Frunze shares practical views on cold-email infrastructure, outbound systems, and tool choices, based on building Ken.
Best for
A deliverability or email-verification vendor, not a campaign-sending platform, focused on separate domains, warmup, and verification.
Not for
A cold-email sequencing platform, CRM, or marketing suite seeking endorsement for cold outreach through its own infrastructure.
Key takeaways
Cold outreach through a main CRM can put more than email at risk.
Ken replaced Instantly, Smartlead, Clay, and Apollo in their current stack.
They call Clay impressive while questioning whether founders want to operate complex stacks.
Check first: Confirm audience demographics, recent comparable results, and rate card or partnership terms.
Main ideas
Cold outreach doesn’t belong on platforms your business already depends on · The bottleneck in outbound is the operator, not the data layer · Tools get replaced when they add operator load, including with in-house builds
Cristian Frunze builds Ken, a cold-email and sales-tooling product, and writes about outbound systems and the tools used to run them.
The available posts show why they built Ken: after trying an eight-person SDR team and a multi-tool setup, they wanted to make outbound easier to operate without requiring a dedicated GTM engineer.
Follow them for practical arguments about cold-email infrastructure, sales-tool choices, and reducing the work involved in running outbound.
Frunze connects operational advice to firsthand decisions and tradeoffs.
One post argues that cold outreach through a main CRM can put more than email at risk, and recommends separating cold-email infrastructure from the primary platform.
Another lays out Ken’s working tool stack alongside tools it replaced, making the build-versus-buy choices concrete.
The posts also complicate a simple anti-tooling message: they call Clay impressive while questioning whether founders want to operate the complex stacks its market represents.
This mix of specific operating details, candid tradeoffs, and a clear preference for simpler outbound systems gives readers material to assess rather than generic sales advice.
Cristian Frunze writes about outbound sales as an operating problem rather than a data problem.
Read more
Across their posts, the recurring argument is that cold email needs infrastructure separate from the tools a company depends on, and that the real cost of a modern sales stack is the person required to run it.
They apply that test consistently: a CRM used for cold outreach is a liability, a multi-tool enrichment stack is a tax, and a tool earns its place only if sending good email doesn’t require a specialist.
The posts support the argument with named products, current tool choices, and explanations of what was removed from the stack and why.
Cold outreach doesn’t belong on platforms your business already depends on
Frunze argues that sending cold email through a CRM such as HubSpot, Salesforce, or Close violates its terms of service and puts the whole account at risk, not just the campaign.
Read more
These platforms are built for warm email to opted-in lists and monitor volume and complaint rates, so well-written copy offers no protection.
Because the risk extends to the account and domain, the downside is losing contacts, deal history, and warm-list reach.
The stated fix is to run cold email on dedicated sending domains and sequencing infrastructure, keeping the primary platform completely separate.
The post describes an account locked with every contact, deal history, and deal note trapped inside it, with no one to call, no appeals, and no grace period.
It puts the danger zone at roughly 5,000 cold emails a month.
In a separate post, Frunze lists how to separate the infrastructure: Warmy for warming the main domain, and Plusvibe, Emailbison, and Warmy for warming cold sending domains.
The bottleneck in outbound is the operator, not the data layer
Frunze argues that complex, configuration-heavy outbound stacks assume buyers want to become RevOps architects, when most founders simply want pipeline.
Read more
Rather than treating enrichment quality as the core problem, the posts ask who operates the tooling: if decent cold email requires hiring a dedicated GTM engineer, the stack has failed the buyer. Frunze’s stated goal is to eliminate that role so anyone can send email at scale without debugging tables, waterfall logic, or sequences broken by an API update.
The account contrasts an eight-person SDR team running a full stack, which produced about 60 calls a month and was, in Frunze’s words, hated, with tearing the team down and learning enough Python to ship an MVP in two weeks.
The post presents the latter as the better trade.
It reads Clay’s reported $3.1B valuation and CapitalG investment as the market pricing in millions of companies that will hire people to operate multi-tool stacks.
Tools get replaced when they add operator load, including with in-house builds
Frunze publishes the stack as a working list of tools and their replacements.
Read more
The pattern is consolidation into fewer tools or home-built ones.
The reasoning follows from the operator argument: each additional tool is another thing to run, so a replacement that removes a step or vendor wins even when the incumbent is well regarded.
They disclose that Ken is their own product and that about half the list consists of in-house replacements, presenting that as the point of what agents make possible rather than hiding it.
