7 ms·
How not to monetise a popular blog
- ams6110 10y agoThe worst designed website I’d ever created was the most popular. Everything I thought I knew about the importance of design had been blown out of the water. Just look at Drudge Report for another example of this.
- tyingq 10y agoChinese websites have a penchant for a very busy design that's really odd. I don't get why this sort of setup is desirable: A large insurance company in China: http://pingan.com/ http://pingan.com/ Similar, a large bank: https://www.dahsing.com/html/tc/index.html https://www.dahsing.com/html/tc/index.html China Postal Office: http://www.chinapost.com.cn/ http://www.chinapost.com.cn/
- kolinko 10y agoMay be cultural differences due to different writing systems.
- Waterluvian 10y agoOdd thought: does their alphabet contribute to that feeling too? On average, each character looks considerably busier with more lines and curves than letters of the Latin alphabet.
- tyingq 10y agoI don't think it's that. Just the raw count of "clickable things" is much higher. Try some of the menus. Or "translate this page"...it doesn't look better. I suspect it's just a cultural thing about information density.
- theartfuldodger 9y agoThere have been articles on HN in the past about the stark difference of "good design" for Asian sites. I remember some reference to the busy crowded streets and their signage. From that article I found lingscars.com which is now my most favorite commercial website. Make sure to read the copy and check out the source code
- theartfuldodger 9y agoThere have been articles on HN in the past about the stark difference of "good design" for Asian sites. I remember some reference to the busy crowded streets and their signage. From that article I found lingscars.com which is now my most favorite commercial website. Make sure to read the copy and check out the source code
- theartfuldodger 9y agoThere have been articles on HN in the past about the stark difference of "good design" for Asian sites. I remember some reference to the busy crowded streets and their signage. From that article I found lingscars.com which is now my most favorite commercial website. Make sure to read the copy and check out the source code
- rhizome 10y agoI read recently that written Chinese has something like a 10x information density over written English, so you might be onto something there.
- ghaff 10y agoIt's not just China. A lot of websites, event signage, ads, etc. in much of eastern Asia look to my Western eyes like something someone's kid put together for a school project. The script may be part of it--it doesn't lend itself to the sort of clean sparse design that's in vogue in a lot of the West. But, as you say, there's also just a lot of clutter and generally cheesy art work, stock photos, etc.
- ValentineC 9y agoI found this (and the corresponding article) to be a good read on why Asian websites are that way (the lower comment threads seem to be more relevant to the topic): https://news.ycombinator.com/item?id=6718067 https://news.ycombinator.com/item?id=6718067
- pswilson14 9y agoSignificant cultural difference makes for really interesting design differences in apps and websites! Americans (maybe Westerners more broadly) tend to prefer simple designs with plenty of white space, where Chinese users find a simple design to be annoying and difficult to use. They want the content all in one place - whitespace is wasted space. I've read once that this might have something to do with slow internet in China, which leads to users wanting to reach content in as few clicks/page loads as possible, but I'm not sure how accurate that is.
- sverige 10y agoActually, Drudge Report is a great example of website design that works. OK, maybe you hate Drudge for political reasons, but his site gets more visitors than yours. You might be able to learn something from that. I have visited it pretty regularly since 1998 and it is awesome that it looks the same now as it did then. There's no stuff that floats up or down or around or in from one of the edges, no blinking fucking arrow telling me to scroll down for more info (as if I hadn't figured that out since the only thing visible is three words in a 400 pt. font and that's not exactly what I was looking for), no autoplay videos, no hamburger buns, no tabs, none of what passes for "good web design." I can see at a glance what I might be interested in, and when I click on it, I get more info. A back arrow gets me back to the place I started. Tell me again how that's bad design.
- ams6110 10y agoI agree actually -- I mentioned Drudge Report as another example of the phenomenon that a website can be popular without a lot of visual design. Maybe the point is that content trumps appearance. Also good point about the design not changing. I hate that seemingly every time I go to pay my Comcast bill the UX is different.
- barrkel 9y agoIt sounds like you're conflating appearance with visual design. I think they're barely related. Appearance has to do with aesthetics, design has to do with use. They certainly intersect in that they both affect the end result, but their purposes are almost completely orthogonal.
