Do you DELETE customers' content when payment lapses?
doublemeat
Registered Users Posts: 24 Big grins
I got an alarming email on Sept 9, which raises a very serious general issue/question for me, and I would imagine also for every paying smugmugger out there.
It all started when someone noticed my auto-pay credit card was expired. The support person mistakenly thought my account was past due (it wasn't!), by the length of time the credit card was expired.
She notified me that you guys had been "patient", and that if I didn't pay up, MY CONTENT WOULD BE DELETED.
Frankly, this scares the s***t out of me. You mean, you actually delete your users' content when their payment lapses? (Or lapses for too long?)
I've worked in ecommerce since before the WWW, and with or for at least a dozen web-based businesses (most larger than smugmug, some smaller). In all cases, those businesses had a policy of not deleting customer content when their subscriptions expired! This was just good business: it encouraged users to make good, and it prevented potentially very p*ssed off customers!
I know you guys have a very storage-intensive business model. But other photo sites have this policy as well (e.g. my PBase user account - not paid since I joined smugmug, all content still there.) There are plenty of viable options for dealing with this: e.g. one or more of: nearline storage on cheaper media [I know you guys use high-octane storage]; offline storage; and/or absolute worst-case: conversion of large media files [e.g. TIFF] to smaller [e.g. high-quality jpeg].
I look at it like the Death Penalty. Personally, I'm all for it in principle, and I don't feel it violates any natural laws of nature or violates human rights. The problem is, that so many people falsely convicted, are on death row and or have been killed and exhonorated too late. Likewise, there are plenty of reasons a user may lapse on payment for a significant length of time for reasons that any reasonable human would find excusable, then be horrified to learned that countless hundreds or thousands of hours (and possibly even the only surviving copies of their images/video) have been deleted. Some reasons might include:
Please, please tell me you don't delete user content for non-payment. And please be transparent: I know that you might wish to say you do, but really you don't, for fear that telling people the truth might compromise your profitability. Just tell us the truth. You are a privately held company and truth, transparency, and customer intimacy "should" count for more than quarterly earnings.
And I should be transparent also: If you say you do (whether you do or not), then I am finding another service ASAP and strongly recommending my friends to the same! I don't mean this to be a threat, but I can't think of a single "feature" (if you wish to call it that) that could be more singularly important. (And I'd bet that if other users thought about it, they'd agree; far higher than a non-arbitrary limit on gallery nest levels, or functional smart galleries!)
Thanks,
Jim
It all started when someone noticed my auto-pay credit card was expired. The support person mistakenly thought my account was past due (it wasn't!), by the length of time the credit card was expired.
She notified me that you guys had been "patient", and that if I didn't pay up, MY CONTENT WOULD BE DELETED.
Frankly, this scares the s***t out of me. You mean, you actually delete your users' content when their payment lapses? (Or lapses for too long?)
I've worked in ecommerce since before the WWW, and with or for at least a dozen web-based businesses (most larger than smugmug, some smaller). In all cases, those businesses had a policy of not deleting customer content when their subscriptions expired! This was just good business: it encouraged users to make good, and it prevented potentially very p*ssed off customers!
I know you guys have a very storage-intensive business model. But other photo sites have this policy as well (e.g. my PBase user account - not paid since I joined smugmug, all content still there.) There are plenty of viable options for dealing with this: e.g. one or more of: nearline storage on cheaper media [I know you guys use high-octane storage]; offline storage; and/or absolute worst-case: conversion of large media files [e.g. TIFF] to smaller [e.g. high-quality jpeg].
I look at it like the Death Penalty. Personally, I'm all for it in principle, and I don't feel it violates any natural laws of nature or violates human rights. The problem is, that so many people falsely convicted, are on death row and or have been killed and exhonorated too late. Likewise, there are plenty of reasons a user may lapse on payment for a significant length of time for reasons that any reasonable human would find excusable, then be horrified to learned that countless hundreds or thousands of hours (and possibly even the only surviving copies of their images/video) have been deleted. Some reasons might include:
- User error
- Company error (Never happens right? But you guys almost did this to me through your own error!)
- Family tragedy
- Unemployment
- Extended illness (coma, depression, cancer...etc.)
