Kasten Culture Deck Breakdown (2018, 19 Slides)

What this deck actually is

Nineteen slides, 960 x 540 points — 16:9 , built and exported in Microsoft PowerPoint 2013 on 31 January 2018 at 04:41 UTC . No author field, no company address, no confidentiality mark. The title slide reads "Kasten Culture Deck" with a superscript v0.1 and one line underneath: "This deck, just like the company, will always be a work in progress. Feedback welcome and appreciated!"

Say the obvious thing first: this is not a fundraising deck. There is no problem slide, no market, no product, no traction, no team page, no ask. Kasten does not appear in it as a business — the word "Kubernetes" never appears, the word "backup" never appears, the product K10 is never named, and no founder is named. It is an internal values document that happens to have been published.

It still belongs in this series, and here is why. Founders increasingly send a culture deck alongside the pitch deck, and later-stage investors read one in diligence — not for the values themselves but for what the document reveals about how the company makes decisions when nobody is watching. A culture deck is a management-quality artifact. It can be torn down on exactly the same terms as a pitch deck: is every claim specific, is it falsifiable, does it survive contact with a hard case, and does the author know what they are trading away.

Kasten, for context outside the file: a Kubernetes data-management company based in Los Altos, California, co-founded and led by Niraj Tolia, which emerged from stealth at KubeCon on 5 December 2017 with the K10 platform for backup, disaster recovery and application mobility on containerised workloads. This deck is dated eight weeks after that launch, when the team was small enough that the entire values system fit on nineteen slides of default PowerPoint theme with no images except five logos on the last page.

Slide-by-slide walkthrough

Slide 1 — Cover

The Kasten wordmark, "Kasten Culture Deck", a superscript v0.1 , and the work-in-progress line.

The version number is the single best design decision in the document. Values decks fail in a predictable way: they are written once during a burst of enthusiasm, posted to the wiki, and then quietly contradicted by the first hard decision. Numbering the file v0.1 does three things at once. It admits the document is provisional. It creates an expectation that a v0.2 will exist. And it makes the deck safe to argue with — an employee who disagrees with a bullet is disagreeing with a draft rather than attacking the company's scripture.

What the cover does not carry is an owner or a date. Nowhere in nineteen slides does the deck say who wrote it or when. The only date in the file is the PDF creation timestamp. For a document whose whole premise is "feedback welcome," there is no name to send the feedback to and no way for a reader eighteen months later to know whether they are holding the current version. If you copy one thing from this deck, copy the version number — and then add the two fields it forgot.

Slide 2 — "Culture is a shared outlook"

This is a definition slide and it is doing more work than it looks. "Shared outlook" is a deliberately narrow claim: not shared values, not shared personality, not shared background — a shared way of looking at problems. It sets up everything that follows, because a set of behavioural rules only coheres if you accept that culture is about perception rather than affinity. It also quietly rules out the "culture fit" hiring trap, where teams select for people they would like to have a beer with. Nothing later in the deck cashes that in — there is no hiring slide — but the framing is right.

Slide 3 — "Culture is Strategy" — Jim Collins

The small one is provenance. The famous line in this territory is "culture eats strategy for breakfast", popularly credited to Peter Drucker — an attribution Quote Investigator could not trace to anything Drucker wrote or said , with the earliest print citations from 2002 crediting Ford's Mark Fields. "Culture is Strategy" is a different claim, and Collins is associated with it as an explicit rebuttal of the Drucker line rather than as its author: the sourcing runs through secondhand accounts of talks, not a Collins book or essay. Collins's own written argument, in Good to Great , is about a culture of discipline making strategy executable. So the slide is not wrong so much as unsupportable — it puts a named authority behind wording that cannot be traced to anything he published. Three of the four quotations in this deck are attributed; this is the one a reader cannot check, and once they fail to find it they will read the rest of the file slightly more sceptically.

The larger problem is that "Culture is Strategy" is the opposite of what the deck then does. The claim is that culture replaces strategy as the driver of outcomes. But the next slide justifies culture instrumentally — as a way to outcompete and build a profitable company — which is culture serving strategy. Slide 3 and slide 4 argue for different things, and slide 4 is the argument the deck actually makes.

Slide 4 — "A Great Culture"

Two lines: enables us to outcompete others; helps build a successful and profitable company.

This is the slide that separates a serious values document from a poster. Most culture decks justify culture in terms of how it feels to work somewhere. This one justifies it commercially, in two sentences, and names profitability explicitly — an unusual word for a venture-backed infrastructure startup in 2018, when the prevailing norm was growth at any burn. It also sets up the Frugality value eight slides later, so the deck's economics are internally consistent even where its philosophy is not.

