Agentic Index
Closely vs Kakiyo (2026)
Both vendors last verified
Closely and Kakiyo both work LinkedIn, at different levels of autonomy. That verdict is the Agentic Index coverage score, graded from each vendor's own published materials.
Closely is outreach automation, with campaigns, contact collection and verification, and email alongside LinkedIn, from $65 a month or $39 billed annually. Kakiyo is an autonomous LinkedIn SDR that runs entire conversations, from connection request through objection handling to a booked meeting, from €125 a month with every feature on every plan.
This comparison is published by Agentic Index, an independent agentic AI vendor research platform. Closely and Kakiyo are each graded against the same 14 capability Agentic Index taxonomy, from the vendor's own public materials under the Agentic Index verification standard, alongside 1,165 researched vendors. No vendor pays for placement and no vendor has reviewed this page. How this evidence is graded
Choose Closely if
- You want automation you drive, sequences and campaigns, not conversations run for you.
- Email plus LinkedIn in one tool covers your two channel motion cheaply.
- Starter at $39 a month on annual billing, for one LinkedIn account, keeps the entry cost low.
Choose Kakiyo if
- Conversation handling is the bottleneck, with replies and objections sitting unanswered.
- Autonomy all the way through booking means the calendar fills without your attention.
- Volume based tiers with identical features keep the upgrade decision honest.
| Feature | C Closely |
K Kakiyo |
|---|---|---|
| Action & orchestration | ||
|
Integrations & Tool Calling Ability to connect agents to real systems through native integrations, OAuth-authenticated actions, custom tools, APIs, webhooks, or MCP-compatible tools. |
||
|
CloselyIntegrations & Tool Calling Named CRM connectors for Salesforce, Pipedrive, HubSpot and GoHighLevel come on every plan, with Gmail and Outlook over OAuth and custom SMTP for sending. The agent acts through LinkedIn, email and CRM writes. A webhooks integration that the customer wires to their own URL pushes contact data out of campaigns, My Network and the Inbox on tag and note events, with emails and phones included by default, and it is built for use with Webhooks by Zapier. That lets the customer connect further systems on their own. The route runs one way. Closely says there is no API, and nothing pushes data into Closely. Sourcecloselyhq.com/pricing and help.closelyhq.com webhooks articleread 2026-10-01 |
||
|
KakiyoIntegrations & Tool Calling The HubSpot integration syncs prospects as contacts, logs timeline events, associates companies and creates deals automatically. Kakiyo also offers a LeadShark integration, sync for the do not contact list, webhooks and a public API of its own, and CRM, API and MCP integrations are on every plan. Sourcedocs.kakiyo.com HubSpot Integration and kakiyo.com/pricingread 2026-10-01 |
||
|
Workflow Orchestration Ability to sequence, branch, retry, route, and combine deterministic workflow nodes with autonomous agent steps. |
||
|
CloselyWorkflow Orchestration Sequences chain profile visits, connection requests, follow-up messages, InMails, post engagement and email steps across every connected LinkedIn account. They run unattended in Closely's cloud, not on the user's machine, advance or halt on reply detection and connection acceptance, and push outcomes to the CRM. Behavior based branching moves a prospect to email when a connection request goes unaccepted. The scheduler is deterministic and spans many accounts, and the AI only drafts message copy. Closely describes no goal decomposition, no dynamic replanning, no branching the agent decides beyond configured conditions and no tool selection. The Sales AI Agents research a lead's profile and company site and write the message, but they do not choose what to do next. Sourcecloselyhq.com/linkedin-automation and closelyhq.com/multichannel-campaignsread 2026-09-02 |
||
|
KakiyoWorkflow Orchestration The agent runs the whole conversation, branching on what the prospect says. In Autopilot it handles everything from the first message to the meeting booking. Prompts, offerings, follow up messages and campaign rules the customer configures shape qualification and objection handling, and the conversation statuses, Active, Paused, Completed and Inactive, carry it between steps. Sourcedocs.kakiyo.com Inbox and Conversation Management and Understanding the Promptsread 2026-10-01 |
