Search Memories

Use search in Memories to find the exact customer moments, quotes, and evidence snippets you want to review, share, or group into clusters.

Memories search combines:

• Keyword search for the words inside memory text, titles, and descriptions
• AI-powered semantic search to find memories that are related in meaning, even when they do not use the exact same wording
• Filter-based search to narrow results by metadata like date, tags, source, creator, contact, cluster, and account details

What search looks at

Search on memories is based on the memory itself. In practice, that means search can use:

• The memory text
• The memory title
• The memory description
• Memory metadata such as record date, created date, tags, source, contact, creator, linked cluster, source record, and linked account fields

Memories search is best for finding the exact evidence you want to work with.

Start with a simple search

Start by typing a normal phrase, question, customer pain point, or theme.

Examples:

• pricing objections
• handover issues
• slow onboarding
• customers asking for SSO

If you want to search for an exact phrase, put it in quotes.

Examples:

• "pricing objections"
• "customers asking for SSO"

This is useful when you want a literal match instead of a semantic one. Semantic search matches based on meaning, so the exact words do not have to appear. Exact matching is useful when you are looking for product names, phrases you remember, or other specific wording.

For plain-text searches, results are sorted by most recent record date first.

Use filters to narrow results

You can combine a free-text search with filters, or use filters on their own.

Filter	What it does
date:	Returns memories recorded during the selected date or date range
created:	Returns memories created in NEXT during the selected date or date range
tag:	Returns memories that include all selected tags
tag_or:	Returns memories that include any of the selected tags
creator:	Returns memories created by this person
contact:	Returns memories from this contact
source:	Returns memories from this source
origin:	Returns memories created from this origin
cluster:	Returns all memories that were assigned to this cluster
source_recording:	Returns memories from this source record
account.name:	Returns memories from this account
account.industry:	Returns memories from accounts in this industry
account.region:	Returns memories from accounts in this region
account.segment:	Returns memories from accounts in this segment
account.account_type:	Returns memories from accounts of this type
account.employee_size:	Returns memories from accounts in this employee-size range
account.status:	Returns memories from accounts with this status

Examples:

• pricing tag:Survey
• date:Last 30 days
• created:Last 7 days
• cluster:Churn Risks
• source:Zoom
• account.name:Spotify

Filter tips

• Use multiple filters together to make a search more precise.
• Use multiple tag: filters when you want memories that include all selected tags.
• Use tag_or: when you want memories that match any of the selected tags.
• date: uses the record date.
• created: uses the date the memory was created in NEXT.

Suggestions and autocomplete

As you type, NEXT suggests filters and values to help you build your search faster.

For memories, suggestions focus on:

• Date ranges
• Tags
• Account names

When you start typing a specific filter such as creator:, source:, contact:, cluster:, or an account... filter, NEXT will suggest matching values for that filter.

Date suggestions support both quick ranges and natural language. For example, you may see options like:

• Last 7 days
• Last 30 days
• Last 90 days
• Yesterday
• Last week
• Last month
• Last quarter
• A parsed date range from something like date:December 2024
• A rolling range from something like date:last 500 days

The same date suggestions also work for created:.

Once you already use a filter in the current search, NEXT avoids suggesting the exact same filter value again.

Calendar date syntax

Calendar date filters are used when the search should follow calendar boundaries instead of a rolling window.

For example, date:Last 7 days depends on the exact time the search runs. If it runs at 07:00, it covers the previous 7 days from 07:00. A calendar filter such as date:Last week covers the complete previous calendar week, from Monday 00:00 through Sunday 23:59:59.999 in the teamspace time zone.

Preferred calendar filters:

Filter	Meaning
date:cal.d.prev-cal.d.now	Yesterday
date:cal.w.prev-cal.w.now	Last week, Monday to Sunday
date:cal.m.prev-cal.m.now	Last month
date:cal.q.prev-cal.q.now	Last quarter
created:cal.d.prev-cal.d.now	Memories created yesterday
created:cal.w.prev-cal.w.now	Memories created during the last week
created:cal.m.prev-cal.m.now	Memories created during the last month
created:cal.q.prev-cal.q.now	Memories created during the last quarter

The general form is:

date:cal.<period>.<anchor>-cal.<period>.<anchor>

Periods:

Period	Meaning
d	Day
w	Week
m	Month
q	Quarter

Anchors:

Anchor	Meaning
now	Start of the current period
prev	Start of the previous period
prev_N	Start of N periods back

Examples:

Filter	Meaning
date:cal.d.prev_2-cal.d.prev	Last 1 day from 2 days back
date:cal.w.prev_2-cal.w.prev	Last 1 week from 2 weeks back
date:cal.m.prev_2-cal.m.now	Last 2 months
date:cal.m.prev_4-cal.m.prev_2	Last 2 months from 4 months back

Times can be added when a calendar boundary should start at a specific local time. If no time is provided, NEXT uses 00:00.

Examples:

• date:cal.d.prev@09:00-cal.d.now@09:00
• date:cal.w.prev@12:00-cal.w.now@12:00

Calendar filters are resolved using the teamspace IANA time zone, such as Europe/Amsterdam.

If a teamspace does not have a time zone configured, NEXT falls back to UTC. This fallback keeps searches working for incomplete configuration, but it can shift calendar boundaries compared with the team's local working day. Configure the teamspace time zone when calendar filters need to match local day, week, or month boundaries.

Relative date syntax

Relative date filters are used when the search should follow a rolling window relative to the current time instead of calendar boundaries.

For example, date:now_sub_7d-now means "from exactly 7 days ago until now". Unlike date:cal.w.prev-cal.w.now, this range does not snap to Monday through Sunday. If the search runs at 14:37, the end of the range is 14:37 right now, and the start is 14:37 seven days earlier.

The general form is date:now_sub_<days>d-now, but you can also use two relative offsets such as date:now_sub_2d-now_sub_1d.

Examples:

Filter	Meaning
date:now_sub_1d-now	Last 1 day
date:now_sub_7d-now	Last 7 days
date:now_sub_30d-now	Last 30 days
date:now_sub_2d-now_sub_1d	From 2 days ago until 1 day ago

Use relative syntax when you want a true rolling window. Use calendar syntax when the range should start and end on local day, week, or month boundaries.

How to get better results

• Start broad, then add filters once you see the result set.
• If an exact keyword search does not find what you expect, try a more descriptive phrase instead of a single term.
• Use quotes when you want to search for an exact phrase.
• Use tags and clusters when you already know the theme you want.
• Use account filters when you are looking for evidence from a specific customer segment.
• Use created: when you want to focus on recently generated memories, and date: when you want to focus on when the conversation happened.

FAQ

Why isn't my tag showing up in search suggestions?

Search suggestions include all tags in your teamspace, even if they are not assigned to a memory yet. Tags used on more memories appear first, so unused tags are usually near the end of the suggestions.

There can also be a short delay before a tag appears in the search suggestions. Teamspaces check for updated suggestions about every minute, so recently added or retagged memories should normally appear shortly afterward.

How do I search for memories with all selected tags?

Use tag: when you want an AND search.

Example:

• tag:Interview tag:Pricing

This returns memories that have both tags.

How do I search for memories with any of several tags?

Use tag_or: when you want an OR search.

Example:

• tag_or:Interview tag_or:Pricing

This returns memories that have either tag.

Can I combine tag: and tag_or: in the same search?

Yes.

Example:

• tag:Interview tag_or:Pricing tag_or:Churn

This looks for memories that must have the Interview tag and must also have either Pricing or Churn.