The SaaS Org Chart
sacks.substack.com
sacks.substack.com
--
CMO (15)
Director, Product Marketing, plus 3 PMMs (4)
Director, Demand Gen, plus 3 marketers (4)
Director, Sales Enablement, plus 2 marketers (3)
Director, Brand Marketing (3)
PR & Analyst Relations (1-2)
Events/Community (1)
--
What I posit is that Product Marketing is useless as a position and should be eliminated most of the time.
The role ends up being: - Jane Doe is the Product Marketer for product X. She's responsible for growing it as much as possible. - She works with Anne X. in demand gen to make ads to promote the product to new customers. - She works with Hillary Y. in email marketing to cross promote to existing customers - She works with tech to do landing page/funnel testing.
However what I find is that the role just adds friction and useless tests to show value. Let the subject matter experts (in demand gen, email, website testing) create the plan for each product holistically. Product Marketers are dead weight 80% of the time.
I would only say this anonymously but its true. Source: I'm a marketing consultant that has worked with dozens of fast growing SaaS companies.
In an ideal world, demand gen and content already have messaging guidance, target segments and titles and their pain points from product marketing (and field marketing for regional guidance), and can just run with it except for perhaps a once over from product marketing on some initial copy for accuracy or tone. Otherwise, product marketing has their hands full researching customers along side PMs, writing more specialized content that’s harder to outsource, or building and delivering sales enablement. Essentially, they’re making sure you’re targeting the right folks at the right time, prepping sales dev to receive those folks, and working with sales on scalable motions to close them. There’s no reason product marketing should be weighing in on the nuances of how demand gen is being executed.
Short version: there’s lots of facets to marketing, a good org keeps people focused where they’re needed based on their expertise.
Part 0: ask yourself if product marketing is actually what you want to do. It is (as you can see) a widely misunderstood role. There may be some other facet of marketing that’s more appealing - demand-gen is a challenging area on its own, as is partner GTM, or branding. If you’re looking for strategy, you may really be searching for a corp dev, product strategy, market research, or chief of staff role. No way to really know except to talk to folks in those roles and check out some job descriptions in your target industry (seriously, the only thing that really galls me when I get asked for career advice is when I inquire “consumer marketing or enterprise / B2B?” and the person shrugs…these are wildly different marketing roles and I wouldn’t dream of weighing in on a consumer marketing job search).
Assuming you’re still interested in B2B product marketing: Part 1 is get as close to the customer as you can. The goal is to know the market (this includes competitors) and the customer as well as possible. Every single product marketer I know is sitting in on sales or customer success calls, doing research interviews with existing customers, investigating prospects we lost, and talking to people for whom the product is never even under consideration. If in a time crunch, sales people - particularly sales engineers and your top sellers - will do, but eventually you want to follow that up with more robust research designed with product management and demand-gen, as sales will only capture bottom of funnel info secondhand for you. In your role, it sounds like you can ask for 30 mins to talk to a sales leader or top seller under the guise of serving the customer better as part of product ops or content, and go from there.
Part 2 is build some content that demonstrates specialized knowledge. The goal here is to take that knowledge from Part 1 and demonstrate that you can actually turn it into something that can support other teams, at scale. The scale part is important - sales or content marketing can deliver content, too, but normally they’re looking to product marketing for anchor content to customize for a particular customer, or a webinar, or an article. No, creating this stuff is not ‘strategy’, but if you didn’t come from sales, you’re already fighting a credibility gap, and no one likes working with a product marketer who can’t make a deck, write a decently technical white paper or blog post, assemble a messaging document, or design and build a sales enablement curriculum. It’s best if you can demo your own product - and the less technical it is, the more you’ll be expected to do it. For example, analytics or productivity solutions would likely be within a non-technical product marketer’s capabilities, while something like a managed database or DevOps tools requiring integrations with a wide set of solutions means being able to work with a PM, technical marketing, or sales engineer. Check out a release blog or an overview video from a product you’re interested in doing marketing for to get an idea of what to expect.
The good news is if you have what I think is required of strategy consultants, you’re probably already a strong writer, good at boiling down market analyses down to the essentials depending on audience, and experienced with building and speaking to a good storytelling deck. Hiring managers will ask for work samples.
PM's role is not to run campaigns, either brand or 'demand gen'.
That's regular operational marketing.
