Beta

Use Twenty CRM with an AI Browser for Candidate Sourcing

Run candidate sourcing in Strawberry using Twenty CRM as one of the inputs. Specific surfaces, example prompt, real output, and tradeoffs vs alternatives.

Diagram of Strawberry AI browser workflow using Twenty CRM for candidate sourcing

If you use Twenty CRM and you regularly need to source candidates, the bottleneck is usually the same: Twenty CRM holds part of the context, but candidate sourcing also needs signals that live outside it - on the public web, in LinkedIn, in news, in other connected apps. Strawberry is built to combine the Twenty CRM context with the rest of the browser, and run the full workflow as a companion you can re-trigger every week.

This page describes specifically how Strawberry handles candidate sourcing when Twenty CRM is one of the inputs. It names the Twenty CRM surfaces involved, the signals the workflow actually needs, an example prompt you can paste, and what a good output looks like.

The job a recruiter, founder hiring, hiring manager is trying to do

The goal of candidate sourcing is to build a shortlist of 10-30 candidates who match the role and have at least one signal of openness. The success metric is concrete: 30% reply rate to first outreach, 5+ first-call conversions per 30 sourced. That definition matters because it shapes what Twenty CRM needs to contribute to the workflow.

What signals candidate sourcing actually needs

For each signal below, here is whether Twenty CRM can contribute directly or whether Strawberry has to find it via the browser:

  • Current role and tenure - Twenty CRM stores or surfaces this directly. Strawberry reads it through the connected integration.
  • Recent role changes (often visible on LinkedIn) - Twenty CRM stores or surfaces this directly. Strawberry reads it through the connected integration.
  • GitHub or content output for technical roles - Twenty CRM stores or surfaces this directly. Strawberry reads it through the connected integration.
  • Company stage match (someone leaving a Series B is more likely to talk to a seed-stage co) - Twenty CRM does not contain this directly. Strawberry uses the browser plus public sources to fetch it.
  • Geo match for hybrid roles - Twenty CRM stores or surfaces this directly. Strawberry reads it through the connected integration.
  • Openness signals (LinkedIn open-to-work, recent comments about job search) - Twenty CRM stores or surfaces this directly. Strawberry reads it through the connected integration.

What Strawberry can do inside Twenty CRM

Native Strawberry integration with read + write across all core objects; ideal for CRM hygiene, batch enrichment, and pipeline triage.

Twenty CRM surfaces Strawberry uses for this workflow: people, companies, opportunities, notes, tasks.

How Strawberry runs candidate sourcing with Twenty CRM

  1. Strawberry opens the Twenty CRM people that contains the relevant context.
  2. The companion pulls related context from Twenty CRM (companies, history, attached files) where it exists.
  3. For the parts Twenty CRM does not store, Strawberry uses the browser - web search, LinkedIn, news, the prospect's website.
  4. Strawberry synthesises the output in the shape this workflow needs: A shortlist with one row per candidate.
  5. A human reviews before any external action (send, update, post). Then the approved output is saved back to Twenty CRM or your system of record.

Example Strawberry prompt

Paste this in a new Strawberry chat with Twenty CRM connected. Adjust the specifics to your actual ICP, role, or topic.

Read this Twenty CRM people and any linked context.
Then run a full candidate sourcing workflow on it. Use the browser to fill any gaps not in Twenty CRM.
Return the output in the shape we use for candidate sourcing: A shortlist with one row per candidate: name, current role, target role fit (1-5), one personalised opening line, contact link.
Do not send anything externally. Save the draft to me to review.

What a good candidate sourcing output looks like

Here is what a finished output for candidate sourcing should look like in practice. The specifics will change for your use case, but the shape should look similar:

  • Role: Founding Engineer (Stockholm or remote EU)
  • Candidate: Marek Novak - Senior Engineer @ Klarna, 4 years
  • Fit: 5/5 (worked on payment systems, contributed to Rust open source, recent talk on type-safe APIs)
  • Opening line: noticed his RustConf talk on type-safe API contracts and our backend lead's tweet about Marek's library
  • Contact: LinkedIn DM + GitHub email

Why Twenty CRM for this, and where to use a different tool

Twenty CRM is strong for this workflow because Native Strawberry integration with read + write across all core objects; ideal for CRM hygiene, batch enrichment, and pipeline triage.

Where Twenty CRM falls short Some create endpoints require sequential calls (~800ms apart) to avoid rate limits; certain custom-field updates must go through update_record.

Consider also Google Sheets for one-off lists.

Common mistakes when running candidate sourcing

  • Spray-and-pray DMs that mention nothing specific
  • Missing the obvious signals (someone just posted 'thinking about a change')
  • No quality bar - putting 200 names on the list to look productive

Connecting Twenty CRM to Strawberry

Twenty CRM is a connected app; uses API key. Once connected, the companion can read the surfaces above without re-authenticating, and any write action still requires explicit human approval the first time the workflow runs.

Caveats

Do not let any AI agent send emails, update CRM records, or change shared systems without a clear approval step. Strawberry is strongest when the workflow combines browser context with connected-app context and a human review for sensitive actions.

How Twenty CRM + Strawberry runs candidate sourcing

1 Twenty CRM

Read

Open the relevant Twenty CRM people; pull related context.

2 Browser

Augment

Use the browser, LinkedIn, news, and other connected apps for signals outside the CRM/tool.

3 Output

Compose

Synthesise into the candidate sourcing shape: A shortlist with one row per candidate.

4 Human

Approve

Human reviews before any external action; approved output is saved back.

FAQ - Twenty CRM + AI browser for candidate sourcing

Can Strawberry do candidate sourcing entirely inside Twenty CRM?