The weakness is that both lines are unfalsifiable as written. "Enables us to outcompete others" is a hypothesis with no test attached. A stronger version names the specific competitive advantage the culture is supposed to produce — faster decisions, lower coordination cost, better retention of senior engineers — because that is a claim you can check a year later.

Slide 5 — "Values"

Four lines: values define and exemplify our culture; define our individual behavior; attract amazing coworkers; and are "used to promote, reward, and part."

That last line is the best sentence in the deck. "Part" means fired. Stating on slide 5 that the values will be used as the basis for promotion, compensation and termination is what gives every subsequent bullet its teeth — without it, the seven values are a wish list, and with it they are a performance rubric. Netflix's deck made the same move and it is the reason that document mattered while a hundred imitators did not.

Having made the claim, the deck never operationalises it. There is no promotion ladder, no example of a values-based performance conversation, no description of how a values violation is raised or adjudicated, and no statement of who decides. A rule that determines whether someone keeps their job needs a process attached to it, and the process is the part employees actually need. This is the single largest gap in the document.

Slide 6 — Our Core Values

Seven bullets: Transparency; Trust and Responsibility; Independence and Autonomy; Focus; Simplicity; Frugality; Being Impact and Result-Driven.

Seven is too many, and the deck's own structure shows why. Trust and Responsibility and Independence and Autonomy are close enough that their detail slides restate each other — one says trust "increases autonomy," the other says to use "the trust placed in you." Focus and Simplicity are two applications of the same discipline, one pointed at the market and one at the codebase. Two of the seven are compound values joined by "and," which means the real count is closer to nine.

The practical test of a values list is whether an employee can recall it in a meeting where a decision is being made. Three to five single-word values survive that test. Seven, two of them compound and one of them a seven-word phrase ("Being Impact and Result-Driven"), does not. A tighter version would collapse the two trust/autonomy values into one and fold Simplicity under Focus, leaving five: Transparency, Autonomy, Focus, Frugality, Impact.

Slides 7 and 8 — Transparency

A quotation slide — "Sunlight is the best disinfectant" , attributed to Louis D. Brandeis — followed by five bullets: applies to everyone on the team; share wherever possible and legal "and even when it hurts!" ; encourages consensus but does not dictate a democracy; implies clear and concise communication; and strongly encourages public channels (Slack, Wiki, mailing lists) over 1:1 messages.

This is the strongest value in the deck because it is the only one that translates into an observable daily behaviour. "Public channels over DMs" is a rule you can audit — you can look at a Slack workspace and see whether it is true. Everything else in the document describes an attitude; this describes an action.

Two clauses show unusual care. "Wherever possible and legal" concedes the limit up front, which matters at a company that will handle customer data and, eventually, an acquisition process. "Encourages consensus but does not dictate a democracy" pre-empts the standard failure mode of transparency cultures, where open information becomes an expectation of a vote on every decision.

What is missing is the hard case. Transparency decks are easy to write and hard to keep, and the moments that test them are always the same three: runway and burn, compensation bands, and departures. The deck says "even when it hurts" and never says whether that includes telling the team the company has four months of cash. Naming one of those three cases explicitly would have made this value credible rather than merely well written.

Slide 9 — Trust and Responsibility

Five bullets: use good judgement in everything you do; reduces friction and process; lets us depend on each other and increases autonomy and job satisfaction; be reasonable with expenses and "treat company's money like your own" ; and "do take time off to recharge and don't worry about sick time or needing to take care of personal stuff."

The lineage here is obvious — "use good judgement" is a direct descendant of Netflix's expense policy — and the adaptation is sensible for a company of this size. The time-off bullet is written unusually warmly for a deck that is otherwise clipped, and the phrasing "do take time off" is imperative rather than permissive, which is the correct framing: unlimited-PTO policies reliably reduce leave taken unless leadership pushes in the other direction.

The bullet is also the deck's clearest unforced omission. It describes a discretionary time-off arrangement without ever naming it, setting a floor, or saying who approves. Unlimited PTO with no minimum and no manager modelling is the version of this policy that fails, and nothing in the file guards against it.

Note also that "be reasonable with expenses" duplicates Frugality four slides later. In a seven-value list, spending a bullet twice on the same instruction is evidence for the consolidation argument above.

Slide 10 — Independence and Autonomy

Three bullets: use the trust placed in you to Do The Right Thing ™ ; be biased towards execution; and — the good one — "no matter where it might be, it is your duty to speak up if you see problems and, even better, see if you can fix it."