||
|
Triggers & Channel Coverage How agents wake up and where they work: schedules, webhooks, message events, CRM events, inbox events, chat, email, voice, and collaboration tools. |
||
|
CloselyTriggers & Channel Coverage Two channels carry the sequences. LinkedIn covers connection requests, direct messages, InMails, profile visits and post engagement, and email comes with an integrated finder and sender for hybrid campaigns. Work starts on sequence state changes, such as a connection accepted, a reply received or a step elapsed, and on inbound webhook events from LinkedIn Lead Sync forms. Behavior based rules move a prospect to email when a connection request goes unaccepted within a set period. Phone is not one of the channels. Sourcecloselyhq.com/multichannel-campaigns and closelyhq.com/linkedin-automationread 2026-09-02 |
||
|
KakiyoTriggers & Channel Coverage Every prospect reply starts the agent's next turn with no person initiating it, and follow up messages fire on their own schedule inside a campaign. The agent works only in LinkedIn. Sourcedocs.kakiyo.com Inbox and Conversation Management and Follow-up Messagesread 2026-10-01 |
||
| Knowledge & context | ||
|
Knowledge Grounding & RAG Ability to ground agent behavior in company data through document ingestion, retrieval, external knowledge APIs, semantic search, or RAG layers. |
||
|
CloselyKnowledge Grounding & RAG Data comes from a B2B contact database, a cascading enrichment engine that falls back through sources until a verified work email is found, live LinkedIn enrichment for job titles, companies and industries, and phone lookup. Message generation is grounded in the prospect's LinkedIn profile and company website, read at generation time, and inbound webhook payloads from LinkedIn Lead Sync forms feed personalization. Closely names no grounding structure the customer owns, such as a brand voice or messaging framework object, an ICP or persona model the agents read, a signal taxonomy or a retrieval layer over the customer's documents or collateral. The data belongs to Closely and its data sources, not to the customer. Sourcecloselyhq.com/data-enrichment, closelyhq.com/email-finder-tool and closelyhq.com/business-contact-data-providerread 2026-09-02 |
||
|
KakiyoKnowledge Grounding & RAG The agent is grounded in what the customer configures, assembled for each conversation. Offerings and prompts carry the product, positioning and qualification criteria the agent answers from, and each prospect is enriched before outreach. There is no maintained index, retrieval over the customer's documents or citation. Sourcedocs.kakiyo.com Understanding the Offerings and Understanding the Promptsread 2026-10-01 |
||
|
Memory & State Persistence Ability to persist context across a run, conversation, workflow, user, team, or longer-term memory layer. |
||
|
CloselyMemory & State Persistence The unified inbox keeps LinkedIn and email threads across every connected account, and the agency dashboard keeps campaign state for each client. Sequence state, such as the step reached, a connection accepted or a reply received, persists and drives execution. Closely names no agent memory across sessions, no learned preference carried between campaigns and no retention or forgetting control for the agent. Message generation reads the prospect profile fresh each time. Sourcecloselyhq.com/pricing and closelyhq.com/linkedin-automationread 2026-10-01 |
||
|
KakiyoMemory & State Persistence Thread state and prospect records carry over, but there is no memory layer. Kakiyo says the AI remembers the full conversation history and picks up a paused thread where it left off, and prospect status is tracked through the campaign. That is context within the thread plus the lead record, with no store carried across conversations and no scope, lifetime or delete path. Sourcedocs.kakiyo.com Inbox and Conversation Managementread 2026-10-01 |
||
| Control & trust | ||
|
Human Oversight & Guardrails Approval steps, consent checkpoints, escalation rules, structured guardrails, policy constraints, and pause/resume controls. |
||
|