Product Marketing's job is to define the Product for the appropriate segments in terms of messaging, features, pricing. Then, outbound, they will provide the overall strategy for the various campaigns.
So part 1: They know the market well. Who uses it, why, what their needs are. They'll sit in on sales calls i.e. should be one of the primary conduits into the company from them. They'll do market research, understanding pricing trends. They'll generally do some kind of segmentation, even if crude.
Part 2: Direct product activities in this regard i.e. 'Each nation has different regulatory apparatus re: this feature so we need to break it down'. Or, even though our software has nothing to do with security, this specific issue is a problem and we need to have XYZ feature even if it's perfunctory.
Part 3: For everything outbound, they'll be setting the messaging and the strategy. 'What groups are we targeting' and 'What those groups need to hear'. Probably putting together the 'baseline copy' and materials.
The 'Marketing Ops' shouldn't be making a whole lot of those decisions on their own. They should be taking that from the PM and then deciding which channels to push their budget into.
There's quite a lot of overlap in terms of customer relationship and product features, as the Product Manager also does this, but I usually think of PM's as satisfying satisfying specific customer requests, having a broader vision for the architecture that goes beyond any segment needs.
The Product Marketing team is the one that says 'We'll be selling in 5 tranches with these features at this price'. The Product Manager cares less about that deliniation and more about the product as whole including internal architecture.
Market segmentation is definitely marketing. Usually pricing is as well.
Product Roadmap and direct links with Engineering is Product.
I suspect these particular areas in which you feel they are at deadweight may be missing much of their focus or you've experienced an organization that is misusing this role.
So here's the thing - I get the desired purpose of the role of centralizing GTM strategy. But in practice product managers are too far removed from the customer to be useful 80% of the time.
Basically I find that tactics that work for a given business should drive strategy more than vise-versa. This is a very unpopular opinion for sure.
In the end, messaging and positioning don't exist in a vacuum. The market responds to what it responds to. But because product marketers are too far removed from the stress of actually making messaging turn into revenue, they just don't have the skillset to be useful.
I wish product marketers could make positioning that actually works. But what happens is they make positioning that doesn't appeal to customers, and sales and demand gen pick up the slack and now have two jobs - to actually use messaging that gets results, and to make a middle manager happy with some data that helps them make useless reports upstream.
From the other side of the B2B marketing fence, I see demand gen burning loads of cash on "high converting", highly "tested" ads that look great in reports, except that those leads never turn into customers, because the PPC marketers never track their revenue that far, and those lowest common denominator leads aren't a great fit for many B2B products anyway.
I suspect that if your argument is "Well, they should build the product that those leads want" I can see why (a) that view would indeed be unpopular, and (b) why you would be in conflict with product marketers who want leads that match the positioning, not positioning that matches the leads.
When all you've got is a hammer...
I like how he limits the number of C-level titles in the early stages to just a CEO and CTO. It's a common mistake for startup founders to hand out too many C-level titles to their cofounders and early hires. This causes problems when the company outgrows some (or most) of those people. If you need to bring in a more experienced CMO, CRO, CSO, or other C-level, your only options are to demote the existing C-level and replace them with a new boss or to arrange their exit from the company. Neither are good looks, and neither make people happy.
Much better to start with Director and VP level titles in the early days. You can either bring in a C-level above them when the time is right, or promote them to C-level if they grow into the role.
That said, most SaaS I've worked with would want more people on backend/devops at the earl stages. Having 3 people total handle combined devops and backend development is a big risk, especially if any of them want to go on vacation or leave for other jobs. Downtime and server outages are not a good look for a growing SaaS, so don't skimp on people who can keep it up and running.
The thing that I thought was broken about it is the bread first strategy. I would think it would be better to focus on a segment of the market until they start getting momentum. Many products we see that are targeted at smaller companies are going to be wholly unacceptable to large enterprises.
* This feels way too hierarchical for the size -- there's significant distance from many employees and the CEO even at small sizes.
* Could not disagree more with putting HR and Ops under the CFO (even in a 50-person company). That's a surefire way to have a neglected HR/people function.
What does seem a little surprising is how many “chiefs” the CEO has reporting to them. In my experience with companies around 50 people, things actually tend to not be sliced into so many pieces, but rather there are a few larger slices that are functionally heterogeneous.
Basically have someone own:
1. The product - CTO
2. The customer - (Could be a director or VP of marketing/sales/rev)
3. The company - CEO
Anything beyond those three execs at or around an A round is nuts to me.
