The “agencies of one” storm is coming
elazzabi.com
elazzabi.com
As a contractor, I once had 9 ongoing projects for about 2 weeks, and that was an insane amount of work and context juggling. Handling 30 active clients for a year, let alone a career, seems impossible.
So either the clients aren't asking for much (which is most likely) or I'm missing something. I assume many bigger companies would happily replace a staff member, or a staff member's valuable time, with a ~30k/year contractor. Critically, these companies wouldn't feel the need to "get their money's worth" by keeping the guy busy every day. However, the smaller companies most freelancers see as their bread and butter are more cost sensitive, and will nickel and dime you as much as possible.
Are there enough big, monied companies to constitute a "storm", which I take to mean a shift in the business model for freelancers and boutique contractors? I dunno about that. I'd guess no, offhand.
Seems like DesignJoy has create a neat niche, which is awesome for him! I don't think it scales to the "disruption" level as this article implies.
The same way that some public contractors offer the cheapest rate knowing that it will be inflated, or that low-cost airlines sells you the cheapest ticket, later adding extra luggage, food, and healthy credit card processing fees, a strategy is to give these people what they want.
I could believe after they have their unlimited service, a lot of the customers do not make use of it. In many of these cases the decision making at the client might even be the bottleneck.
I love the non-answer to this question on their FAQ:
> Who are the designers?
> Designjoy is a solo company, meaning that you'll be working directly with the founder.
Then you got microsoft with teams, office, and such trying to be the agency of one for your collaboration needs.
Google is fighting to be your one stop shop for advertising.
I do agree we are moving towards agencies of one.
I don't think it would work for something like software, where the last 10% takes 90% of the time. Maybe you could do it for prototypes, but do any companies really need a "constant prototype" contractor?
> I don't think it would work for something like software
Except that it happens all the time. its fairly prevalent.
> Maybe you could do it for prototypes, but do any companies really need a "constant prototype"
Absolutely.
"There seems to be some correlation between size of the company and the quantity of their needs. The smaller they are, the more needs they have. The opposite is true as well. So yes, many clients pay $2k+ per month and may use the service once or twice a month, equating to under an hour’s worth of work."
Source: https://www.indiehackers.com/post/designjoy-crosses-70k-mrr-...
What becomes challenging at those rates and client densities is managing a tail risk of a collision of multiple big clients suddenly wanting services at the same time. You're just playing chicken with that happening, until you can scale your offerings and revenue up to where you can reduce client density to where you can handle the crunch better when it happens.
There still is often hourly estimation because the retainer frequently will be for a set number of hours, with additional hourly rates if you go over that limit. (Freelancers who offer unlimited hours get in trouble fast.) Companies therefore still triage their needs, to try to get as much done in the set hours without incurring too many additional hours.
Otherwise, that seems like a fast track ticket to burning out. Retainer model and selling bundles of hours is going to keep your client incentivised to not waste your time.
Subscriptions are distinctive because they include terms based on defined outcomes. You're getting X done every month (X may also be use-it-or-lose-it, but it isn't ever Y or Z), and the vendor (mostly) decides how best to accomplish X.
Retainer agreements tend to have bespoke terms; if 5 clients retain you, chances are you have at least 3 different sets of terms and 3 different scopes of work. Part of the definition of a subscription, again, is that it's the same work for everyone you sell it to.
Neither retainer freelancers nor subscription services should be charging hourly; hourly is a consulting anti-pattern.
It's easy to say that but hard to avoid hourly. I say that as somebody who's read your threads and those of patio11 for a decade. I've run my 1-person business [1] for over 15 years and whilst I would like to break away to value pricing I'll still take money-for-time over fixed priced bids any time.
Why? Risk transfer. With fixed bid the risk and management of scope creep is on me. With hourly, on the client. And not having to micro-detail everything in the spec before it gets signed off to make sure no time sinks pop up is nice.