- tombrossman 10y agoMatt is an interesting character and I recommend people watch him at this National Press Club talk back in '98. https://www.youtube.com/watch?v=RvcA75f9_ik https://www.youtube.com/watch?v=RvcA75f9_ik If you can't spare the 40 minutes to watch it all, skip to the end when he takes questions.
- Spooky23 10y agoIt is classic design that just works. The modern design aesthetic often focuses too tightly on optimizing some specific set of use cases. The problem is when they build the journey maps or whatever, the designers often don't understand the use cases completely or are specifically focused on optimizing some metric. My local newspaper does the overdesigned path. The information you need is purposely not easy to find. Their objective is literally to make you abandon your quest and look at the "e-edition" of the paper newspaper because it generates more revenue. In other contexts, this stuff is most obvious in responsive web apps, where the designers just bury features for whatever reason.
- pmiller2 10y agoMy go-to example here is Craigslist. It's pretty fugly, but functional enough as a site, not to mention useful for the content.
- hbosch 10y agoCraigslist is another example. Dense, "ugly", but the wheels continue to spin. There is no magic bullet in design. Drudge Report's design would never work for a product like Facebook, or Stanford.edu for instance - your design says specific things. Drudge is like a Costco; it's intentionally "raw" to give the (false) impression that it's simply so focused on delivering a product that they don't have time/don't care about design. Lots of approaches work for different reasons. The reason why you should avoid using out-of-the-box UIs like bootstrap is similar - by taking a generic approach your design says nothing...
- rhizome 10y agoPut things where people are able to figure out what to click on to see the thing they want to see. A 2px vs 3px border doesn't affect that, it's just an implementation detail.
- bemmu 10y agoI love reading experiments like these with actual numbers included. To me $225 / month (6£ / day) doesn't seem bad at all. Some A/B testing of ad placement could get that to $500 / month. Then you find a freelancer for $100 / month to update it. You could live with 10 such sites in your portfolio, except at least in my case I can never resist spending my time tweaking them instead of starting new ones. "Maybe just one change and it'll be 30% bigger!"
- pryelluw 10y agoYou need to hire a social manager (email is social media). Have them manage while you grow new ones. OT: Why dont you vlog about candy from japan? Id love to see videos about it.Maybe stores that sell it, taste comparisons, history of a specific candy, etc.
- bemmu 9y agoI guess I became a bit disillusioned after I couldn't attribute a single sale to this popular video: https://www.youtube.com/watch?v=dSCgXVkYQcA https://www.youtube.com/watch?v=dSCgXVkYQcA (here's my writeup on it with stats: https://www.candyjapan.com/behind-the-scenes/sales-results-from-getting-3-million-views-on-youtube https://www.candyjapan.com/behind-the-scenes/sales-results-f...) "Figuring out YouTube" is definitely still something I'm interested in, though. Writing might be more suitable for me, but I don't often come across topics about Japanese candy where I could easily imagine how to get any distribution for them (where to post).
- pryelluw 9y agoContact a known female cosplayer on Instagram and see if they will promote your stuff for a % of the profits.
- JohnJamesRambo 10y agoIt seems his "current blog" failed too lol. Cyclelove.net points to nothing.
- 3stripe 10y agoShould be cyclelove.cc
- TheOtherHobbes 10y agoI clicked through to the front page of his main blog, and my eyeballs melted. Some designers have strange ideas about colour choices. I suspect the number of people who want to read a small block of black text on a monitor-filling solid slightly-off-bright-lime-green background is not huge. I don't think I'm being unreasonably picky here. I actually said "Aaargh!" out loud when the page loaded. (Or words to that effect, anyway.)
- skilled 10y agoI have always had an inkling towards clean design. There's something about having as few fonts and colors appear on any given page. I'm guilty of overspending time on "perfecting" individual pages, e.g., homepage, and posts. Recently, when working on a new project, I went over the most prominent design features of similar websites in my niche, and quickly realized how many of those sites can "get away" with cheap and often default templates. I think some of it comes down to how big social media has become, and how the homepage + posts page design interaction is becoming obsolete for content-rich websites. Everyone is jumping straight to the "good bits". Nonetheless, browsing an aesthetically organized website is much more pleasing than looking at countless, and pointless, UI elements.
