Feed Viewer logo Feed ViewerRead the web on your terms
Curation

How to Handle Paywalled Links in a Curated Reading List

The complaint that finally changed how I do this arrived in March, from a reader who had followed my weekly links for about two years. Four of the six items that week sat behind paywalls.

A weekly link roundup on screen where several entries are marked as subscription only beside a browser showing a paywall prompt

The complaint that finally changed how I do this arrived in March, from a reader who had followed my weekly links for about two years. Four of the six items that week sat behind paywalls. He had clicked all four, hit four different registration screens, and written to tell me the issue had wasted eleven minutes of a lunch break. He was not rude about it. He was right, and I had not noticed, because I pay for three of those publications and my browser signs me straight in.

That is the trap in curating links from a subscription heavy reading diet. Your own feed looks open. Everyone else sees a wall. I keep around four hundred subscriptions in my reader and roughly sixty of them are paid, so I now treat paywall status as a field I check before anything goes out, the same way I check that the link resolves. Here is the system that came out of it, including the parts I got wrong first.

Not all walls behave the same, and the difference matters

Lumping everything together as paywalled is why curated lists annoy people. A reader who hits a soft registration prompt will shrug and move on. A reader who hits a hard wall asking for $19 a month to read one article feels baited. Those are different experiences and they deserve different handling.

Wall typeWhat the reader hitsHow I handle it
Hard wallNo article text at all without a paid accountLink only if the piece is genuinely worth it, and label it
MeteredTwo to five free articles a month, then a wallLink freely, label it, warn that the meter may already be spent
Registration wallFree but demands an email addressLabel as free with signup, which readers tolerate
Leaky wallFull text in the RSS feed or on a syndicated copyLink the open copy, mention where it came from
Delayed openPaid for 7 to 30 days, then freeHold the link and run it when it opens, or note the date

Delayed open is the category most curators miss. Several academic publishers and a handful of trade magazines drop the wall after a fixed window. If a piece is not time sensitive, parking it in a queue for three weeks turns a locked link into an open one at zero cost to anybody. I keep a folder called ripens later with about fifteen items in it at any time, and roughly half of them eventually go out as free links.

Labels readers can act on

I spent a month using a padlock character in front of locked links and then abandoned it. Screen readers announce it awkwardly, it vanishes in some email clients, and readers told me they could not tell whether it meant paid or secure. Plain words beat symbols here.

What works in my list is a short parenthetical at the end of the annotation: the word subscription, or metered, or free with signup. Nine words of real description followed by one word of warning. Readers can decide before clicking, which is the entire point. My click through rate on labelled subscription links dropped by about a fifth, which looks like a loss and is not. The clicks I lost were the clicks that ended in a wall and a grumble.

The one rule I hold to: never bury the label mid sentence where somebody skimming on a phone will miss it. Put it at the end, in a consistent position, every time. If your annotations need a refresher, the way the label sits inside a sentence is covered in our piece on writing link annotations readers actually click.

The archive mirror shortcut is not the answer

Here is where I disagree with a lot of curation advice. The common suggestion is to route paywalled links through an archive mirror so your readers get the text anyway. Mirror services are easy, they work often enough, and plenty of newsletters do it quietly.

I stopped for three reasons that are practical rather than moral. The first is that mirrors break. A mirrored page is a snapshot held by a third party with no obligation to you, and in the sample I tracked across one year, about one in five mirror links I had used was dead or showing a captcha gate within eight months. That is far worse rot than the original URLs, which is the opposite of what you want in an archive.

The second is that mirrors strip the thing you are trying to point at. Charts, interactive maps, embedded audio and sometimes half the images do not survive. I once recommended a data piece through a mirror and the three charts that made it worth reading were simply absent.

The third is the one that actually settled it. The publications I most want to keep covering are small, and a few of them are two person operations whose entire revenue is subscriptions. Routing my readers around their wall while recommending their work is a contradiction I could not keep writing past. I will link the original, label it, and let people choose.

What I do use are the legitimate paths publishers themselves provide. Many subscriptions include a monthly allowance of shareable links, often around ten a month, intended exactly for this. For research, a locked journal article usually has an open preprint or a repository copy, and a two minute search by title finds it. For news, a wire story behind a wall is frequently carried in full by a regional outlet with no wall at all.

