowed to see — conversations you cannot open are never used. In the web app the panel’s Whole base button and the Assistant button on a screen without a record open this page too; a question you typed in the panel stays there, with the record it was typed for. Inside Bitrix24 there is no such page: there Whole base and the Assistant button ask about the whole base right in the panel, without filters. Narrow a question by period, project, rep, type (calls or chats), category or customer; the customer field suggests names as they are stored in your records, and a name is picked from the suggestions. Only customers with conversations under the other filters are suggested; change another filter after picking, and the customer is removed with a note — the name stays in the field, and the suggestions offer it again if it still fits. If nothing matches all the filters together, the answer says so and suggests removing one. The assistant finds the conversations that match your question by meaning, reads up to 12 of the closest in full and answers from them, with quotes that open the conversation they come from; the answer says how many records it searched and which filters it was asked under. If nothing matches, it says so without asking the model. Such an answer takes longer than one about a single record — it can take a few minutes — and is prepared in the background: you can leave the page and come back to it. While it is on its way the filters and the list of conversations stay closed. Changing the filters starts a new conversation, so an answer never mixes history asked under other filters; the page lists your latest five whole-base conversations. Only conversations that have been analysed and are already in the search index are used — one still waiting for analysis, or whose analysis failed, is left out. A record whose text could not be reliably cleared of personal data is not used either: the answer says how many of the records it went through were left out, and if every record found is refused this way (and none is still being analysed), the question ends with an error that repeating it will not fix. If some of the found records are refused this way and the rest are still without analysis or text (or have been deleted since the search), the question also ends without an answer: the refused ones are never read, and once the rest are ready you can ask again. When you see only part of the base — as a rep or a team lead — or filter by customer or category, the assistant searches the 5,000 most recent analysed conversations that match. The daily question limit is shared with the other scopes, and a question and its answer are deleted when any record the answer was built from is deleted, and in any case at the end of your organization’s retention period.

The AI assistant page: past conversations on the left, filters and suggested questions on the right

Switching the assistant off. The organization owner can switch the assistant off, and back on, in Settings → Organization → AI assistant. While it is off, nobody in the organization sees the Assistant button or the AI assistant menu item, and it answers no questions; the change applies to the organization right away (tabs already open pick it up on their next refresh) and can be undone at any time. An administrator sees the switch but cannot change it. If Aelo has switched the assistant off for the organization at the platform level (for example under the contract or compliance requirements), the switch is locked and says so, and turning the assistant on again is a request to support. On plans below Pro the switch is not shown.

Audio Player (Calls)

The player sits at the top of the Transcript tab.

  • Waveform — a single colour; the part already played is filled in. For a recording longer than 5 minutes, or of unknown length, no waveform is drawn — a flat bar stands in for it, and the controls work the same.
  • Controls — play and pause, skip 10 seconds back and forward, speed 0.5x, 0.75x, 1x, 1.25x, 1.5x or 2x, and volume.
  • Who spoke how much — when the transcript is labelled by speaker, a bar under the expanded player shows each participant’s share of speaking time in per cent, with a legend of names (or “Rep” and “Client”).
  • Compact view — in a window under 800 px tall the player starts collapsed; the button under it expands and collapses it. The collapsed player shows a thin progress line instead of the waveform, and has no volume slider (mute only) and no “who spoke how much” bar.
  • Jumping from the transcript — clicking a line moves playback to it and starts playing.
  • Key-moment clip — while the Transcript tab is open, clicking a time in the “Errors & Key Moments” card plays 20 seconds from that moment and stops; the Continue recording button plays on.

Transcript Tab

  • Full conversation transcript with speaker labels; each line shows the time it starts (and the recognition confidence in per cent, when there is one)
  • In a call, clicking a line, or the Play from this line button that appears on hover, jumps to that moment and starts playback; the line playing now is highlighted, and the transcript scrolls along with it
  • Quote highlight: when the record was opened from a quote (for example, from a dashboard card), the line with that quote is highlighted and scrolled into view
  • Note highlight: passages with notes are highlighted in their severity colour — green (low), amber (medium) or red (high); hovering shows the note’s text
  • A chat waiting for a start already shows its conversation: the text arrives with the chat, and only the analysis waits

Analysis

In a wide panel the analysis occupies the right column next to the transcript; in a narrow one it opens on the Score tab. It shows:

  • Quality score with breakdown by category
  • Sale probability
  • Sentiment indicators
  • Rep errors and key moments, topics
  • Compliance flags
  • Strengths and recommendations (see Quality Analysis for details)

Errors & Key Moments. This card in the Insights group puts in one list the rep errors the analysis found (tagged Error) and the key moments of the conversation (tagged Key). A key moment says what happened and, when the analysis gives them, who spoke ([Rep], [Client] or [Role unknown]) and the words verbatim; its time is a button that plays 20 seconds of the recording from that moment (while the Transcript tab is open). A missing time, or one beyond the end of the recording, is not shown; with no player the button does nothing.

