Shared publicly  - 
 
On the Digital Gruntwork Labor Market

I am beginning to realize that there is a whole "Groucho Marx" zone of tech work that is in that gray zone which is beneath anyone who is good enough to do it for pay.

One such task is creating and maintaining simple websites using something like WordPress, and dealing with basic security, upgrades, SEO, themes, no-code customization, ftp-ing stuff, making quick and dirty graphics/logos, writing basic copy, doing social media integration, tweaking the cache on occasion etc., is one such kind of work. Everything just short of CSS/PHP hacking.

It takes a broad but shallow skillset, spanning basic tech-savvy, keeping up with developments, basic writing skills, a basic sense of visual design, all glued together with common sense and an experimental attitude.

The sheer amount of stuff you need to know is a problem. It is simple stuff, but voluminous. If I tried to write down everything I know about running WordPress sites, it would run to thousands of pages, and page n would be obsolete by the time I got to writing page _n_+20. So this is not book-knowledge. It is virtual street smarts basically.

I'd love to outsource this stuff, but it is remarkably hard to find people who are both competent and affordable, and trustworthy enough (though the work is trivial, it does involve giving admin access to people). What's worse, I don't really know how to interview them or filter out people who have dangerous holes in their knowledge. The only real test that somebody can do this stuff is that they are running their own sites with at least modest success.

There are tons of people who can do it easily, but if you try to hire them, it is beneath them, and if they agree to do it anyway, it is way too expensive. OTOH there are also tons of people who simply don't know enough to do even this rudimentary stuff. But the moment they learn enough or you teach them, they graduate to the first class. Which means if you want to outsource this stuff you have to deal with high churn.

There is also a third category of traditionally-educated, smart people who could do this stuff, but they are so completely lacking in Web tech instincts that you have to babysit them through every simple step. They lack the "muck around and figure it out" confidence.

Though the work is simple, they are used to a permission-over-forgiveness attitude towards technology, and also an inability to learn from messy web sources (which is a "metis" in James Scott' sense). They are the types who get anxiety attacks unless they can buy a For Dummies book for every piece of what they need to do. Unfortunately, the set of For Dummies books around something like running a WordPress site still covers only 80% of what you need to know, and it gets out of date very quickly. So if you cannot self-learn in the evolving ecosystem via unreliable blog posts, forum discussions etc., you're screwed.

Despite a decade-long boom in blogging-about-blogging and a flood of Dummies books, surprisingly nobody has really figured out how to bring this last population into the digital age.

I think there is a huge opportunity for somebody smart enough to create a usable labor force around such work. The challenge is to create a bucket of work that is both interesting enough for the people to stick with it, useful enough that it is worth outsourcing for publishers, and cheap enough that it is a no-brainer about whether or not to do it.

The bottleneck problem I believe is that you first need to achieve a mindset shift from "school knowledge" to "artisan/metis knowledge." Without the attitude shift, these people simply cannot keep up.
3
1
Daniel Lemire's profile photoFred Zimmerman's profile photoMarti LaChance's profile photoBill Seitz's profile photo
34 comments
 
You really hit the nail on the head when describing your third category as having a permission-over-forgiveness attitude. Lots of people out there who are just too afraid to click the button
 
+Venkatesh Rao A lot of this falls under the "sysadmin" category. A lot of it is outsourced to Indian folks. I'm sure you could find someone easily. The reason I do it myself is that it is not much work, and it is work I can do when I would not otherwise work (I can't write 12 hours a day)... and I want to really understand it because my web presence matters a great deal to me. I'm not sure I would understand SEO just as well if I didn't do my own sysadmin.
 
Your post is dead on. In my experience, the best site managers are tech-savvy creatives. These folks are willing to slog through troubleshooting the latest update to WP -- if on another day they can use GA stats to tinker with site architecture, or add a new set of photographs, or (yikes) experiment with the site templates.
 
+Daniel Lemire sysadmin is precisely what this is NOT. Sysadmins are overqualified in general. They can generally do Unix shell script stuff, run scripts, analyze logs, all the way up to juggling EC2 servers and administering Hadoop clusters etc. OTOH they generally cannot do (or don't want to do) the "marketing" stuff: writing quick posts, tweeting/FB stuff, quickie logos, picking a decent theme/choosing colors for stuff, composing decent emails to lists etc.You want a Sysadmin-- combined with an Admin Assistant++.
 
Mmmmm... I think this is only the tip of the iceberg and a harbinger of things to come in the world of computers and even more broadly in tech in general.

Our much vaunted "progress", "innovation" and "disruptive technologies" means that there will be more and more of this flurry of quickly obsolete trivia in every domain.
 
