# Read many records to answer a pipeline or reporting question

> Read many records to answer a pipeline or reporting question: 3 companies fully do it, 17 with a limit, 27 carry the tag. Every listing in CRM & customer platforms with the reason and a source.

- Canonical: https://www.anchorterminal.com/niches/crm.bulk-read-report/
- Markdown: https://www.anchorterminal.com/niches/crm.bulk-read-report/index.md (~1,650 tokens)
- Slim: https://www.anchorterminal.com/niches/crm.bulk-read-report/index.min.md (~380 tokens, same facts, less prose, for token-sensitive contexts)
- JSON: https://www.anchorterminal.com/niches/crm.bulk-read-report/index.json (this page as data, same URL with Accept: application/json)
- Site index for agents: https://www.anchorterminal.com/llms.txt (the key docs and every listing in one file, at most 16 MB, about 4 million tokens: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- MCP: https://www.anchorterminal.com/mcp (the whole catalogue as tools, no key)
- DNS: dig +short TXT <slug>.grade.anchorterminal.com (the grade as a DNS TXT record, for a sandbox with no HTTP where names still resolve; also top., best.<category>. and <a>.vs.<b>. under the zone, and the same under grade.letme.dev)
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)
- Updated: 2026-11-15

Two or three companies' docs show a full fit.

| Companies, full fit | With a limit | Carry the tag | Share of the category's work | First-hand accounts |
| --- | --- | --- | --- | --- |
| 3 | 17 | 27 | 11 | 7 |

## What counts as done

An answer computed over every matching record, with the count of records read stated; a truncated read is flagged, not presented as complete.

Measured, for our own tests: under 20 tool calls for 10,000 records.

How agents ask for it: how many contacts came from this campaign; which deals have stalled for 30 days; summarise the pipeline by stage; chart closed-lost reasons this quarter

## Does the job

The docs show it meets the test above, and nothing we read says otherwise.

- **Close API + MCP (Elastic Inc)**: aggregate_objects counts, sums and groups any filtered object set server-side in one call, and search returns the total count of all results with a cursor for the rest.
- **Microsoft Dynamics 365 Sales (Microsoft Corporation)**: Web API pages return up to 5,000 rows with @odata.nextLink, so 10,000 records take two or three calls, and read_query runs SQL SELECT over the MCP.
- **Zoho CRM (Zoho Corporation Private Limited)**: A Bulk Read job exports a module filtered by criteria as CSV in a create, poll and download sequence, and Get Records pages 200 at a time with more_records, so 10,000 records fit well under 20 calls.

## Does the job, with a limit

It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.

- **Affinity (Project Affinity, Inc.)**: Cursor pages (nextUrl) and search_all_companies give totals, but the maximum page is 100, so 10,000 records take about 100 calls against a 100,000-a-month cap on Scale and Advanced.
- **Attio API + MCP (Attio Limited)**: run-basic-report counts and groups server-side, and list-records defaults to 500 a page (20 calls for 10,000), but offset paging has no has-more flag and SQL is Enterprise-only.
- **Breakcold (Logike SAS)**: Cursor paging tops out at 100 records a page, so 10,000 records take about 100 calls, each drawing on the weekly AI usage allowance; no aggregate or export route was found.
- **Capsule CRM (Zestia Limited)**: calculate_value_of_opportunities totals a filtered set server-side, but other reads page at 100, so 10,000 records take about 100 calls.
- **Copper API (Copper CRM, Inc.)**: Search pages hold at most 200 records, so 10,000 take 50 calls; X-PW-TOTAL is only an upper bound and results stop at 100,000.
- **folk API + MCP (Folk Inc.)**: Searches return nextCursor; idsOnly gives up to 500 ids a page (20 calls for 10,000), but full records come 100 a page and there is no count or aggregate tool.
- **Freshsales API (Freshworks Inc.)**: Lists and searches cap at 100 a page, so 10,000 records take about 100 calls against 1,000 requests an hour shared by the whole account, and no aggregate endpoint exists.
- **HubSpot API + MCP (HubSpot, Inc.)**: query_crm_data runs SQL with aggregation and search_crm_objects pages 200 at a time, but a01 counted contacts in 120 loops, 80 minutes and 18M tokens, and a06 found record-by-record access too slow.
- **Nutshell (Nutshell, Inc.)**: The MCP fetches summary or detailed lead reports, so pipeline counts are computed server-side, but list paging limits aren't stated and record reads beyond reports page by page[limit].
- **Payani AI (Payani Media LLC)**: The page says the assistant can read reporting across CRM, campaigns and analytics, but no page size, count or truncation behaviour is published.
- **Pipedrive API + MCP (Pipedrive Inc.)**: v2 lists take limit up to 500, so 10,000 records need 20 calls at 20 tokens each against a company-wide daily budget; no aggregate endpoint or MCP report tool was found.
- **Relaticle (Relaticle, Inc.)**: get-crm-summary-tool and aggregate-opportunities-tool give counts and totals by stage server-side, and lists report total and has_more, but MCP lists cap at 25 a page, so record-level reads of 10,000 take 400 calls.
- **Salesflare (Salesflare BVBA)**: Lists take limit and offset with small defaults (10 to 20) and no documented maximum or has-more flag, and there's no aggregate endpoint, so a full count can't be planned.
- **Salesforce API + MCP (Salesforce, Inc.)**: a30 saw SOQL rows past the first page dropped with no signal, and the bulk and query limits pages answered 403, so the path to 10,000 records in under 20 calls couldn't be confirmed.
- **Streak API + MCP (Rewardly, Inc.)**: The help article shows pipeline summaries by stage and value, but no page size, count field or truncation flag is published for the MCP or the API.
- **SugarAI (SugarCRM) (SugarAI Software Inc.)**: /filter/count returns counts server-side and filter pages report next_offset -1 at the end, but max_num defaults to 20 with no stated maximum, and SugarCloud blocks an IP above 20 requests a second.
- **Twenty API + MCP (Twenty.com PBC)**: group_records aggregates server-side, but page size isn't documented, batches cap at 60 records, and the API allows 100 requests a minute.

## What stops agents finishing

- page size of a few hundred records turns a count into hours and millions of tokens (a01)
- single-page query with no has-more flag (a30, a31)
- monthly or daily call caps shared across the account (Affinity, Pipedrive, Freshsales per context notes)

## Around it

After: documents.chart, chat.post-summary.

## Where this comes from

Weight 11 of 100 in CRM & customer platforms: 7 of 48 accounts (a01, a02, a05, a06, a24, a30, a31); a01 gives the only cost figures (18M tokens, 80 minutes)

As data: https://www.anchorterminal.com/api/v1/use-cases/crm.bulk-read-report.json, or the MCP tool get_use_case with id crm.bulk-read-report. Draft, read 2026-11-15 from public docs and first-hand accounts, not yet reviewed by a person.

---

Query this catalogue directly over MCP: https://www.anchorterminal.com/mcp (Streamable HTTP, no key). How: https://www.anchorterminal.com/agents/#mcp