I think a lot of people underestimate just how much you can outsource at that size. There almost no reason to have a COO when you’re less than 100 people. Same with a CFO. Either of those hires don’t become strategic until you’re doing $10+m in rev. Until then, use other SaaS services for payroll, book keeping, tax savings (MainStreet), etc. These take little effort use and can be setup by one of the founders in less than a day.
This is HN I suppose, so not entirely surprising.
I still think you're going to be light with 100 people and no dedicated VP of Product or VP of Customer Success.
But not absurdly so
One slight contention here is who owns the keeping part. I’ve seen that under both sales and product, and don’t have strong convictions either way. I think it mostly depends on the type of product you’re selling.
There’s two issues then - where do you put HR/People Ops and then effectively CFO needs direct communication with HR to properly resource the org.
A lot of places have combined that work into same stack/JS and call it "frontend".
but then having vc money contributes to decisions most of us, will never understand.
By my count all the non-front end devs do back end, so Client Applications (2), Core Services/Platform (1) & Dev Ops/Infra (1)
DevOps is a mindset of bridging operational and development teams to improve the delivery of work done to a place where it can deliver value.
If the author means an “SRE”, he should say that - though the discussion about “you write it, you run it” can be left for a different day.
I don't agree with any of that, but that's how the cookie has crumbled.
In my case, a back-end team of 4 owned everything "non-app" while about 10 engineers owned the various front-ends.
- $25K ARR per person with >40 employees.
- $50K ARR per person with >100 employees.
- $100K ARR per person with >1,000 employees.
There are plenty of independent SaaS businesses that make it past $200K ARR per person with teams of <10.
Which is great for companies aiming to stay small and self-bootstrapped. There's nothing wrong with this model for founders who want to stay independent.
The VC alternative allows companies to grow faster than organic revenue growth might allow. It can take a long time for a 10-person company with $2mm ARR to grow to $2.2mm ARR so they can hire the next person they need. Alternatively, they can take VC money and hire the next 100 people they need without waiting for revenue to get there first.
One model isn't inherently better than the other. The important thing is for founders to decide which type of company they want to be and stick to it. A founder who wants to grow organically is going to have a bad time if they take money from investors who want growth. A founder who wants to grow as fast as possible is going to have a bad time if they're constantly stuck waiting for organic growth to let them make the next hire.
The point is if you’re in a VC heavy space, you often have no choice but to dance along with the music.
This is what I understood it to mean: a startup grows marketshare rapidly with VC money infusion in an easy money environment. But when credit dries up like it did in 2008, the startup could be left hanging as VCs might not fund the next round of their previously star startups as they want to deploy resources to their absolutely top-rung startups. So, even good startups can be left stranded.
That's the feeling I have. Would like to know if my thinking is on the right track.
How many of them become billion dollar companies in less than a decade though?
In this case I think the ARR per person is probably more of a proxy for "are you growing fast enough?" rather than the number itself mattering.
EDIT: Oh per person, I missed that part. Still though, that puts us at over $50k ARR per person at an org size of 2. However it also makes the comparison completely pointless.
But you don't now need to introduce additional abstractions when you go from 50 to 125. Why introduce a VP of engineering? Can the CTO not handle 3 extra director reports (4 if you add a "director" of QA with 2 reports?) Come on. Who is going to come on to be the "VP of engineering" for a startup, yet not have/want oversight for Security, Analytics, and Infrastructure?
Also in the 50 person startup, the product lead with 5 reports is a "Director" but a product lead with 11 reports is a "VP"? Get outta here, this is silly. It's the exact same role. It is NOT more scope or oversight.
There should be orders of magnitude size growths before you start overthinking the difference between a director and a VP, or introducing new layers.
"technical leader" for a 0-50 org is VERY different job than >100 org.
The CTO is probably very good at 0-1 type of start up work, and probably a cofounder and inexperienced eng leader.
But a experienced VP of eng is required to build teams and process later on. But usually those VP of eng are not good at building startups from the ground up.
So the solution here is either: - fire the CTO co-founder and replace them with a VP of eng type of talents - hire VP of eng and let the cofounder CTO focus on other works
I have seen both happened before.
- tech manager != tech leader, different skills
- CTO is like a Chief Science Officer at a biotech or seniormost Principal/Fellow Engineer at a bigger tech co: ~No boss and ~no reports. The more boring the core tech & scale, the less important to have.