Inside the analysis there is a group filter: All, Score, Script rules, Criteria, Speech, Insights. A group’s button appears only when that group has data, and the filter itself only when at least two groups do.

Built-in criteria in the analysis. A built-in criterion is shown in the Criteria group (the Criteria Results card) only when it has an answer for this processing of the conversation: Yes or No, and a Triggered badge when it fired. Then the matching older row is not shown — the recording disclaimer, TCPA, data privacy and the mandatory-phrase checks in Compliance, and the rep’s profanity in Speech — so one check never appears twice. If the criterion has no answer for this processing (an older conversation, or one the criteria step did not evaluate), the older row stays as before when the conversation was analyzed before the change; a newer analysis has no older row. A built-in that Aelo has paused until it is re-measured keeps the older row, if there is one, with the note “Paused by Aelo until re-measured”. A built-in you paused, with no answer, leaves no row anywhere. The Compliance card disappears from the analysis when all its rows have moved to built-in criteria (a CRM completeness row keeps it); the Speech card disappears when profanity was its only row. The PDF report follows the same rule. Older rows exist only for conversations analyzed before the change; new analyses have none.

Script rules group

If the project has a linked sales script and adherence has been evaluated, the analysis gains a Script rules group holding the “Script Adherence” card and its score. It is a group inside the analysis, not a tab of the panel. See Scripts for details.

Notes Tab

Notes on the record; the tab label carries their count. See the Notes on the text section below for how to add them.

Record details and processing status

Project, participant, channel, duration and the CRM entity link live in the panel header — there is no separate Metadata tab. The processing stages are listed above under “Record Statuses”; the panel has no pipeline-status tab either, but when processing has been halted the panel shows a notice with the reason.

Notes on the text

Owner Admin Supervisor Rep

Add notes to specific parts of the transcript:

  1. Select a passage of the transcript.
  2. Right-click the selection and choose Add note.
  3. In the Add note window (it shows the selected text), pick the Type — “Poor Objection Handling”, “Missing Discovery Questions”, “Tone Issues” or “Other” (the default) — and the Severity — “Low”, “Medium” (the default) or “High”.
  4. Write the Note (required, up to 1,000 characters) and click Create note.

Severity sets the colour: in the transcript the passage is highlighted green, amber or red, and on the Notes tab the note’s tag is grey, amber or red (low, medium, high).

The separate Note button in the action bar opens the same window without requiring a selection in the transcript.

When coaching is on in the organization, the employee whose conversation it is gets a “Manager feedback on your conversation” notification about the new note — without the note’s text. No notification is sent when the conversation cannot be matched to an Aelo user, when that user cannot open the conversation, when they have turned off the Collaboration category, or when they left the note themselves. How it arrives and how to turn it off: Settings → Notifications.

Who can add notes follows the table below: a rep adds notes only their own calls, a supervisor only in their assigned projects. Notes are visible to everyone who can open the record’s analysis; for department heads, see the notes caveat in Team → Org Structure And Supervisor Visibility. Only the author can edit or delete a note, and only while they can still add notes to that record.

Add note: type, severity and the note

CRM Entity Linking

If a CRM integration is configured, records may show linked CRM entities (contacts, deals, leads). Click the CRM link to navigate directly to the entity in your CRM.

Bulk Actions

Owner Admin Supervisor Rep

Bulk actions work in the Table view: tick the checkboxes at the start of the rows. A bar appears above the list with the number of selected records, a Clear button and these actions:

  • Analyze N records — only for records that wait for a start (manual mode), and only for the owner and admins; it shows the price and asks for confirmation first; in a project in Transcript only mode, Analyze transcripts does the same for records that have only a paid transcript
  • Reprocess — run the selected records through processing again, skipping the stages already done; for an Owner or Admin, and for a Supervisor on records of their projects
  • Export — a menu with Export as CSV, Export as JSON, Copy to Clipboard and Export to MD (the transcripts of the selected records in one Markdown file); only the selected records are exported; every role can export, a Rep only their own records
  • Archive — move records to archive; the same roles as Reprocess
  • Delete — permanently remove selected records; Owner and Admin only

A confirmation dialog appears before Delete and before Analyze (with the price); Archive runs at once.