+Jean-Luc Delatre exactly. And it takes a curious sort of personality to be entertained by the keeping-up. I am calling this hypothetical species the "Creative Technology Admin Assistant." creative enough to pick colors and crop photos. Non-flaky enough to be trusted with simple but consequential things like API keys, DNS stuff etc, and creative enough to work off an instruction like "I just bought this domain, get the basic site running with such and such broad parameters."
 
+Marti LaChance you got it. The problem I've seen is that tech-savvy creatives usually have creative itches of their own that makes 'em quickly unhappy doing others' grunt work. If I could figure out how to make the role just sufficiently creatively satisfying to get 'me to stay, I'd make a gazillion dollars running the talent agency around this stuff.
 
If it helps, I am thinking sub $20/hr wages. Sub $15 to really hit the affordability sweet spot. People in that range are usually outsourced, students or liberal arts majors from 2nd or lower tier colleges. And you can generally only get 2-3 of the 4-5 skill buckets needed in one person: basic writing/graphic arts, tech savvy, DIY/learner spirit and admin asst skills (I'd include basic book-keeping in this). If they had basic stats, I'd value them at $25-$30. Most virtual personal assistants seem to mainly offer traditional secretarial services like calendar management and invoicing etc.
 
Damn right. G+ has helped a little on the social media side (I have three circles: Essential, Signal, Noise, and use them accordingly), but filtering RSS has been harder. There are some high-volume feeds that I would love to simply sample from randomly. Ie., I could grab CBC, BBC, Al Jazeera and a couple other high-volume news feeds, and grab 10 random articles from each. Would reduce my cognitive workload immensely.

Similarly, I'm trying to reduce the filter success problem (my filters are so trained after 8 or so years, that the quality of content could keep me occupied 24/7. This is a bad thing for the rest of life, so I have to put in forced hiding spots for things I likely will never read but can't bear to simply ignore).

I'm sitting at home tonight, VPNed into the company network to upgrade a Drupal install from 6 to 7 simply in order to convert an entire class of content over to another site that is Drupal 7. Long, tedious work that is precisely obtuse enough to be impossible to delegate.
 
Impossible to delegate for under $150/hour, that is. :P
 
Hence the success of Facebook.

It doesn't really make sense to try to outsource these jobs at a per-customer granularity. The negotiation cost is too high compared to the negotiable value of the work.

The way it makes sense to outsource / automate it is at platform level : Software-as-service and sysadmin replaced by virtualization technologies that make "bringing up a new instance" automatic.

If Blogger hadn't been bought by Google and languished without any kind of visionary leadership, it's quite possible that it could have provided everything you need from a professional blogging platform. (Including other social networking, reader relation management, accounting etc. services)
 
+phil jones yeah, aggregation will be necessary. 1creative admin per 5-6 bloggers for example.
 
People still don't get this ... why doesn't Blogger have a template market built into the dashboard? So I can click a button and post up a request for a unique template for my blog and fire off a 99Designs type competition? Or an author market if I want to run a blog and hire people to write content for it?
 
+phil jones You know, it looks like you have pretty good business ideas. But I think that you and +Venkatesh Rao are making a bit of mistake. I don't think there is enough of a market in blogs to ever make real money. There are few pro-bloggers, and they already have people doing their designs and stuff.
 
+Daniel Lemire a) it isn't just bloggers. Every small biz is in the same boat. WP as a CMS for such businesses is growing faster than blogging I believe+other use cases. B) Your causality direction is wrong IMO. The market hasn't emerged because of this support labor problem.
 
+Daniel Lemire I don't think the fact that people have been successful making things by hand implies anything about whether or not that market can be automated or done at scale.
 
Slightly side-topic, I've been wondering if it's possible to hire a "CSS doctor". Not a designer who wants to create your look, just someone who knows how CSS works on all browsers and who can, for a couple of hours pay, fix the CSS so that things that are meant to line up, do.
 
Right on point! A view from the other side ("outsource side" if you may :) ).. a couple of my friends tried providing these outsourcing services to the same effect, the staff who gets good enough to handle "most of WP" quickly jumps to a better job / asks for uneconomical pay-hike leaving the founders with mangled workload / depressed margins. Both of them have now stopped providing these services standalone.
 
When you choose a platform, you get what it does out-of-the-box, and you take on the work of making it do what it does not. That is never the problem. The problem is when you want to change what it does out-of-the-box. Then, you will do Quixotic battle with all of the assumptions and design decisions that the platform designers made.

Wordpress is very full-featured and powerful. Altering it is a nightmare of cascading architecture failures. If a project needs more than theming, you're better off building what you need from scratch. 
 