Named removals and replacements include Instantly and Smartlead replaced by Ken, Apollo replaced by Prospeo plus Ken, Gamma replaced by Claude Design, Calendly by Cal.com, Loom by Cap.so, and Firecrawl and Apify replaced by an in-house scraper. The post calls Clay genuinely impressive and credits its valuation before listing it as replaced by Ken.
Across Cristian Frunze’s posts, the build-over-buy position extends from one product decision—building instead of buying Clay—to a broader operating habit of replacing paid tools across cold email, CRM, design, and scraping.
Specific, named picks instead of general recommendations
Frunze presents tool preferences as current operating choices with reasons, making the advice checkable rather than abstract.
Read more
When several credible options exist, they rank them; when relevant, they give a straightforward cost or convenience rationale and acknowledge that a category is contested. It’s the same list-driven method as the stack post, applied to individual buying decisions.
LeadMagic is named for email enrichment ahead of FindyMail and Prospeo, with the instruction to start with LeadMagic if you pick only one.
Mailtester is used for unlimited email verification and described as dirt cheap.
Twenty CRM is called open source and underrated, with the prediction that it is going to be big.
Kimi K2.6 replaced Claude Sonnet for writing and personalizing emails.
This post explains the problem Ken is meant to solve and offers Frunze's clearest statement on outbound tooling.
Read more
Newcomers should start here.
Frunze says that at 17, they were told to buy Clay.
Instead, they built an eight-person SDR team, ran a full stack and made roughly 60 calls a month—work they hated.
The post credits Clay as a real product with a $3.1B valuation, then reads that valuation as a market bet that buyers want to become RevOps architects running multi-tool stacks.
Frunze responded by tearing down the SDR team, learning enough Python to ship an MVP in two weeks, and trying to remove the question of who operates the tooling, so sending decent cold email would not require a dedicated GTM engineer.
For brands planning a campaign
A sales-tooling brand can use this as a debate or counterpoint: the claim that complexity, not data quality, is the bottleneck gives a vendor a specific position to answer, rather than another generic founder profile.
This answers the infrastructure question buyers usually ask second: what gets an account restricted, and why separation—not better copy—is the fix.
Read more
It is the most operational of the three posts.
The post argues that using a CRM such as HubSpot, Salesforce or Close for cold outreach violates the platforms’ terms and can lock an account, trapping contacts, history and deal notes with no appeals or grace periods.
Frunze says the platforms monitor volume and complaint rate, not copy quality, puts the danger zone at around 5,000 cold emails a month, and warns that the risk extends to the main domain, warm list, calendar and search visibility.
The stated fix: use dedicated domains and infrastructure for cold email, separate from the primary platform.
For brands planning a campaign
This is a natural sponsorship or co-authored follow-up for a deliverability, sequencer or email-infrastructure product. Its concrete threshold claim also opens the door to a data-backed piece that tests or qualifies the number—an opportunity for a vendor willing to publish its own benchmark.
This answers the practical question of what the toolchain looks like day to day.
Read more
It is the most concrete of the three posts and the fastest way to see which categories and products Frunze works with.
Frunze lists tools by category: cold email (Ken for campaigns, LeadMagic for enrichment, with FindyMail and Prospeo as alternatives, Mailtester for verification, and Warmy, Plusvibe and Emailbison for domain warming), sales (Twenty CRM, Client Giant, Claude Design), and content (Kimi K2.6, Hyperagent, Beehiiv, Adobe Podcast).
A second list covers tools dropped and their replacements: Instantly and Smartlead replaced by Ken, Apollo replaced by Prospeo and Ken, Calendly by Cal.com, Loom by Cap.so, Clay by Ken, and two scraper providers by an in-house scraper.
They note that about half the list consists of in-house replacements and frame that as the point of working with agents.
For brands planning a campaign
A stack teardown offers an unusually direct placement format: individual tools appear by name with a reason for inclusion or removal. A brand in one of the listed categories can propose a build-versus-buy or stack-audit collaboration; mark any placement clearly as commercial, not as Frunze's independent recommendation.
Cold email does not belong on your main CRM domain
In a post about CRM bans, Cristian Frunze argues that cold outreach through tools like HubSpot, Salesforce, or Close violates those platforms’ terms of service and puts the whole account at risk, not just the campaign.
Read more
The post describes accounts locked with contacts, deal history, and notes trapped inside, and says there is no appeal path.
Frunze identifies volume and complaint rate as triggers, putting the danger zone at roughly 5,000 cold emails a month, and says good copy does not protect you.
The recommended fix: dedicated sending domains and cold-email infrastructure, kept completely separate from the primary domain.
Sets up a specific failure scenario, addresses the reader directly, then gives the fix.
