Tuesday, December 14th, 2010

We Need New Ways of Judging the Success of Websites

We have such terrible metrics for judging websites! There's income, and there's traffic, and that's about it. But neither of those take into account burn rate, overall expenditure or organization size, just for starters. One way to look at things might be: unique visitors per month, divided by employees. Size of staff is something of a predictor of size of traffic, it turns out! If you have no staff, you cannot make the traffic, for one thing. Obviously there's a slight variable in this metric—which has to do with number of part-time contributors, freelance and marketing budgets and, of course, certainly at the big behemoth, unpaid contributors. Speaking of, let's look at the Huffington Post!

The Huffpo has a staff of 203 (among those are 97 full-time editorial staffers), and they're currently at 26 million monthly uniques (according to their numbers). That's 128,000 uniques per employee.

The Daily Beast has 65 full-time employees (positively tiny in comparison), and, in November, 6 million uniques, as per Omniture. (comScore reports it as 3.7 million.) The Beast also doesn't have "free bloggers," and yet (using their numbers, of course), it still comes up with a pretty good 92,307 uniques per employee.

Slate is now doing 6.48 million uniques (comScore for July to September, as a three-month average). It has but 30 editorial staffers and counting the full team (some of which are shared across the Slate Group), they're 42 people, coming in at a whopping 154,285 uniques per employee. Slate is also, you know, seven times older than the Beast—and all that time online makes for a nice stream of search results and archive pages.

What other ways can we start looking at what websites are and how they perform? Well, pay rate, sure (though try prying that number out of anyone).

Then there's traffic versus site income, certainly. Take the HuffPo again—that's $30 million this year, they say, and they were reporting 26 million uniques via comScore back in March, so let's just totally ballpark it and say they did 312 million uniques this year. That puts their income at just about 1:10 with readers. Fascinating!

Don't compare that to print, by the way: the print edition of the New York Times has, very roughly, just a million subscribers. How tiny and quaint, you think? Well, the news division of the Times Co. alone takes in—mostly from the print product—somewhere around $150 million… each month.

43 Comments / Post A Comment

deepomega (#1,720)


DoctorDisaster (#1,970)

I am wary of putting "number of staffers" in the denominator of success.

jfruh (#713)

If you have too many staffers and not enough income, it's a measure of failure.

iantenna (#5,160)

yes! and can the new paradigm not judge success in such a way that i have to load 10 individual pages to see such and such magazine's "top 10 m. emmet walsh movies" or whatever?

Flaneur (#998)

"Top 10 M. Emmet Walsh Movies" is a highly amusing concept.

deepomega (#1,720)

1. Christmas With The Kranks

David Roth (#4,429)

2. Blood Simple/Red Scorpion (tie)

iantenna (#5,160)

@deep: you're sick, ironic, or both. also, did you see this from yesterday?

Abe Sauer (#148)

3. Blade Runner/Wildcats (also a tie!)

iantenna (#5,160)

@abe: only if it's the director's cut.

deepomega (#1,720)

@iantenna: Oh yes. I would notch it just below the Jingle All The Way review.

Abe Sauer (#148)


iantenna (#5,160)

it's mostly just footage of ll cool j preening.

sailor (#396)

I think "wardrobe malfunction" is as good a metric as any.

boyofdestiny (#1,243)

How about "number of restraining order-ish readers"?

Neopythia (#353)

So uniques–those of us who click back here relentlessly throughout the day are only counted once per month?

saythatscool (#101)

I think you need to go feudal on this shit with a peasant vs. royalty breakdown.

Can I be a Viking?

saythatscool (#101)

I would have it no other way.

Mindpowered (#948)

Ohh. I want to be a Hun

NFK (#8,747)

I got dibbs on Vandal.

garge (#736)

Yeah, so I guess I'll go stand with the vassals.

S_W_S (#4,296)

Net profit is usually the measure of success in a business, which relates more to your ability to attract advertisers, and ad revenue is not totally dependent on traffic data. It's also possible to sell against the quality of your audience. They are "influential/cool" or "super rich" or "very gullible", etc. For a growing/niche site like The Awl, I would think the obsession with macro-scale traffic numbers is more of an issue and that you'd be hoping that people/advertisers more thoroughly considered the qualitative view in their metrics for success.

sigerson (#179)

EARNINGS BEFORE INTEREST, TAX, DEPRECIATION AND AMORTIZATION. EBITDA is the benchmark for all business and EBITDA multiples are the main formula for valuation.

MollyculeTheory (#4,519)

Combined IQ of commentariat times cube root of number of cocktails imbibed while on or as a direct consequence of the website, divided by number of mentions by Glenn Beck, plus the log of the number of bear videos. THERE.

kneetoe (#1,881)

I like this. Any negative impact my low IQ has on the Awl's success metric can be more than outweighed by number of cocktails!

BoHan (#29)

How many Huffington Post staffers does it take to screw in a lightbulb? A lot.

Abe Sauer (#148)


narnio (#38)


BoHan (#29)

This is why you're a famous writer and I'm not. Per Abe Sauer, we now have the correct answer. A: 203, 1 to screw in the lightbulb and 202 to prepare the Slideshow.

Rod T (#33)

This is all fine for "mass" websites, but I prefer to shop at a boutique instead of Wal-Mart. And why is all of this quantitative? Why is there no mention of fucking quality and uniqueness and adding to the ambience?

Sure your ad might reach millions if you advertise on Perez Hilton, and if you have a shitty mindless product, it might be the right place for you. If you are, say, Saab, though that's probably not the site for you. If you are, say, Rag & Bone, you might want to advertise on my lovely site. If you are a purveyor of bear repellant and cat toys or, say books of words, you might want to advertise on this one.

It's just a symptom of the gutless society that we live in that we hide behind numbers to rationalize these things. If you love something, stand behind it with your ad dollars. If you love content from one person more than another, pay them thusly. Have confidence in your preferences and quit being slaves to statistics.

neel (#239)

And how does The Awl size up re: this metric? No sarcasm, genuinely curious!

Rod T (#33)

My point is to abandon metrics and starting to actually think about marketing. Metrics are designed to aggregate data to the lowest common denominator so that decisions can be made by inexperienced drones in large quantities (such as media buyers, straight out of college). My proposal is to think more about what is the right ad in the right place.

KarenUhOh (#19)

Absoposilutlely. Numbers are great if there's no such thing as having enough. Or too much. Which, many people say, Why not? Well, KUO says bullshit. That's why not.

If the business is sustainable on its fare–if the income generation suits the clientele–if the owners/managers can be satisfied with satisfactory standards of production/profit–then who cares if you have 100,000 more eyes in Idaho than Tina Brown or Lisa Douglas?

I swear. Capitalists have no idea how much is how much.

jordank (#9,051)

Simple formula:

(number of pages created + time) x relevance / cost = success

as for "judging" websites, that's subjective. a matter of taste, need and perceived value in the mind of the audience member (fka user).

rex (#557)

It is obvious what the new metric for success should be: percentage of employees who can cite from memory the number of employees at HuffPo, DailyBeast, and NYT.

narnio (#38)

But that doesn't even make sense!

NinetyNine (#98)

"They like compromised user databases generally"

Post a Comment