- Simple forgetfulness
Please, please tell me you don't delete user content for non-payment. And please be transparent: I know that you might wish to say you do, but really you don't, for fear that telling people the truth might compromise your profitability. Just tell us the truth. You are a privately held company and truth, transparency, and customer intimacy "should" count for more than quarterly earnings.
And I should be transparent also: If you say you do (whether you do or not), then I am finding another service ASAP and strongly recommending my friends to the same! I don't mean this to be a threat, but I can't think of a single "feature" (if you wish to call it that) that could be more singularly important. (And I'd bet that if other users thought about it, they'd agree; far higher than a non-arbitrary limit on gallery nest levels, or functional smart galleries!)
Thanks,
Jim
0
Comments
I'm sorry that our Support Hero made a mistake - as Josh said in his follow up, you're good to your anniversary date, March 2011.
Portfolio • Workshops • Facebook • Twitter
Fair enough in principle: I didn't ask if you kept them forever.
But you, I, and others may have extremely different ideas on what "a long, long, long time" is - with some potentially ugly results.
It would be nice to know have - and know - a written policy. 1573 milliseconds? One day? Week? Month? One year? A decade? It would be a very hard customer argument to make that you do have a written policy, but wish to keep it a secret. Sure, some may abuse it, but that should be greatly overshadowed by the reduction in furious lost customers, say, who lost their homes, income, savings, and images in a regional Katrina-like disaster (or maybe one more local but no less disastrous that you guys have know news knowledge of).
(And even if people do abuse the written policy by not paying until the day before you have documented to be deleted...well they still have to pay what they owe you in arrears at minimum - plus a year more to access their content in anything other than read-only mode; so the worst you'd stand to lose from abuse is your IRR on that outstanding money for that amount of time...which you could also recoup via a "reconnect fee". Some might find that annoying or usurious, but far better than deleted media!)
And if you don't have an internal documented policy
Thanks,
Jim
Now, to maintain decent customer service and good relations with customers who accidentally screw up, it makes sense that Smugmug wouldn't immediately delete your images and if you personally contact them, you MIGHT be able to get access to them again. But, that's a pure bonus. They are not obligated to do that at all. No customer should count on that. If you want continuous access, pay your bill. If you're not going to be able to pay your bill and want the content, then download the content and save it before your bill lapses.
If Smug wants to commit to some period of availability, that's all fine, but I don't 'think they owe it to us and I don't think any customer should count on it.
Homepage • Popular
JFriend's javascript customizations • Secrets for getting fast answers on Dgrin
Always include a link to your site when posting a question
Yes, photos and especially video are very storage-intensive. And yeah, if you store it on your primary RAID 1+0 15k RPM fiber channel SAN arrays (or whatever you use), then of course it is super expensive.
And I'm not trying to remove the focus on a written policy or debate technology, but most business highly reliant on high-performance storage, have a "near-line" model; where they move lesser-used content to vastly cheaper storage (e.g. RAID 6 arrays of cheap consumer 5400 RPM SATA drives). This is built into the business model from the very start, as tacking such a solution on after-the-fact is often prohibitively complex and expensive.
Either way, some kind of tiered (or flat) performance to usage [or profit] model is important. And whether you have such a model or not (i.e. it's all just high-octane), then you still need some kind of mathematical model to help you understand when it is most profitable to delete content. Just flying by the seat-of-the-pants could be very expensive, either in angry vocal customers and/or (probably "and") keeping some non-paying users data too long.
It is simple economics: Delete user data too soon, and you lose. It is an inescapable law of ecommerce, particularly involving user-created content.
In other words: If you delete data too soon on average, then the cost of losing customers (who then scare away new potential ones) could be vastly more expensive than the one-time and ongoing costs of keeping it. Or, if you keep the data too long on average, then your are spending more on hardware then on losses to to angry customers (who may not even notice after some point). These things are hard to model and good data for basing one on isn't easy to come by - but you need something. Just a spreadsheet with adjustable assumptions and forming a worst/likely/best-case scenarios would be infinitely better than driving in the dark without a model and a policy.