CloselyHuman Oversight & Guardrails The configurable guardrails protect the sending account. They include smart daily activity caps, randomized delays between actions, working hours windows, a warm up mode that ramps activity gradually, a dedicated residential proxy per account, and automatic pausing of a sequence when a prospect replies. Each keeps automation beneath LinkedIn's detection thresholds and protects the customer's account from restriction. None of them gates an agent decision, reviews generated copy before it sends, or protects the person receiving the message. People get involved in the unified inbox after messages have gone out. Closely publishes no approval workflow, review before sending, sign off by role or policy engine. It does publish an anti-spam policy and an opt-out manual, a commitment about recipients that is not a mechanism in the product. Sourcecloselyhq.com/linkedin-automation and closelyhq.com/anti-spam-policyread 2026-09-02 |
||
|
KakiyoHuman Oversight & Guardrails People can step in and set limits, but nothing is approved before the agent speaks. Copilot mode lets a person step in at any time, pause a conversation and send manually, then resume the AI, and daily outreach limits and the do not contact list are constraints the customer sets. Messages are not approved one by one, so nothing holds the agent before it sends. Sourcedocs.kakiyo.com Inbox and Conversation Management and LinkedIn Account Safety Guideread 2026-10-01 |
||
|
Security, Identity & Governance RBAC, SSO, auditability, encryption, least-privilege tool access, compliance posture, and data handling policy. |
||
|
CloselySecurity, Identity & Governance A full DPA with standard contractual clauses governs transfers out of the EEA. Closely commits by contract to background checks on every employee or contractor before granting access to client personal data, and after termination it keeps data only on an aggregated or de-identified basis. The privacy policy honors GDPR Articles 12 to 23 with a one month erasure window and gives Article 13 and 14 notice to EEA data subjects, and Closely publishes an anti-spam policy. Closely holds no certification. It says it undergoes audits aligned with ISO 27001 standards. Closely publishes privacy, anti-spam, terms and cookies pages, but no security page, trust center, sub-processor list or penetration test evidence. It cites AES-256 at rest and TLS 1.3 in transit, though it also says the exact protocols are not disclosed. Closely holds the customer's authenticated LinkedIn session inside its own cloud, and that login is not easy for the customer to rotate. Sourcecloselyhq.com/legal/data-processing-agreement, closelyhq.com/pp and closelyhq.com/anti-spam-policyread 2026-09-02 |
||
|
KakiyoSecurity, Identity & Governance Roles exist, but there is no attestation. One Owner handles billing and team administration, and invited Admins get full access to campaigns, profiles, prospects and analytics but none to billing or team settings. There is no access control per campaign. There is also no SSO, SAML, SCIM, attestation or trust page, even though the product holds LinkedIn credentials and operates accounts for the customer. Sourcedocs.kakiyo.com Team Managementread 2026-10-01 |
||
|
Observability & Auditability Traces, logs, execution histories, metrics, audit events, and debugging detail for production agent behavior. |
||
|
CloselyObservability & Auditability A unified dashboard reports performance for each sequence across every connected LinkedIn account. It tracks connection acceptance, reply rates, InMail replies, email opens and clicks, meetings booked and step completion, and an agency can monitor multiple client accounts from one place without logging in and out. Outbound webhooks push activity events into the customer's own systems in real time, so a customer who wants a durable trail can build one there. Closely names no audit log, no record of what the AI generated against what a human edited, no trail of actions tied to a user and no retention policy on activity history. Sourcecloselyhq.com/pricing and help.closelyhq.com webhooks articleread 2026-09-02 |
||
|
KakiyoObservability & Auditability Each conversation shows its transcript, not a record of actions. The full message history, with AI messages, prospect replies and manual messages, appears with its status, and Analytics reports campaign results. There is no view of the agent's decisions or tool calls and no separate audit log. Sourcedocs.kakiyo.com Inbox and Conversation Management and Analyticsread 2026-10-01 |
||
|