Keeping the overall mix honest

The number I track is the share of locked links per issue. Over the last year my weekly list has averaged around twenty three percent subscription only, and I get complaints when it passes forty. That threshold will differ for your audience; a list for working professionals in finance can sit much higher, a general interest list cannot.

Finding the balance is easier if your sources are deliberately mixed rather than accidentally skewed. I made a point of adding open access and independent sources until roughly half my reading inbox was free to anyone, which took about six months of patient additions rather than one big session. The same discipline applies when sources come and go, and the process I use for that lives in vetting sources with a thirty day trial.

Two smaller habits that cost nothing. When a publication opens a piece that was locked, I go back and remove the label, because a stale warning makes the archive look worse than it is. And when I cannot verify what a reader will see, I open the link in a private window with no extensions and look. That takes fifteen seconds and it is the only way to actually know.

If somebody cannot read the article, your sentence about it has to carry more weight. For an open link, one line of context is enough. For a locked one, I write two lines and include the single concrete thing the piece establishes: the figure, the finding, the name. Not a summary that replaces the article, but enough that a reader who never clicks has gained something.

This is also the test for whether the link belongs there at all. If I cannot name what the article adds, it was probably going in because it was interesting to me rather than useful to anyone. Those are the links I now cut, and my list got shorter and better received at the same time. If you want to keep those surviving links reachable for years, the companion to this is our guide to preventing link rot in a curated archive, and the rest of the method sits in curation.

FAQ

Should I ever link a paywalled article with no label?

Only if your whole audience is known to subscribe, which is rare outside internal team lists. For a public list, the label costs one word and buys you trust that is hard to earn back once spent.

Use them, sparingly, for the pieces that genuinely matter. The allowance is small, usually about ten a month, and some expire after a set number of days or opens. Note the expiry in your own notes, because a dead gift link in an archive is worse than an honest wall.

Is it worth subscribing just to curate a source?

Sometimes. I pay for two publications purely because they are upstream of things my readers care about, and at roughly $90 a year each I treat that as a research cost. I would not do it for a source I link four times a year.

How do I spot a leaky wall quickly?

Check whether the feed carries full text in your reader. If the whole article arrives in the feed, the wall is porous and your readers may well get through on the first visit. That is also a reason to keep reading in a reader rather than in a browser tab.

LF
Leocadia Fennimore

Leocadia has used RSS since the days of Google Reader and never stopped. She edits the site's feed and reader coverage and keeps around four hundred subscriptions under control.

More posts by Leocadia

More in Curation

01doorganico 084goad 3hot8 4465456 after5internprogram ahandfulofblessings ahmadpc ahomcybersecurity allodialresources annodex antiquesandperiodcollectibles aotc asianworldkashmir beedlectrics bloggertemplatefree bluefeatheragency bodyweightsolution campaign08blog carfixadvisor catstreetcollective cbstvsports cnfanspurchasing committeeof500yearsofdignityandresistance covertopsdjs cs-kaiyuangame cuestacomunicaciontotal cxoutlet dancebetgame deskgearhub destinyvillastay detectmagazine di33 diamondsteelbuildings donmo dyanadropship emarsysfinance emimattl escapechan eskitarih evautoplay everythingweddings-orlando farmzone foxstudiosscratch gamevui123 gleasongrade greetingsday hacklabasu haskell iw-kaiyun jagoanproperti jilidskripsi joltbusinessfinance juegodepokergratis juegospucca kaeru-festival latestmehndidesigns lestanzediottaviano litemagazines localcommunityhero lustrsazan mariashebets menswheels newwebstechs1 nikkit ootheday oxigenvpn plurfitness prefetial probudgetcarpetcleaning psychicmrsmartin pyaten qvantor rejoignezlogis rexzbot riselayouts ruchin rwxqsgd5 sakaltime shibiaowang shop-bot skrtips spotlesshaven strcoding tetapatrium topselller uneplanteartificielle vipertechcr webtech1223411 wo-kygaming yamaha-mekar yeknava yoybuysheets zenunwa yamaha-mekar jilidskripsi ankertoto ankertoto ankertoto