For brands planning a campaign
Strong fit for deliverability, domain, or outbound-infrastructure vendors.
A collaboration could walk through separating cold sending from the primary domain or show what a ban would cost a team.
Brands selling warm-email or CRM tools should expect the argument that their category is the wrong place for cold outreach.
Cristian Frunze shared a list of 19 tools in use, grouped into cold email, sales, and content.
Read more
The cold-email group includes Ken for full campaigns, LeadMagic for email enrichment, Mailtester for verification, and Warmy, Plusvibe, and Emailbison for domain warming.
Sales tools include Twenty CRM and Client Giant; content tools include Kimi K2.6, Hyperagent, Beehiiv, and Adobe Podcast.
The post also lists dropped tools and their replacements: Instantly and Smartlead replaced by Ken, Clay replaced by Ken, Apollo replaced by Prospeo plus Ken, and Gamma replaced by Claude Design.
Frunze notes that Ken is the poster’s own product and that roughly half the list consists of in-house replacements.
A structured, category-by-category list with an explicit dropped-tools section and plain disclosure of self-promotion.
The post mixes independent tool picks with the creator’s own product, so any brand collaboration inherits that comparison.
For brands planning a campaign
A natural format for tool-comparison or migration content: why a specific tool was dropped and what replaced it. Vendors in the cold-email, CRM, or content-tool categories can pitch inclusion in an updated stack post, but should expect the creator’s own product to appear alongside theirs and accept direct comparison.
Build versus buy: operator burden over stack complexity
Cristian Frunze describes being advised to buy Clay and instead building in-house.
Read more
The post recounts running an eight-person SDR team, getting about 60 calls a month, and disliking the work, then learning enough Python to ship an MVP in two weeks.
Frunze reads Clay’s reported $3.1B valuation and CapitalG investment as the market pricing in companies that will wire up multi-tool stacks and hire GTM engineers to run them.
They argue that most founders want pipeline, not to become a RevOps architect debugging tables and broken sequences.
The stated goal is to send good email at scale without needing a specialist to operate the tooling.
A personal before-and-after narrative: the advice received, the path taken, the result, and the broader argument.
For brands planning a campaign
The strongest fit is for products that reduce operator load: done-for-you services, simpler sequencing tools, or anything sold on “no GTM engineer required.”
Complex, configuration-heavy platforms pitched as a power upgrade run against the creator’s stated position.
A collaboration should address that objection directly.
Specific tool picks in enrichment and deliverability
Beyond the general stack list, Cristian Frunze gives concrete preferences: LeadMagic for email enrichment, named ahead of FindyMail and Prospeo as the single pick to start with; Mailtester for unlimited email verification, described as very cheap; and Warmy for warming the main domain, alongside Plusvibe and Emailbison for warming cold sending domains.
Read more
These are current operating choices, not abstract recommendations.
For brands planning a campaign
Useful for enrichment, verification, or deliverability vendors seeking a credible operator endorsement, and for brands that want a head-to-head framing. A collaboration could include a side-by-side test or a list of criteria behind the pick, fitting the creator’s list-driven style.
Cristian Frunze writes about the working parts of outbound sales: where cold email can be sent from, which tools run it and who has to operate the stack.
Read more
Start with the post warning that cold outreach sent through HubSpot, Salesforce or Close breaks those platforms' terms of service and puts contacts, deal history and the primary domain at risk; dedicated sending domains are the fix.
The 19-tool stack post serves as a running index of current choices in enrichment, verification, domain warming, CRM and content, with a dropped-tools list showing which products were replaced and by what.
The build-versus-buy post lays out the argument behind both: they describe running an eight-person SDR team for about 60 calls a month, tearing it down and shipping an in-house MVP instead, and read Clay's reported $3.1B valuation as a market bet that buyers want to become RevOps architects.
Read the three together for the position, then revisit the stack post as tools change.
Writes about outbound as one layer in a larger go-to-market system, not a standalone tactic.
Read more
Names the tools doing each job and publishes layered frameworks, with layers numbered in build order. Both treat list quality, signals and sending setup as prerequisites to copy, and point to concrete client outcomes to show why the setup works.
Vacca writes about agency client work across five parallel fronts: AI search visibility, signal-based outbound, founder content, paid ads and RevOps.
Their named results include 830 contacts at a 5.5% reply rate and $7.83M in pipeline.
Frunze writes from a product position, arguing against stacks that require a dedicated operator and documenting their own tool removals.
Vacca's unit is a multichannel program; Frunze's is a tool decision.