- VP.E might do more on internal people processes, like run overall technical recruiting, while CTO stays involved like helping close candidates & joins bigger sales calls, but is overall more peripheral to the people management. They are not fully removed -- they might focus more on say code culture due to being an experienced technologist & deeply familiar with the core code.
- Both VP.E & CTO might say fix little bugs to make sure code culture is sane and help add productivity tools to the stack, but CTO will more likely run major arch advances & new system prototypes, while VP.E will help trains run on time by seeing ahead around coordination issues. Both need to be leaders (consensus building, ...), but one leads the company's technology and reads deeply on technology, and the other, its technologists, and reads deeply on management.
Lacking a good VP.E. leads to inability to ship: cannot execute tactically. Lacking a good CTO leads to a stagnant technology org: cannot execute strategically.
I suspect this changes at bigger scales, where my cynical side suspects, at most orgs, corp dev (M&A) > Office of CTO for technical leadership :(
IT and Eng. roles will probably be sub divided.
Instead of a VP it might be possible to have just a manager or director, but it depends on the seniority of the people you're bringing in.
The CTO's job is not just downward focus, they have strategic things to to.
A few titles could be mushed around but it's just about right.
For the VP product with only 11 reports - that's a staff position, it doesn't have to imply a lot of headcount.
You might have a Colonel in charge of an entire Regiment of Soldiers, but another Colonel in charge of a much smaller detachment of Medics, or Intel, Communications, or even just a Chief of Staff etc..
Aside from handing out VP roles to possibly people that are too young or inexperienced, the 'tree' itself is just about right.
The VP has 28 - the vast majority of the org.
This role is redundant at this size, and wasteful.
Just have one technical leader who maanged SIX directors (and 4 IC's, and it is still a reasonable size team. What is the VP of Eng providing here that the CTO can't do?
BTW, arguably this is still true at the Series C 400-sized company. The technical org is 114, but the VP of Eng managed 92 of them.
Consider that at 16 direct reports, 8 hours a week is spent just having 1:1s
In my experience much of what gets discussed 1:1s should be discussed with the larger group during retro times. This can be setup for anonymous feedback if that makes people more comfortable(it also comes with benefits anyone involved so you can "read the room").
Coaching can work well with a rotating peer coaching setup.
Now you can get 20+ people with leads reporting to a single person who owns the employee relationships. Dump all the useless middle management.
Really, you only have about five roles:
- make the service - run the service - market the service - accept money for the service - support the service
and a little later, talk to a pro about taxes.
For example, you don't need any of the management jobs until you've hired enough people that you have to manage them. You don't need to do sales and customer management jobs when you have no sales and no customers. Etc.
Of the job titles that remain, these are all specializations. The projection of the work that you'll be doing onto any of the roles will be so small, that there's no point in even identifying it. E.g. you're not really doing "Dev Ops/Infra" if your project is running on a single server in whichever cloud you had the most free credits for.
For a indie one-person SaaS, this list is useful only as an identification of broad areas of concern. I wouldn't even try to read anything from relative number of any type of positions - jobs don't scale at the same rate. E.g. sales workload may grow faster than dev workload in B2C, the opposite in B2B.
Next time may just stay a 1-man show, tempeting for sure.
If you want to go indie you can make similar choices to Basecamp (read It Doesn’t Have to Be Crazy at Work) and stay lean and small.
I also find way more insightful how you get from x founders to 50 employees than from then on.
I'm sure this seems obvious to some, but I'm not enough of a business/finance person to have really thought about this before. It helps crystallize some dynamics between, for example, my current employer and my immediately former employer; the former is able to pay dramatically more (around 150% on salary, plus equity) for essentially the same work, and gets more and higher quality talent as a result.
I wonder how a smaller org can escape this problem, if at all.
The market is pretty brutal but I’ve seen many extremely talented folks work at mediocre organizations because of reasons other than financial. I guess everyone is happy with the arrangement.
It was an outlook I never quite understood. The core business was and is strong, reliably profitable and has made plenty of people quite wealthy and employed dozens more gainfully. Why mess with that success? What's in it for them? What's the point of (fake numbers) $100m in sales vs $50m when it comes to our 70-person company and small number of private, non-institutional owners?
But I am, of course, not the one who makes such decisions.