My point: $20/hr for this work is undoable. You're asking for custom software. The fact that you want it built on Wordpress makes it harder, no easier. Use posterous. 
 
+Brandon Hudgeons I've never done any custom hacking of WP ever. At most I've cut and paste bits of code into theme files. I am talking about much more basic stuff that is well within the capabilities of WP+plugin universe. I pretty much stay entirely within the WP architectural paradigm for everything I do with it. But even within that space, there's a ton of tedious stuff I'd like to cheaply outsource.
 
+Stefan King You can have Blogger blogs hosted on your domain. At least you used to be able to. I never bothered to do it (a big mistake on my part, though I had good reasons at the time). But, sure, Blogger is definitely not setting itself up as a pro-blogger platform. And doesn't seem to show any inclination to.

My point was just that it could have been. And that the answer to +Venkatesh Rao's problem of how you can subcontract a lot of small maintenance tasks for things like pro-blogging or other small business "faffing" is for someone to build a platform which aggregates many users of identical services.

Actually, one of the often unrecognised rules of the web is that there's ALWAYS room for someone to come along and reinvent a tired "genre" of platform in a radically better way. There's a great opportunity for someone who does want to create a pro-blogger platform. You could definitely go beyond where Automattic are today.
 
How about (1) outsource WordPress hosting to the experts, then (2) a stable of cheap specialists for the rest? The latter has the advantage of getting you to ponder whether a task is really "worth" doing at all.
 
+Bill Seitz "Stable of specialists+WP host" is a statement of the problem, not the solution. Why is it that I can sort of do it all myse
F, but when it comes to outsourcing I have to manage 6 relationships with a transaction cost total that makes it pointless? This tells me there's market here. And I just used WP as an example. An analogy is to gun-making before autumation. Used to tke 40 separate skilled trades. When the process was mechanized, it took a couple of medium skilled workers to replace the 40. What we need is commodity creative work. We seem to be going all the way back to guild/artisan labor economics instead of synthesizing industrial+guild/artisan=post-industrial. Many people with romantic 18th century yearnings want this, but the reality of guild/artisan economics is endlessly annoying.
 
Well, part of my point was that all those most-technical tasks can be dumped on 1 service/group as part of the hosting. Leaving you with a bundle of more-related bits. Which might allow you to keep that stable small because they fall into the soft/creative arena.

I think part of the challenge here is that so much is changing all the time (new tools, practices) that there's a high overhead of learning. And it's very artisinal, it's hard to even figure out what the right/best approach is.

Perhaps lowering your standard on some of these things is another part of the answer?
 
Also note that these tasks are pretty hard to automate, so you're really buying time by the hour. So if someone billed out 2000hrs/yr (leaving their learning time for night), your pay-rate would give them $30k/yr income. Which works for ThirdWorld FreeAgent-s and recent LiberalArts grads living with mummy, but not many others.
 
+Bill Seitz its actually less about outsourcing what I currently do (eminently manageable) and more about what I could do if efficient outsourcing were on offer. Opportunity costs. Ive had to discard many excellent ideas because outsourcing at what I think are achievable levels hasn't happened yet. And yeah, 30k/yr isn't much, but it's hardly mommy-couch level. Living standards in America are going to plummet in the next decade and 30k/single, 60k/couple is going to look real good soon. And if you get good you can probably graduate to managing a stable of providers. The killer that makes this income level seem like poverty in the US is healthcare costs. The distant second place killer is ridiculous inflation in child-rearing costs, thanks to runaway expectations on the part of both kids and parents. 
 
+Venkatesh Rao What tasks are taking up your time and blocking you? Ie. what are the pain points you're experiencing?
 
+Venkatesh Rao , +Bill Seitz , +Daniel Lemire , +Jordan Peacock Bumping this because re-reading I found it died at a tantalizingly interesting point. I think we can all agree that it would be great to automate / outsource a lot of small-scale annoyances. We probably all, to varying degrees, share the intuition that it's not cost-effective to do it on an individual basis, but a "platform provider" could probably make it work.

But the question is, what is it? What are the 3 most important routine-but-cannot-be-automated tasks that have the worst effort-to-value ratio that you actually experience in real life?

My first thought is that for me, as a developer, CSS is the difference between doing something that looks good enough to show the public and something that isn't. A "CSS Doctor" would be valuable if it was cheap enough. Though, now I've discovered Twitter Bootstrap this is improving the situation.

Also, that's quite a specialist requirement. What things are more general across a lot of small businesses, free agents, "pro-bloggers" etc?
Add a comment...