That third bullet is a genuine engineering-culture rule. "No matter where it might be" removes the team boundary as an excuse, and "see if you can fix it" converts a complaint into an action. It is the closest the deck comes to describing how work actually crosses org lines.

The trademark symbol on Do The Right Thing ™ is a joke, and it is the wrong joke in the wrong place. This is the slide where an employee facing an ambiguous call is supposed to find guidance; the trademark gag signals that the authors know the phrase is a platitude and are winking at it rather than replacing it. "The right thing" is exactly the term that needs a worked example — one paragraph describing a real decision someone made and why it was right would have carried more weight than the other two bullets combined.

Slide 11 — Focus

Three bullets and a sub-bullet: commit to the mission even though the temptation will be strong; build for the customers we are targeting and don't get distracted; "we should not sell to customers we cannot satisfy" ; with the caveat that engaging customers to guide the roadmap and "coverge" later is encouraged.

"We should not sell to customers we cannot satisfy" is a remarkable line for a first-year startup to put in writing, because it is a rule that costs money in the short run. Early-stage infrastructure companies die of exactly this — a large logo signs, demands a feature set the product cannot support, and the roadmap belongs to them for eighteen months. Committing in advance to walking away from that revenue is a real constraint, and it is the only value in the deck that is expensive to keep.

Two problems. First, the deck instructs the reader to "commit to our mission" and never states the mission anywhere in nineteen slides. A values document that references a mission it does not contain assumes a reader who already has the context — fine for a five-person team, useless as the onboarding artifact it claims to be. Second, "coverge" is a typo for "converge," and it survived into the published version of a deck that lists Simplicity and "clear and concise communication" among its values.

Slide 12 — Simplicity

An unattributed quotation — "Everything should be made as simple as possible, but no simpler" — then: we simplify a complex world for customers and should not ignore our own house; reduce complexity in product, processes and teams, and refactor everything when complexity seeps in; complexity is the bane of growth; build for the long term.

Note the attribution inconsistency. Brandeis is credited on slide 7, Collins on slide 3 and Netflix on slide 17, but the Einstein-adjacent line here — itself a paraphrase of something Einstein did not quite say — runs bare. If the deck is going to cite, it should cite consistently; a mixed pattern reads as carelessness in a document that a new hire is meant to treat as authoritative.

The content is strong on one axis and silent on the other. Extending simplicity from the product to processes and teams is the right generalisation, and "refactor everything when complexity seeps in" gives engineers explicit licence to spend time on cleanup, which is usually the first thing a young company defunds. But "build for the long-term" sits directly against "be biased towards execution" two slides earlier, and the deck never says how to resolve them. That tension — ship now versus build to last — is the single most common argument inside an infrastructure engineering team, and a values document earns its keep precisely by saying which side wins by default.

Slide 13 — Frugality

Two bullets: money is finite and we want to control our destiny as much as possible; invest in people, tools, and the things that will make us successful.

This is the thinnest slide in the deck, and the first bullet is nonetheless the most strategically literate sentence in it. "Control our destiny" is the honest reason for capital efficiency at a venture-backed company: every dollar not spent is a month of runway, and runway is negotiating leverage in the next round and the terms of an exit. Saying that out loud to employees, rather than framing frugality as a virtue, respects the reader.

The second bullet then cancels the first. "Invest in people, tools, and the things that will make us successful" is an exception broad enough to cover any purchase anyone wants to make, which leaves the value with no operative content. Frugality needs a boundary to mean anything — a spend threshold that needs a second pair of eyes, a stance on travel, a default on tooling — and there is none. Compare it with the Transparency slide, which names actual Slack channels. Two bullets, one of which is a blanket exemption, is not a value; it is a sentiment.

Slide 14 — Being Impact and Result-Driven

Six bullets: we don't measure time in your seat — but "as a local company, we do expect you to be there for your team during reasonable working hours" ; we measure the impact you make on our plans; we value people who are agile and biased towards execution, because as a startup no plan survives contact with the customer; we value a strong sense of ownership; and we value people who are humble and inquisitive .

The first bullet contains the deck's sharpest internal contradiction, and it is a useful one. "We don't measure the time you spend in your seat" is a results-only claim; the immediate sub-bullet reinstates an attendance expectation and adds a colocation requirement ("as a local company"). Written in January 2018 this was simply an honest description of a small in-person startup, and the qualifier is more candid than the frictionless remote-first language that was fashionable at the time. Read now, it is a clean illustration of why results-only rhetoric usually survives only until someone asks where you sit. If a company wants both, it should say so directly — we measure outcomes, and we work in the office because the team is small enough that proximity is worth more than flexibility — rather than asserting the first and retracting it in a nested bullet.