A growing company that was bootstrapped and is trying to make something actually worth selling in order to fund further growth is an entirely different situation.
Other option is provide equity to your engineers to help make up for compensation gaps.
Please let us know which VC firm you're with giving those terms so I can apply.
This sentence was a little bit hard to parse because of conflicting meanings of the "former" (in sentence vs in your employment history). You meant that your new employer have you 150% salary and equity, and the previous one couldn't compete, right?
The CEO seems to manage the CPO & CTO relationship in these models as well, which seems like a risk given a CEO's value is investor and market / key customer facing, and any time they spend on managing the peers in the CPO/CTO relationship is opportunity cost against that value.
It also means the CPO & CTO as peers will be jockying for the CEO's attention as a deciding factor, which seems like unnecessary friction. The CPO's key relationships are to the board and supporting the CEO activities and sales key accounts with strategic feature alignment that may include M&A.
The CTO's key relationships are outward with technology partners, suppliers, vendors, etc. and downward in the org. A CTO should be a partnership with a COO role, as they are solving scale and optimization problems together.
All that is to say, depending on org maturity, I don't think CTO/CPO should be peers under the CEO. IMO one of the CPO/CTO should report to the other, as otherwise you're going to get unproductive running battles between them that cost CEO time. If you have this model and are wondering why your staff aren't working well together, it's you.
This is why it can be useful to keep a founder around as a facilitator with convening power, provided they aren't too marginalized or a loose cannon. That said, I've only consulted to orgs like this and see it across them, and this is not the view of a CEO.
There's definitely overlap between the two, but a collaborative approach works better in my opinion. As a disclaimer, I'm a CPO myself - and enjoy the working relationship and collaboration with the CTO, while reporting directly to the CEO and the board. Also, I guess that sometimes it really comes down to the types of personalities that are present in the management team - as well as the type of product being built.
There is only one fundamental truth: you need a good product that people are willing to pay for. Everything else is optional — you can get by and run a successful business without marketing, without mailings and mailing lists, without A/B testing, without AWS, without Kubernetes, but you cannot run a successful business without a good product that people are willing to pay for.
Anyway, the advice in the article might actually be good for large venture-funded SaaS businesses. I didn't mean to mock it, I just found it amusing how different my viewpoint is.
Not sure if this is the correct rule of thumb, but one day from Software Engineering I might want to move into management or have my own startup and this would help.
So it's for companies that aren't profitable even at IPO. Not my cup of tea.
On the ML side you probably have more people working on data gathering/cleaning/client specific tweaks than actual modeling. (Practically, I’d say is 80%-20% at best in the projects we do at work).
Cargo-cult leadership might be the demise of the modern SaaS/VC workflow.
People that doesn’t understand what some position even are trust this kind of things and hire a lot of people for the sake of it.
Measuring growth in headcount is disfunctional and just bad. I’ve seen many of startups hiring tens on engineers because they can, and to show “growth” to the outside world, not because they added any value
Better to take external advice on typical team sizes / talent needed to achieve specific milestones specific to your business. Hire to get meaningful work done. Not based on a funding-centric "target" org chart.
By the point when you got round-A, you have already passed the most difficult time for a startup: to survive from bootstrap or angel-fund to round-A.
the key is actually how to get angel-fund or bootstrap so you can reach to round-A, instead of being those 95% failed to launch ones and never saw the light.
Though the final scenario with hundreds of employees does feature a General Counsel as the final entry in the chart.
Have a sharp general counsel _early_. If you want real (i.e. enterprise) customers you will get destroyed if you agree to their terms for SLAs and penalties.
My experience is probably skewed, mostly working for SaaS companies providing enterprise B2B software, but the product and engineering team represents only 15-20% of the headcount - the customer success team is often 2-3x larger.
I've also usually noticed a lot of inefficiency on the dev-side too (bike-shedding, over-engineering, rewrites).
What does this mean? I had the impression employees are a liability.
You should hire people based on a market need for that role. While I agree that this is the org chart that most often organically arises, it’s not because CEOs read this article on sub stack.
This is about as useful of information as telling me how to decorate my office space when I have a 50 person company.
Definitely worked for enough C-level people who said "We now need a 6-person data team in our startup because my friend Bob said so, and he's a big deal" or whatever the trend of the day was, to find such prescriptive templates questionable at best and dangerous at worst.
I'd be very surprised if a CEO at any large company has more than 50 direct reports.