I don't know if the breakthrough I've never managed to escape time based billing is mindset, target audience, offering or location (UK), but I haven't made it happen.
Tell your clients what the project is going to cost, and then work hard not to blow past that estimate. If you go over and it's your fault, eat the overage. If you go over because the customer wasn't ready for 3 days at the calendar start of the project, bill them the overage.
Unless your clients are whackjobs, nobody is ever going to take recourse to the contract terms and the law department. Just engaging the lawyers at all will create costs that swamp the overage dollars. Your customers true recourse is simply never doing business with you again.
If you want to do straight, metered T&M, that's fine. I think it's suboptimal but cromulent. But even then, there's still no reason to ever give your clients an hourly unit of work. Just bill days.
My reading though is what you were doing was still time and materials. Sure you're billing in units of days or weeks, but as I understand it the underlying principle is the same as hourly billing: track time and charge for it. It's not linked to the value of outcomes, it's not eating up unknowns that come out and each has to be negotiated with the client ("We didn't expect $EvilCorp we're integrating with would take 2 weeks to answer every query").
In another comment you say "Time and materials billing does the opposite of keeping everyone on the same side". Can you expand on how you kept people on the same side at Matasano?
> and a clause in the proposal and the SOW that said overages were billed pro-rata additional days.
Did you have an approach that stopped each one becoming a lengthening negotiation with the client over what were true overages and which you should eat? I have had plenty of brusing conversations doing this and in my experience the client's expectation of what's "My fault" can be unreasonably wide, which is one reason for looking to other ways to price.
We never had any negotiations about overages. If we were going to bill an overage, we'd tell the client, and if they told us "no", the project would presumably have stopped there. It didn't come up.
I realized working for large agencies that I'm already the person doing all the work, so why not get compensated accordingly? It was also amazing to me the amount of outsourcing that happens at an agency, most of them really don't have inhouse designers or developers (regardless of their massive size).
The leaders of these companies are well connected charismatic used car salesmen.
Something to consider though.
The site is intentionally very over the top, and not everyone will like it. That's sort of the entire point, you take risks and you're always going to have people that dislike it. There's lots of money in the margins, and that's where we want to exist.
Anywho, I'm honestly surprised not more people hated it.
The retainer model may be more common among legal firms than design agencies (largely because clients imagine their design needs are occasional or episodic rather than continuous), but it has absolutely been used for design.
The hourly rate is in place so that in this relationship, work means more time for the consultant and more resources for the company wanting said work, this is the way it should be and keeps everyone on the same side.
If you mean, like others mentioned here, a model where a company pre-pays for some amount of hours this contractor promises to allocate for this company (at minimum) - this is called the "retainer model", this is fine and is a good idea for small tasks and not a big ongoing project
My on this topic, ad nauseam, for over a decade on this forum:
https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...
Particularly this fella right here:
Context is important here, the work I was doing is helping companies with big complex software projects.
Did you try something else? do you have different experience?
Get unlimited requests and revisions, source files included, for $2,499/m.
I mean, it is unlimited, within a reasonable use, and the agency is the one to define what is reasonable use and what is not.
> Yes! Once subscribed, you're able to add as many design requests to your queue as you'd like, and they will delivered one by one.
You can make as many requests as you want, but it doesn't seem like he makes any guarantee as to how fast he'll get to them.
> Because each and every request is different, it's hard to guarantee anything here but my general rule of thumb for a typical request is two business days.
- https://intercom.help/designjoy/en/articles/5509274-how-fast...
As the number of requests increases, the latency of a response will go up. If a client needs a response in order to evaluate it for the next request, this will naturally lower the total number of requests. Also, clients will at some point realize they get where they want faster (i.e. overall throughput) if they are more parsimonious with their requests.
In practice, healthy relationships and good faith clients and agents are probably sufficient for this to work out.
I'm building a company as a solo engineer right now, and I've been thinking a lot about growth in terms of firepower. I've decided to not anyone for a long time.