Also publishes long, list-shaped posts about lead-generation tools and AI connectors, mapping which products cover prospecting, enrichment, CRM and outreach.
Read more
Both argue that tools deliver value when connected into one system rather than run separately, and name vendors one by one so readers can act on the lists.
Panjuta's posts offer general inventories of Claude connectors and tool combinations for a broad audience learning to use AI at work, and direct readers to a paid community.
Frunze's lists draw on the stack at Ken, include a dropped-tools section and make a build-over-buy case.
Panjuta surveys what exists; Frunze reports what was kept.
For a campaign aimed at B2B sales-tooling practitioners, Cristian Frunze offers a small, self-selected LinkedIn audience: the profile had 3,964 followers on October 8, 2026.
Read more
The three posts make direct, specific arguments about cold-email infrastructure, CRM terms-of-service risk, and tool-by-tool stack choices.
That content may appeal to founders, operators, and outbound practitioners who make or influence tooling decisions, but the posts don’t verify who reads them.
Before committing budget, ask for recent results on comparable topics, such as tool-stack breakdowns, and for any audience survey or platform demographics covering professional role and company size.
If seniority or employer composition matters and no such data exists, treat that audience fit as unproven.
Cristian Frunze's posts do not publish sponsorship formats, a media kit, a rate card, partnership terms, or a paid-partnership disclosure.
Read more
They do show a stated commercial interest in Ken, described as their own cold-email product: one post lists Ken as what they use for full cold-email campaigns and as the replacement for Clay, Instantly, Smartlead, and Apollo in their stack; another argues that founders want pipeline rather than the work of operating tools like Clay.
For a B2B sales-tooling buyer, this is a competing commercial position, not a sponsorship offer.
Other products named in the stack, including LeadMagic, Prospeo, Mailtester, Warmy, Twenty CRM, Beehiiv, and Kimi K2.6, are presented as personal recommendations, with no indication of payment, affiliate arrangements, or sponsorship terms.
Before planning a collaboration, ask directly about scope, deliverables, disclosure wording, exclusivity around competing cold-outreach tools, and how to make an inquiry; no official inquiry route is published on this page.
A dedicated deliverability and email-verification vendor—a tool for separate sending domains, inbox warmup, and list verification, not a campaign-sending platform—is the strongest fit among the briefs tested here.
Read more
Frunze’s most operational post makes the vendor’s case directly: running cold outreach on a primary CRM domain violates the platform’s terms, volume and complaint rate trigger restrictions, and the stated fix is to keep dedicated sending domains separate from the primary platform.
Verification and warmup products make that argument actionable, and Frunze already names Mailtester for verification and Warmy, Plusvibe, and Emailbison for warming as current choices in the stack.
Because Ken is the campaign layer, not the verification or warming layer, the vendor complements their product rather than competing with it, removing the main structural objection.
The tradeoff is reach: they had 3,964 LinkedIn followers on October 8, 2026, too few for an awareness play.
Ask for audience demographics to assess seniority and buying roles.
Fit rests on relevance to a niche operator audience, not scale, so sell the campaign on that basis.
When to choose a specialist
A cold-email sequencing platform in the Instantly, Smartlead, or Clay category, or a CRM or marketing suite seeking endorsement for cold outreach through its own infrastructure, is a poor fit.
Read more
Frunze states plainly that Ken replaced Instantly, Smartlead, Clay, and Apollo in the current stack.
A collaboration would ask them to endorse the category Ken is sold against—a conflict no deliverable resolves.
For a CRM or warm-email platform, Frunze argues that using HubSpot, Salesforce, Close, or Mailchimp for cold outreach risks a locked account and endangers the primary domain, warm list, and deal history, with no appeal path.
A collaboration built on the opposite premise would be a debate, not an endorsement, and would need to be framed and priced as one.
The brief also falls short on size: 3,964 followers cannot carry a broad awareness campaign.
Ask for audience demographics before making any claim about decision-makers.
A worked campaign example
Illustrative angle for a campaign aimed at deliverability: a domain-separation teardown.
Frunze walks through one concrete setup: a clean primary domain, a separate sending domain and subdomain configured for cold outreach, verification before sending (the Mailtester slot), warmup (the Warmy slot), and the specific signs of a throttled domain.
The piece uses the thresholds they already cite—around 5,000 cold emails a month plus complaint rate—and shows what a locked CRM account costs in contacts, deal notes, and calendar access, letting readers weigh the risk against the vendor’s fix.
Agree on deliverables: one long-form post written as their operating view rather than a product recommendation; a one-page checklist hosted by the vendor for readers to use outside the article; and an explicit commercial disclosure, since existing stack posts read as personal recommendations and readers may assume the same unless told otherwise.