- Finnucane 10y agoAn unmonetized blog on the Internet? Oh, the humanity.
- mcguire 10y ago"I didn’t build up a mailing list. At one point I had almost 20,000 people subscribing via Feedburner. Imagine if I had converted half of them to email subscribers. That would have been a huge platform to work from. And email is the single most effective way of communicating with people online." I don't get it. I'm on like 27 mailing lists from places where I bought something or had to set up an account for some reason, and the first thing I do is set up a filter that shovels their traffic into thr trash. Why does everyone say, set up a mailing list?
- pryelluw 10y agoBecause most people want to read what you publish. No further explanation needed. Now, being included on a mailing list because you bought something is a dick move. However, most people dont mind.
- Swizec 10y agoI do the list on buy thing and the first email people get is "If you don't want this, unsubscribe now [big fat link]" Almost nobody takes me up on it. My guess is they care about the free updates and such. The reason I do this is because I too once made the mistake of not building a mailing list for a moderately popular blog.
- detaro 10y agoI wonder how many people just delete that mail without even consciously reading far enough. I've observed some interesting cases of "banner blindness" in myself with helpful information that was too big and fat...
- Swizec 10y agoGood point. But hey, every email I send has an unsub link and the unsub rate is around half a percent. I can live with that and apparently so can my audience.
- kasparsklavins 10y ago"Does anyone other than a graphic designer ever look at a website and think “Holy crap. That thing is a pixel out of place!”. Probably not." Users won't notice when something is perfect. But when anything is misaligned, it sticks out like a sore thumb. Dont give up, we care about the pixel.
- reitanqild 10y agoI and I guess many others don't see it and don't care.
- coldtea 9y ago>Users won't notice when something is perfect. But when anything is misaligned, it sticks out like a sore thumb. No, it really does not. Users don't care for Drudge Report not being perfect, Craigslist not being perfect, and thousand other huge volume sites.
- Nullabillity 9y agoUncanny valley, not trying at all looks a lot better than trying hard and being slightly off. And besides, Craigslist and Drudge Report don't look nice, but they still look aligned in the places where they're trying.
- coldtea 9y ago>Uncanny valley, not trying at all looks a lot better than trying hard and being slightly off. Nice counter-argument, but I'd still say no. There are still tons of sites that try but are still off, and people don't care at all, they're still huge. It's all about the service, unless one's is catering to OCD customers.
- i336_ 9y agoI agree. - Stackoverflow horizontally scrolls at 1024x768. Reaction: Growl at VCs, zoom site to 90%. - "?" button on Q/A editing toolbar pops down a little explanation bar that is misaligned off-by-one to the right (when site zoom is at 90%). Reaction: fidget in chair and be sad :( So it is the little things! PS. I rehashed the 1024x768 thing recently, not much happened. https://meta.stackoverflow.com/questions/344521/request-for-redesign-to-have-better-compatibility-with-1024px-wide-displays https://meta.stackoverflow.com/questions/344521/request-for-...
- coldcode 10y agoMy programming blog has been around more than a decade and gets around 200K visitors a year. Early on I played around with various ad networks but discovered (not surprisingly) that programmers hate ads and won't click on anything. I decided that writing and getting people to talk about stuff (elsewhere, I removed commenting too) was enough pay for me to continue. I support RSS and twitter as subscription methods and leave it at that. I am sure there are ways to monetize but I just don't care to.