Deployment & Data Residency Deployment modes and options, including SaaS, dedicated cloud, VPC, on-prem, hybrid, local runtime, and self-hosting. |
||
|
CloselyDeployment & Data Residency Delivery is in Closely's own cloud only, which is what lets sequences run when the user's machine is off. Closely names no region selection, on premises option or private VPC, and no way to keep the LinkedIn session on infrastructure the customer controls. Its operating entity is Esterson Limited in Cyprus, and the DPA governs transfers out of the EEA under standard contractual clauses, which implies processing in the EEA by default. The location is fixed, and the customer has no say in it. Sourcecloselyhq.com/terms and closelyhq.com/legal/data-processing-agreementread 2026-10-01 |
||
|
KakiyoDeployment & Data Residency Kakiyo runs the cloud, and customers have no choice of where data lives. Each LinkedIn account runs on Kakiyo's own virtual machine with a proxy placed to match the prospect's geography, which is sending infrastructure. There is no region, VPC, single tenant or on premises option for customer data. Sourcedocs.kakiyo.com and kakiyo.com/pricingread 2026-10-01 |
||
| Solution readiness | ||
|
Prebuilt Agents, Templates & Packs Ready-made workflows, packaged employees, templates, blueprints, industry solutions, and role-specific agents that reduce time-to-value. |
||
|
CloselyPrebuilt Agents, Templates & Packs Reusable campaign sequences combine connection requests, messages, InMails and email steps. Message templates carry smart variables that AI personalization then varies for each prospect, and campaign patterns are offered by job title and segment. Agencies get a white label layer with a master dashboard, portals for each client, custom branding and domain. Sequence templates configure one automation engine. Closely offers no catalog of named agents or of templates for named jobs, and no marketplace, importable play library or community exchange. Sourcecloselyhq.com/pricing and closelyhq.com/white-labelread 2026-09-02 |
||
|
KakiyoPrebuilt Agents, Templates & Packs Each customer builds one agent, and there is nothing to adopt. The customer writes the AI Builder setup, prompts and offerings for its own offer, and the optimization guides are advice, not assets to adopt. There is no template gallery or set of prebuilt agents. Sourcedocs.kakiyo.com AI Builder Quick Setupread 2026-10-01 |
||
| Platform extensibility | ||
|
Model Flexibility & Routing Ability to work across multiple foundation models, route tasks to different models, or let buyers bring their own providers and keys. |
||
|
CloselyModel Flexibility & Routing Sales AI Agents analyze a lead's profile and company website and craft tailored messages, and AI personalization is metered at two credits per use. Closely names no model provider. It publishes no model selection, routing, bring your own model or key, or temperature or style control beyond message templates. Every message the AI writes goes out under the customer's own name and LinkedIn identity. Sourcecloselyhq.com/linkedin-automation and closelyhq.com/pricingread 2026-09-02 |
||
|
KakiyoModel Flexibility & Routing The customer chooses across model makers. An AI Model dropdown on each prompt offers Claude Sonnet, Haiku and Opus, GPT models, Gemini Pro and Flash, and Grok, with Claude Sonnet 4.5 the recommended default. Sourcedocs.kakiyo.com How to Choose Your AI Modelread 2026-10-01 |
||
|
APIs, SDKs & MCP Extensibility Composability layer: stable APIs, SDKs, MCP tool consumption/serving, custom tools, and integration into internal systems. |
||
|
CloselyAPIs, SDKs & MCP Extensibility A webhooks integration pushes information from Closely to the customer's system, and in Closely's words it does not push data the other way. A Zapier listing reaches downstream apps, and a Chrome extension handles LinkedIn data extraction. A customer can react to Closely events but cannot drive Closely programmatically. Closely publishes no API documentation, developer portal or reference, and says there is no API. Sourcehelp.closelyhq.com webhooks article, Closely support response and closelyhq.com site navigationread 2026-09-02 |
||
|
KakiyoAPIs, SDKs & MCP Extensibility An API covers Kakiyo's own platform, with endpoints for campaigns, prospects, prompts, products, agents, workspaces, analytics, do not contact lists and webhooks, and it uses API key authentication. A webhooks guide and an MCP server for assistants come with it. Sourcedocs.kakiyo.com API Reference and MCP Serverread 2026-10-01 |