Settle directly with Frunze before committing whether Ken appears in the piece and in what role, exclusivity against competing verification and warmup tools, usage rights, and disclosure wording.
Judge success by checklist downloads or trial starts from a small, targeted audience, not reach.
Before budget approval, ask Frunze for a rate card, media kit, partnership contact, history of comparable sponsored work, and audience role composition.
Alex Vacca is the better choice when the campaign sells an assembled outbound system rather than one piece of infrastructure, and the buyer wants named client outcomes in the same narrative.
Read more
Vacca's posts lay out five parts of go-to-market — AI search visibility, signal-based outbound, founder and employee content, paid ads and RevOps — name the tool doing each job, and report results: 830 qualified contacts with a 5.5% reply rate for Hemlane, $7.83M in qualified pipeline and $1.52M closed-won for AirOps with no new SDR hires, and 3,655 accounts in a self-refreshing Salesforce database for Aircall. A vendor whose pitch is that the whole stack has to work together gets more usable proof from Vacca than from Frunze, whose posts concentrate on the cold-sending infrastructure decision.
Those results come from agency delivery, so the collaboration reads as a services case study rather than a founder explaining the tool they built.
Vacca's posts also recommend competing products by name — Clay, Instantly, Prospeo.io, lemlist and Apify among them — so a vendor in any of those slots inherits an endorsement problem.
A verification or warmup vendor appears as one component inside a five-layer system, not the argument of the piece.
Frunze offers a narrower trade: one tested claim, that cold sending must be kept off the primary domain, made from the operator's side of the problem.
Denis Panjuta is the better choice when the objective is tool discovery and integration sign-ups rather than credibility with people who run cold email.
Read more
Panjuta's posts are built for saves and shares — fifteen Claude connectors for a lead-generation workflow, twenty Claude tool combinations — and each ends with an invitation to join a Skool community, so a collaboration can ride a distribution routine the creator already runs weekly. A vendor launching a connector, integration or self-serve product, and willing to be described in one line inside a long list, fits that format directly.
The cost is specificity.
The posts explain what each connected tool is for in general terms; they do not test sending domains, verification, complaint rates or what breaks when volume rises, and report no client results or named accounts.
Frunze's posts do not show the hands-on cold-email operations a buyer in this category is trying to borrow, and the list format is easy for competitors to replicate.
Where Frunze supplies a tested operating argument, Denis Panjuta supplies reach into adjacent AI-tooling interest without posts reporting operational results.
Brigitta Ruha is the better choice when the campaign is aimed at people who set GTM budgets and headcount by revenue stage.
Read more
The ARR-stage post maps rules, roles and tool spend from $0–1M to $25M+ — when to add the first SDR, LinkedIn Ads at roughly $8–15k a month, account research under $0.10 per account — which lets a vendor attach its product to a planning decision at a named stage instead of to a sending practice. The AI lead-generation post adds a workflow a reader can copy: choose buying signals first, turn qualification into a Claude Code skill with a two-signal rule, and teach the model a house writing style with real examples and banned words.
The tradeoff is depth.
The content is shaped as save-and-repost lists, the full rule set sits in an image rather than in the post text, and the tool recommendations span many categories, so a single vendor becomes one line among dozens.
Ask Ruha for evidence of domain setup, verification or deliverability work, and client outcomes.
Frunze gives a narrow but tested argument about infrastructure; Ruha gives a planning frame that places spend, at the cost of showing that the tool works.
Cristian Frunze's three LinkedIn posts say that running cold outreach through a CRM such as HubSpot, Salesforce, or Close risks the account and its contacts and deal history; list 19 tools in the Ken stack with the tools each one replaced; and describe building Ken instead of buying Clay.
The posts carry no publication dates.
LinkedIn was the only channel available, so no interviews, videos, newsletters, or longer-form writing inform this page.
The LinkedIn account displayed 3,964 followers on 8 October 2026, and no audience demographics were available, so this page does not claim a specific reader seniority or buying role.
Similar-creator comparisons draw on two LinkedIn posts each by Alex Vacca, Denis Panjuta, and Brigitta Ruha, published between 30 September and 8 October 2026.
Several gaps remain open and would change a reader's decision if filled: Frunze's posts do not state their formal role or ownership at Ken; no partnership contact, media kit, rate card, or sponsored-work history appeared; and the posts do not establish whether books, shows, or earlier work also explain searches for the name.
These findings reflect a small, undated collection of posts, not everything Frunze has said.