And, may I say, communicating the policy would be very useful to your customers. Frankly, I'm not very comfortable with "a long, long, long time"! Uncomfortable customers cost you money!
Sorry John but in the real world, that is not the way these things work. Fact is, it's just not good business practice from a pure profitability perspective. If you read what I posted before, I've worked for/with at least a dozen ecommerce sites with written policies of customer content retention. (Usually also involving privacy scrubbing...e.g. deleting billing information for security and liability reasons.)
And sorry, but there is not a single viable online backup service in business that deletes your data when you fail to pay. There is no other business where this is a no-brainer and baked into the business models. Consumer data backup has extremely high rates of non-renewal; mostly because people don't understand the odds of catastrophic data loss, and don't think that far ahead or abstractly if they do. So, when it's time to re-up, the do the math all wrong in their heads about which bill to pay (MasterCard or CrashPlan?). Then, when they really need it, they freak out and will pay anything. Such is the nature of that business.
Edit: Should acknowledge that backup business is also all about minimizing storage cost, rather than maximizing performance. But that doesn't mean their customer model isn't applicable to all forms where user-created content is the basis of business. This is where near-line storage comes in, where minimizing cost rather than maximizing performance is also the goal, and for the same reasons. Furthermore, SmugMug is a backup business (just not it's primary business). I know half a dozen people who have lost their primary storage, and the only copies of photos they have are on photo sharing sites (be it picasaweb, flickr, smugmug, etc.) Even I have recovered more than one photo from smugmug. My best friend consciously and deliberately uses SmugMug as his only backup. (He even uploads photos to private galleries with no intention of displaying, just as backup copies.)
Peace...Jim
So, with a Pro level account bringing in $150 or so per year for SM, how much should they invest in keeping someone's data to maybe get a retroactive fee of a few hundred dollars (while paying for storage of potentially hundreds or thousands of GBs). If they archive locally, they still need to maintain the archives, pay for space for them (if just tapes) or worse, even cheap disk still has to be powered, cooled, and maintained over time.
For the cost of the membership, I just can't see this making it worth it for them.
One thing I have been curious about, but probably won't be told the answer, is how much data does SM actually store overall? That's off topic though...
.02
Where we do agree is that it would make good business sense (for their own selfish customer relations reasons) for Smugmug to keep the data for some short period of time and allow you to recover it. But, after some short period of time, they are just paying Amazon S3 by the GB/day and it's just costing them more than they are ever likely to recover in either good will or renewal.
I can see why they would not want to publish the time period. Suppose they actually keep it for 6 months. Do they want to tell the world that you can fail to renew and then actually have 6 months to decide that you actually wanted to keep your account. No, I don't think they would want to advertise that. It's too easy for people to abuse if they promise that.
Homepage • Popular
JFriend's javascript customizations • Secrets for getting fast answers on Dgrin
Always include a link to your site when posting a question
Good, this is a more productive conversation. My understanding was they had their own cohosted servers, but I could easily be wrong, or outdated. Either way, it's expensive (I use Amazon S3. Way too expensive for personal [or any] backups, that's for sure. I would never use SmugMug's backup/archival service for that reason. Good for reliable enterprise data. There are services for backups that are vastly better suited to that purpose; the premium is on density, cheap, and data reliability [achieved through datacenter architecture rather than expensive hardware] - rather than speed or even necessarily lots of nines.)
For "near-line" storage, the cost per TB is rediculously cheap, and for anything less than a TB of storage it is literally more expensive for someone to go in and (manually) clear out a users data, than it is to leave it there literally forever (or the life of the service). This calculus changes if/when you throw in total automation.
So, if SmugMug utilized "near-line" storage for archived accounts (which could still be accessed read-only and at lower priority), and migrated accounts automagically according to business rules, then the same rule of thumb would hold true: cheaper to keep forever, unless total automation is involved.
So, if it were me running SmugMug, for my own self-interest, I would impliment near-line storage and keep user data forever; thus encouraging users to come back in order to download their original files any time in the future (e.g. after a fire etc.), or to manage it in other ways. (Of course having to pay first.) Also, this would make the service more attractive as it would appear vastly more popular. (E.g. PBase has probably a fraction of the size of active users as it would appear, since they don't delete data...ever.)