No, and that is the point. candidate sourcing needs signals Twenty CRM does not store - public web, LinkedIn, news, other apps. Strawberry combines Twenty CRM with the browser, which is where the real value comes from.

Does Twenty CRM need to be the primary CRM or system of record?

Not necessarily. Twenty CRM can be one input among several. Strawberry can read it as context even if your primary system of record is somewhere else.

What permissions do I need on Twenty CRM?

Read access to the surfaces you want Strawberry to use (people, companies, opportunities). Write permissions are only needed if you want Strawberry to update Twenty CRM after a human approves the change. Twenty CRM is a connected app; uses API key.

What is the realistic success metric for candidate sourcing?

30% reply rate to first outreach, 5+ first-call conversions per 30 sourced - that is the target Strawberry helps you hit, not the only thing it measures.

What is the biggest mistake to avoid?

Spray-and-pray DMs that mention nothing specific.

Run candidate sourcing in 10 minutes with Strawberry and Twenty CRM

  1. Open Twenty CRM

    Connect Twenty CRM so Strawberry can read people, companies, opportunities and combine them with the rest of the brief. Pin the specific record, list, or query you want to start from so the agent doesn't drift.

  2. Tell Strawberry the brief

    Drop the prompt below. Replace the placeholder with the actual recruiter target - one name, one URL, or one Twenty CRM reference is enough. Keep the goal explicit: build a shortlist of 10-30 candidates who match the role and have at least one signal of openness.

  3. Let it gather signals

    Strawberry pulls current role and tenure and recent role changes (often visible on LinkedIn), then layers public web sources in parallel. You should see citations next to each fact - that is the audit trail. Watch the Twenty CRM side: Some create endpoints require sequential calls (~800ms apart) to avoid rate limits.

  4. Review before write-back

    Output lands in the shape you asked for: A shortlist with one row per candidate: name, current role, target role fit (1-5), one personalised opening line, contact link Read it once. Fix anything off. The success metric is 30% reply rate to first outreach - if the draft doesn't hit that bar, send it back with a one-line correction.

  5. Save it as a routine

    If you'll source candidates again next week, click Save as routine. Pick a cadence (daily, weekly, on-trigger). Strawberry re-runs the whole flow on schedule and pings you when the new output is ready.

Paste-ready prompt for candidate sourcing with Twenty CRM

You are helping me source candidates. Use Twenty CRM as one input and the public web for the rest.

Target: [paste one recruiter target here - a Twenty CRM reference, a name + company, or a URL]

Goal: build a shortlist of 10-30 candidates who match the role and have at least one signal of openness.

Signals to gather:
- current role and tenure
- recent role changes (often visible on LinkedIn)
- GitHub or content output for technical roles
- company stage match (someone leaving a Series B is more likely to talk to a seed-stage co)
- geo match for hybrid roles
- openness signals (LinkedIn open-to-work, recent comments about job search)

Output shape: A shortlist with one row per candidate: name, current role, target role fit (1-5), one personalised opening line, contact link

Rules:
- Cite every fact with a link or a Twenty CRM reference. If you cannot find a signal, say so explicitly rather than guessing.
- Do not invent specifics. Use real, dated signals from the last 90 days where possible.
- If a fact would change the outcome and is missing, pause and ask me before writing the final output.

When the output is ready, surface it in this chat. Do not write back to Twenty CRM or send anything externally until I approve.

Paste this into Strawberry's chat field. Replace the target placeholder before running.

When Twenty CRM + Strawberry is NOT the right fit for candidate sourcing

Skip this setup if any of the following is true:

  • You don't actually need Twenty CRM signals. If everything you need lives on the public web, drop the Twenty CRM step and let Strawberry run on URLs alone - it's faster.
  • A known Twenty CRM constraint blocks the speed gain: Some create endpoints require sequential calls (~800ms apart) to avoid rate limits.
  • The buyer (recruiter, founder hiring, hiring manager) doesn't own the decision. If the brief gets handed to someone who'll redo the research, the audit-trail-in-Strawberry advantage is wasted.

3 mistakes that kill this workflow

  1. Spray-and-pray DMs that mention nothing specific. Twenty CRM is one input. Strawberry's edge is combining it with everything else. Stop at Twenty CRM-only signals and you'd have been faster with native Twenty CRM reports.
  2. Missing the obvious signals (someone just posted 'thinking about a change'). Pre-check Twenty CRM for a recent touch or duplicate before Strawberry acts on the output. A duplicate hit burns the relationship.
  3. No quality bar - putting 200 names on the list to look productive. Strawberry is built so a human reviews before any external action. Skipping that review to save time is how you ship a wrong fact to a real person.

Honest tradeoff vs alternatives

You could source candidates inside Twenty CRM alone using its native features, or with a dedicated candidate sourcing tool. Twenty CRM alone gives you tighter data fidelity but misses every signal that lives off-platform. A specialised candidate sourcing tool gives you better dashboards but its scope ends where its integrations end, and most of the real signal still lives on the open web.

Strawberry's edge with Twenty CRM: Native Strawberry integration with read + write across all core objects; ideal for CRM hygiene, batch enrichment, and pipeline triage. The price you pay: an agent run takes 30-90 seconds; a native Twenty CRM action loads in 2. For a one-off question you already know the answer to, use Twenty CRM directly. For an output you'll redo every week or every account, route it through Strawberry as a saved routine so the synthesis happens once and re-runs automatically.

What a real output looks like

  • Role: Founding Engineer (Stockholm or remote EU)
  • Candidate: Marek Novak - Senior Engineer @ Klarna, 4 years
  • Fit: 5/5 (worked on payment systems, contributed to Rust open source, recent talk on type-safe APIs)
  • Opening line: noticed his RustConf talk on type-safe API contracts and our backend lead's tweet about Marek's library
  • Contact: LinkedIn DM + GitHub email
Download