Part of the reason is that I hate being an employee, and I don't want to subject anyone to that: https://www.adama-platform.com/company.html
Instead, my bias is to find free people that enjoy their craft, and then use a waterfall model for various efforts.
We niched down to helping B2B software teams design better products (UI/UX). [1]
That focus helped us get really good at tackling design problems for these companies super effeciently.
Companies then benefit from just plugging us into their workflow and hit the ground running.
In the end, we as designers get to focus on what we love doing without the hassle of classical agencies (networking, endless prospecting). So less time is wasted doing things we hate.
And companies benefit from a more economic offering.
Why more economic?
Because we structure our business like a SaaS, our marketing is also structured like a SaaS. Clients find us, we don’t chase them.
The reason classic agencies are super expensive is because they spend many unpaid hours looking for projects. Then when they have one, the client gets charged a premium to cover for the previously unpaid hours.
A subscription model and treating the agency like a SaaS is more fair IMHO for both parties.
As someone who helped launched such a service, it's nice for me because the revenue is reliable, but I suspect both our services and contractor rates are much, much higher than what this is looking at.
It's possibly more in line with the "small art design" markets in which you can hire a graphic designer or musician or whatever out of the global south that is doing solid work for a fraction of what a more developed country agency would charge. Which is great for everyone except the existing freelancers in said developed countries.
Of course, this is better - there will be of course no sub-contracting. No siree, won't be seen. And nobody is ever going to slow-roll "unlimited". Or phone it in. Not at all. And of course there will be no client ever who would abuse an "unlimited" contract and have you work 120 hours a week.
Oh. Wait. Yes, there will, and it'll be just like it's now. There will be people you build a trust relationship with, and they'll get the juicier deals - because you trust them. They'll scale up, because they'll be asked more. And presto one-two-three you have an agency.
Yes, it's a slightly different billing model. And it's more conducive to a "stardom"/"free agent" approach, but that also means it will only work for people at the top of the game - who don't want to scale out and make more money that way.
"Unlimited requests" for 33 simultaneous clients seems... unreasonable
I bet there's a normal distribution for the work that sees new clients creating 90% of the workload, fading down towards zero as they age.
I've had contracts like that renew for years with little work most months because a client just wanted to know I was incentivised to be available when they suddenly needed me.
Sometimes this is called a "network agency".
In practice many companies have a similar model they just dont say it. I mean technically as a dev I'm going to make as many revisions to my code as is requested and I get paid the same amount at the end of the month. My team just knows if its 500 revisions then there are probably some issues with the asks or my code.
[0] (but sub-contracts work out extensively)
Just about guarantee that's the situation.
Yes it is.
I know several people who charge for an initial development fee and then offer recurring monthly packages which handle new development, updates, and maintenance on a retainer. They've been doing this for years. The best part about this scenario is that since the payment is already baked in, there's really no incentive for the "agency" to clamor for more unnecessary work. They are free to be advocates for what's best for the client since they will usually end up making more money by using the extra time to bring on yet another client than they would by servicing this existing client with more custom work.
It is when a contractor becomes a business.
It's the lawyer business model, they pay you a retainer just to stay a client. Largely to ensure you are around when they do need you. This works well with software as moving companies is near impossible, you will have to retrain a whole new team.
It's hard to manage multiple clients and why I moved to high ticket stuff.
Your agency of one soon turns into a agency of N.
which is me
(Now that doesn't get rid of the nightmare you're likely to face if you have to _use_ the insurance for anything major... but that's just as true on an employer health plan, too.)
I would totally do it again -- it's about 1/2 the price of my current BCBS PPO for similar benefits. There's a lot more choice in healthcare when you're not part of a group plan that chooses your options for you.
Companies offering SaaS where there's very low friction - yes. But agencies like that, unlikely.
I'm already in this boat the last thing we want is every single person to be on different terms.