Talking about 3 to 6 months is a start. But I would wager that SmugMug's super-secret internal policy double or quadruple that already. It just makes purely self-interested financial sense.
As to the comment of my "tone" someone made, not sure what to make of that. If you read that as a tone of "entitlement" you are badly mistaken and obviously didn't read everything I wrote in the thread before opening mouth. (No harm done...kind of the rule on any impersonal forum like this where anyone can talk s**t and rarely expect to be called on it.) As someone who deals with these very issues every day and often have to "sell" CxO's and other leaders on the simple, self-evident brutality of the $ figures right in their face, it can be frustrating at times (to me and everyone that hired me) that they make decisions against their own self-interest based on profound ignorance and an astonishing unwillingness to look at the hard, unsympathetic data. So yeah, you might say I don't have a whole lot of tolerance for people with obviously little to no knowledge of the subject of enterprise data storage and/or customer-content-centric business models, pretending to "know" or somehow intuitively understand the cost/benefit metrics and decisions involved.
But, this is well off-track from the original subject. My original point was that my data was nearly deleted by a mistake that SmugMug--not me--made. Andy said their (apparently informal) policy is "a long, long, long time". I'm not comfortable with that policy. As a paying customer, that is my prerogative. I have three choices: 1) Shut up and take it; 2) Complain; and/or 3) Migrate to another service with a published policy of retention [almost doesn't matter how long it is]. I don't want another company mistake risking my hard work; or at minimum, I want enough time, in writing, to discover and fix it. I got "lucky" this time with SmugMug's mistake.
So as it stands now, my choice is both 2 (complain) and 3 (migrate away). No one has to agree.
The challenge with your proposal to use a cheaper kind of storage for archiving old accounts is that this would be a bunch more engineering to set up, configure and to move the data around. If one was committed to keeping data for a long time and wanted to do it cheaply, putting it on a lower class of storage would make sense.
But, personally I would not want Smugmug to invest their precious engineering or ops resources in this. I'd rather they just keep the data for a month or two and then get rid of it. IMO, they should optimize for the paying customers rather than expend any precious resources (storage $$ or engineering resources) for the non-paying customers.
Homepage • Popular
JFriend's javascript customizations • Secrets for getting fast answers on Dgrin
Always include a link to your site when posting a question
Hi, your data wasn't nearly deleted. Our support hero simply read your account info screen wrong, and then wrote you an incorrect email. Your account isn't up for renewal until March 2011. It was never in any way, shape, or form in a position where it would have been closed, or images deleted.
Just wanted to clear that up, thanks.
Portfolio • Workshops • Facebook • Twitter
Even approximately. Is it:
a) Less than 1 week;
b) 1-2 weeks;
c) 2-4 weeks;
d) 1-3 months;
e) longer than 3 months?
Personally, I'd find c, d, or e perfectly acceptable. But if it is too short a period, something like a bank error could result in serious business loss, which would not be cool.
Thanks for saying that Andy. Since you can see the support thread, I'm sure you could understand how it seemed that way to me :-).
I'm also sure that you could understand - as a consumer yourself - it could be hard to take the "word" of a business that is not transparent about their policies (whatever your internally rational - reasons might be if there are even consistent ones). At least on the surface. But in this case I've read enough of your posts over the years, and have read posts and blogs of the other employees and owners, to take your word for it.
It does change the metrics some for me. But I don't think it changes the overall outcome. Fortunately for smugmug I have so much family life stuff going on right now, and have already invested so much time already with redesign for (broken) virtual galleries, that I won't be able to tend to anything probably until well after my subscription is up for renewal, so you probably have me for another 1.5 years... (Maybe by then you'll have a published policy, fixed virtual galleries, and non-arbitrarily limited gallery structure!
f) none of the above
Portfolio • Workshops • Facebook • Twitter
I'm sure there's an internal policy that is written down somewhere and is used in training, but I don't think that's going to be public information. I'm just going to hope that if there's a lapse or mistake that every effort will be made (depending on the size of the account) to make sure a user really wants to stop using SM and didn't just forget to update payment methods.
Want faster uploading? Vote for FTP!
Malte