||
|
Testing, Debugging & Optimization Testing, debugging, scoring, retries, fallbacks, quality gates, and optimization loops for improving agent workflows before and after deployment. |
||
|
CloselyTesting, Debugging & Optimization Campaign analytics cover connection acceptance and reply rates, InMail replies, email opens and clicks, step level completion per sequence and across accounts, campaign timelines and team performance comparison. A/B testing of message variations across LinkedIn and email lets a customer compare one sequence against another. Closely names no sandbox or dry run mode, no evaluation harness, no output scoring, no regression suite and no way to check a change to message generation before it reaches real prospects. Messages go out under the customer's own LinkedIn identity, so a bad message puts the account at risk, not just a campaign. Sourcecloselyhq.com/pricing and closelyhq.com/multichannel-campaignsread 2026-09-02 |
||
|
KakiyoTesting, Debugging & Optimization The Sandbox is a rehearsal the user reads, with no verdict. It simulates a LinkedIn conversation between the configured agent and a fictional prospect, and a checklist covering tone, product knowledge, qualification and boundaries guides the user's check by eye. Nothing is scored, compared across prompt versions or gated. Sourcedocs.kakiyo.com Understanding the Sandboxread 2026-10-01 |
||
| Specialist automation | ||
|
Browser & Computer Use Browser, desktop, or remote/local computer control for workflows that cannot be handled through stable APIs alone. |
||
|
CloselyBrowser & Computer Use The automation works inside LinkedIn itself. Closely operates the customer's authenticated LinkedIn session to perform profile visits, connection requests, direct messages, InMails and post engagement. Each account runs on a dedicated residential proxy so the traffic presents as a real user, with warm up mode and daily caps set to mimic human behavior. LinkedIn publishes no sanctioned interface for these actions at this scale, so the product necessarily operates the interface instead of calling an API. A Chrome extension, listed on the Chrome Web Store, also extracts data from LinkedIn pages the user is viewing. The automation is built for one site, with no general browser or computer control and no way to point it at another application. Closely does not say whether it runs a headless browser or an emulated client. Sourcecloselyhq.com/linkedin-automation and closelyhq.com/extensionread 2026-09-02 |
||
|
KakiyoBrowser & Computer Use The agent works the real LinkedIn interface on infrastructure Kakiyo runs. Each connected LinkedIn account runs on its own virtual machine that simulates human behavior, with a dedicated IP, proxies matched to location, encrypted credential storage and rate limits, and the agent sends connection requests and messages through it. Sourcedocs.kakiyo.com How Kakiyo Works Behind the Scenes and LinkedIn Account Safetyread 2026-10-01 |
||
Pricing snapshot
Sourced from the Index pricing dataset · open each vendor's profile for full detail.
| Pricing | C Closely |
K Kakiyo |
|---|---|---|
|
Entry price Lowest public entry point |
Starter $65 a month for one LinkedIn account ($39 a month billed annually). Growth $159 ($95 annual) with three accounts and 3,000 credits. Essential $259 ($155 annual) with five accounts and 5,000 credits. Custom tier above that. | Pioneer EUR 125 a month (EUR 1,500 a year) · Hunter EUR 315 · Conqueror EUR 495 · 10+ accounts white-label on request · every feature on every plan |
|
Pricing confidence How public the numbers are |
Public, exact | Public, exact |
|
Billing Primary billing axis |
connected LinkedIn accounts plus metered credits | Monthly or yearly subscription by LinkedIn accounts and team members; every feature on every plan; AI credits per agent |
|
Variable cost Workload / overage exposure |
Medium variable cost | Medium variable cost |
|
Free tier / trial Try before you buy |
No free tierTrial
|
No free tierTrial
|
|
Buying motion Self-serve vs sales call |
Self-serve | Self-serve |
|
Pricing verified When this pricing was last checked |
Other matchups in GTM and revenue agents
Not the pairing you were after? These compare a different set of GTM and revenue agents on the same 14 capabilities.