- squeakynick 9y agoI’m happy to share my data too. I’ve been blogging for half a dozen years at http://datagenetics.com/blog.html http://datagenetics.com/blog.html and have slowly built my traffic up to about 300k PV a month. A couple of years ago, I installed a single Google Adense skyscraper advert on each page. As others have mentioned, it doesn’t pay much more than a couple of beers a week. On a good month I’ll generate about $100/month from these adverts (which is, at least, enough to cover hosting costs). I’ve had an RSS feed forever, I’m not sure it does a lot. I’ve recently put up a link allowing people to sign up for a newsletter. This mailing list has grown to about 750 recipients. Open rates on mailings are 30%-40% and click-through rate has a higher variance and ranges from 8%-20% (If you are going to experiment with email, I highly recommend a product called Sendy, it works great). Of course, I have a Facebook page to promote each article, and I also tweet a link to each article (where I have just a couple of thousand followers). The biggest generator of traffic is being featured on places like here on Hacker News, Reddit, or being featured on one of the large portals/news sites; these bring supernovae spikes in traffic for the time the links are in the sun. I also get a non-trivial amount of traffic from localized versions of my pages; People seem to like to translate my articles into foreign languages and usually put link-back credits, which is how I find out (I guess there could be people who do it without giving credit, and for those, I’m just ignorant about!) I’ve embedded the occasional link to Amazon to allow users to purchase, if I mentioned a book I like. Amazon’s affiliate program gives me a small percentage cut of anything a user buys within a short time after following the link of my page. It works, and I get some revenue, but in total, it’s not yet broken $100 total lifetime earnings. As regards content, I try to publish a new article every week. Bottom line - until I can get, at least, two orders of magnitude increase in traffic (or CPM), it will remain a hobby
- fzilla 10y agowhy stop at a new paragraph for each sentence? why not a new paragraph for each word?
- tobltobs 10y agoSelling a side for 3 _month_ earnings? Too cut losses and cash in? And I always thought 3 years earning would be a steal. The one who later bought shipmentoffail.com for 5000$ seems to have failed even more. shipmentoffail.com is a parked domain now.
- sanswork 10y ago3 months isn't that bad for something that is currently a meme since tastes move so rapidly it's likely the traffic will be home in a year. Ford most sites 1 year is around a good price but you're almost never going to get 3 traffic is just too unpredictable in most cases.
- Houshalter 10y agoCould a blog like this succeed today? Seems like this function has been taken over by Reddit Tumblr and Facebook. None of which pay the content creators and steal most their content from other sites.
- shouldbworking 10y agoI would say probably, if you're a good enough writer and game the search results a bit. One of my employers gave me permission to spend a day a week writing articles for a couple months. I wrote 5 articles and they brought in over 90% of our site traffic for years, even thought we had hundreds of articles on the site and new ones added a few times a week. The quality of writing matters an extreme amount, more than any of the content farms have you believe. If your content is unique, useful, or brilliant, even to a small number of people, it will bring you good traffic for a long time. We didn't have Facebook sharing or any advertising. People simply shared the links organically
- sanswork 10y agoMy wife use to run a travel blog that got kind of popular at points. We were never able to make any income from online advertising(though if we put the effort into direct sales it probably could have done better) but we got more than enough free stuff over the years to make it worth operating. Hotels, restaurants, attractions, tours, transport, etc are all often willing to give you free access, free upgrades or a huge discount with the promise of a post. You have to be honest in your approach with the fact that you don't guarantee a positive post just a post though(and you have to follow through with negative posts when they are deserved).
- mdotk 9y agoPeople going to a meme site don't want to buy anything, he didn't lose a thing by not setting up an email list.
- ozim 9y agoYeah, second that. Lesson from this post is that 500$ was good deal for such site and "amount of work" he put in is not actual value of site. It is something to learn that things are only worth as much as someone is willing to pay for not amount you think it is.
- mattbgates 9y agoIf anything, the most important lesson in blogging: blog for passion. Don't blog for money. If you can figure out ways to monetize for money, than do so. But never expect it. Been running http://www.confessionsoftheprofessions.com http://www.confessionsoftheprofessions.com for a few years now. In the beginning, it wasn't making any money at all. I eventually decided to sell ad space for $2 a month (now $5 a month due to increased popularity), and I have quite a few companies that purchase ad space. Occasionally, I'll get a sponsored post (they pay me to prioritize their article or pay me to write it), but most of the income comes from the ads. It is not much, but it does pay for the server it sits on and did buy me a laptop when mine broke. Wish I could quit my day job and do it full-time, but I'm passionate about it to keep it going. My visitors and contributors are the ones who really keep me going when I get bored or lose interest in maintaining it. If it weren't for them, I'm sure it would've been in the archives of the Internet. My interest in the topic of the website is much more important than the amount of money I could make from the website. So anyone who comes along and offers a monetary donation is certainly helping out.