"Humble and inquisitive" is the deck's implicit hiring bar, and it is buried as the last sub-bullet of the last value. It is also the only place in nineteen slides that describes the kind of person the company wants to hire, which is a strange allocation for a document whose slide 5 promised the values would be used to "promote, reward, and part."

Slides 15 to 18 — Teams and closing

Slide 15 is a one-line transition: "We will succeed because we, as a team, live these values." Slide 16 adds a team section that reads like a late insertion — value 10x teams via common development standards and code hygiene, and a culture that onboards and develops members, with a parenthetical noting this is "a common thread in our culture decks," plural, referring to documents this file does not contain. It closes with the hard line: "We will not tolerate disruptive personalities that poison teams."

That is the no-brilliant-jerks rule, and it is stated more plainly than most companies manage. It is also the second promise in the deck with no mechanism behind it. Who decides that a personality is disruptive, how the person is told, and what happens between the first conversation and the exit are all unaddressed, and those are the questions that determine whether the rule is real.

Slide 17 is the deck's only diagram: three chevrons reading Customer → Team → Self , under the Netflix quotation "We're a team, not a family," with three explanatory lines — customers first builds a successful company, strong teams rather than individuals produce success, and customer and team success will naturally drive individual success.

The priority ordering is the most useful artifact in the file, because it is the only thing here that resolves a conflict. Values lists tell you what to care about; a ranking tells you what to do when two things you care about collide. Customer over team over self is a decision rule, and it is the kind of statement you can hold a manager to.

The word doing the most work is "naturally." Individual success following from customer and team success is an assertion, and the promotion and compensation system is what makes it true or false. Slide 5 said values would be used to promote and reward; slide 17 says individual success follows automatically. Neither slide describes the mechanism connecting them.

Slide 18 closes on "We (and that means you) are all torch bearers for our culture," and slide 19 credits five companies by logo — Twilio, HubSpot, Netflix, Asana and one further mark — under "This Deck Stands on the Shoulder of Giants." Crediting the sources of a culture deck is rarer than it should be and entirely to the authors' credit; it also tells a candidate precisely which lineage they are joining.

What is missing

Read as an onboarding document rather than a manifesto, the omissions matter more than the contents:

The mission. Slide 11 instructs the reader to commit to it. Nineteen slides never state it. · Any mechanism. Promote, reward, part; no process. Do not tolerate poison; no process. Speak up about problems; no channel named. · The hiring bar. Two adjectives in a sub-bullet, no interview loop, no description of how values are assessed in an interview. · Compensation philosophy. A deck that names Frugality and promises values-based reward says nothing about how pay is set — the omission most likely to be noticed by the employee reading it. · Feedback and conflict. A transparency-first culture with no stated method for giving hard feedback or resolving disputes leaves its most load-bearing value unsupported. · Stories. Not one example, not one real decision, not one named person. Values are learned from cases, and there are none. · Inclusion. Absent entirely, in a deck that spends four slides on who belongs on the team.

What to copy

The version number. "v0.1" plus "feedback welcome" converts a decree into a draft and makes the document survivable. Add an owner and a date, which this one forgot.

"Used to promote, reward, and part." Say what the values are for. Everything downstream inherits its seriousness from that sentence.

A priority ordering. Customer → Team → Self resolves conflicts in a way a list of seven nouns never can. If you write only one slide, write this one.

One behavioural rule per value. "Public channels over 1:1 messages" is the model. It is specific, it is observable, and it is the only bullet in the deck you could verify from the outside. Frugality, which got a blanket exemption instead, is what a value looks like without one.

The verdict

As a first-year values document from a small infrastructure team, this is above average and unusually honest — it names profitability as the goal, admits the office expectation instead of pretending otherwise, credits the decks it borrowed from, and numbers itself v0.1. As an operating document it is unfinished in the specific way its own version number predicts: it makes four strong promises — values decide promotion and termination, transparency even when it hurts, we won't sell to customers we can't satisfy, we won't tolerate people who poison teams — and attaches a mechanism to none of them.

The lesson for founders writing their own generalises straight back to pitch decks. A claim without a mechanism is a mood. "We will part with people who don't live the values" and "we will reach $10M ARR" fail on the same axis when neither says how. The Transparency slide names actual Slack channels and is therefore the only value in the deck a stranger could check. Write every value — and every slide of your pitch deck — so that a stranger could check it.