The Reprocess, Archive and Delete buttons are shown to every role that ticks rows, but the server refuses an action the role has no right to.

Action bar for selected records

Export

Owner Admin Supervisor Rep

The Export button is in the page header, to the left of the List / Table switch. Its menu offers:

  • Export as CSV — a spreadsheet-compatible file (records-YYYY-MM-DD.csv)
  • Export as JSON — structured data
  • Copy to Clipboard — the same data as tab-separated text that pastes straight into a spreadsheet

The export covers the records already loaded in the list, with the current filters applied. The list loads 50 at a time as you scroll, so scroll to its end first to export everything. While records are selected, the header button is replaced by the one in the selection bar: it exports only the selected records and adds Export to MD (see Bulk Actions).

Export to MD has no fixed limit on the number of records: the transcripts are loaded in your browser, five at a time, so a very large selection takes a while. A record of a project in Transcript only mode has no analysis, so its speakers keep generic labels such as speaker_0 unless the transcript cleanup named them.

Export menu

Re-analyze & Retry

  • Re-analyze (Owner Admin Supervisor): Resubmit a completed record for fresh analysis (useful after changing evaluation profiles). The button reads Analyze while the record has no analysis yet.
  • Retry processing (Owner Admin Supervisor): Retry a failed record that encountered a processing error.

Both actions are available from the record detail panel: Re-analyze in the action bar, Retry processing in the failure notice above the tabs.

Re-analyze: choose the profile (Auto — category profile by default) and start

In manual mode, Retry processing runs nothing: the record waits to be started, like any other record waiting for a start, and no minutes are spent until it is started. The same happens when Auto-Analysis is off in the record’s project and it was the analysis that failed, or the record already had an analysis (it was started by hand); a failed transcription or delivery of a record without one is run again the way the project is set up, without an analysis. Only an owner or admin can start a record, so where the record would have to wait, a supervisor’s retry is refused. A record whose processing was already paid for is not charged again when started, and the price shown before the start leaves it out. A call whose transcript is saved but whose minutes were never charged is charged once when a start, a retry, a delivery retry or a re-analysis analyses that saved transcript — for the length of the conversation, or for the length of the recording when the conversation was not measured. The exception is a short call, with less conversation than the minimum set in the project: it gets no analysis, so none of these charges it minutes, and it does not wait for its recording either. When neither length is known, a start, a retry of a failed record or a delivery retry does not analyse the saved transcript: the call waits for its recording, like a call that arrived without one. A re-analysis of such a call, whose length is not known, is refused: nothing is analysed and nothing is charged. When the source answers that it has no recording of a call whose transcript is saved but not charged, a start analyses that transcript and charges it instead of closing the call.

For a record of a Uspacy installation whose processing is paused or withdrawn, or not yet activated, Re-analyze and Retry processing — including a retried transcription or media fetch — are refused with a message: the recording is stored but not analysed until processing of its source is resumed. Records of other sources are not affected.

In a project in Transcript only mode, the record is marked “Transcript without analysis” and offers Analyze with its price. Re-analyze of such a record, and Retry of one that finished, are refused with a pointer to Analyze: only the transcript was paid for, and the analysis is a separate full-minute charge. Retry of a record that failed runs again without analysis — it rebuilds only the transcript.

Access by Role

FeatureOwnerAdminSupervisorRep
View records list✅✅✅ (projects + subordinates)✅ (own records)
Upload records✅✅❌✅
View record detail✅✅✅ (projects + subordinates)✅ (own records)
Add notes✅✅✅ (project-scoped)✅ (own records)
Bulk Analyze✅✅❌❌
Bulk Reprocess / Archive✅✅✅ (project-scoped)❌
Export✅✅✅✅ (own records)
Re-analyze / Retry✅✅✅ (project-scoped)❌
Delete records✅✅❌❌

Tips & Best Practices

  • Your filters are remembered between sections, so you can leave for the dashboard and return to the same view (e.g., “Failed records this week”).
  • After uploading, check the record status. Most records complete within 2–5 minutes depending on duration.
  • Use Re-analyze after updating your evaluation profiles to see how scores change.
  • Supervisors: filter by rep to review individual rep performance.
  • Reps: your own records are automatically filtered when you access the Conversations page.