Losing Money on Stamp Recruitment

[Cormac Stark]

On my previous round of stamp recruitment, which has concluded, I sent recruitment telegrams to New:10000. Of those telegrams:

* 8,357 Delivered
* 2,445 Read (29.26%)
* 89 Recruits (conversion rate: 1.06%)

A pretty dismal yield, but also pretty standard as far as I know, and I recognize there is no way for site administration to improve this yield without turning stamp recruitment into more of an unfair pay-to-play advantage than it already is, which is not something I would want to see happen. However...

* 1,479 Blocked by Category
* 151 No Such Nation
* 13 Class Member
* 1,643 Not Delivered

Not to sound cheap, but I find this pretty unacceptable. I want to contribute to NationStates as much as the next guy -- I became a site supporter the first day it was made available, if I recall correctly, and I've since bought stamps pretty liberally for both recruitment and WA campaigning -- but I really don't like the idea of literally throwing money away. I'm not asking site administration to give preference to stamp recruitment or to ensure better outcomes for stamp recruiters, but I don't think it's unreasonable to ask for stamps not to be spent on telegrams that are never even delivered. This also goes for other types of telegrams that aren't delivered, like blocked WA Campaign telegrams.

[Klaus Devestatorie]

I've sent more stamp campaigns than I care to admit (probably approaching a 7 digit figure of stamps), and Cormac's rate is probably above my average. It's difficult for me to justify that on region building project that itself may or may not be successful medium-long term. Stamps to no such nation and class members, at minimum, should ideally not be count as spent. That said, there's no way for scripts to get their 180 seconds back or for manual recruiters to get their progression-towards-carpal-tunnel back, so I'd understand if this didn't happen.

[[violet]]

I've spent the day hacking at this and I think it's doable. There are a few good reasons against it, which is why we don't have it already, but at this point I do agree it's just fundamentally frustrating to feel that stamps are going to waste.

So what I think we'll get is a system that refunds stamps upon non-delivery every few minutes.

Reasons against, raised in the past, just FYI: Stamps are consumed on sending, not delivery, in the real world. The code consumes stamps at delivery time and doesn't know at delivery time how the telegram was originally sent, so a whole new system is required. It often winds up being insignificant amounts, like refunds of a few cents. The stamp price has the assumption that not all will be delivered built into it, so we'd have to raise the price after implementing the feature, defeating the point. It would encourage more blind blasting of recruitment messages, since you wouldn't have to care how many bounced, when what we really want to encourage is more targeting.

[Ever-Wandering Souls]

If it's more targeting you seek, perhaps code for finer sorting of targets? API systems, including NS++ which I believe you have access to code for (that and it's on github), often allow for targets including by the nation's political stance (issue-based, not player based), by targeting active nations, etc. While that applies less to "new," "refounded," and other targets (at least in the activity sector), their presence (and the ability, perhaps, to stack multiple options) could allow a campaign to target, say, only active WA nations with low political freedoms, or only new nations with leftist ideologies.

I would also make a point on the "for" side of refunds - I'd be interested to look at any NS++ stats, should they be available, on how many new nations are founded via that. Knowing that some raiders will churn out 60 new ones for a tag run via that service, on top of other users, and nations founded via it's puppet creator have all mass TG's blocked by default, I'd be curious to see how many stamps get wasted via that service's consequences alone. It's also worth noting that one of the things most often told to new players in telegrams they do receive is "oh btw this is how you block more of these messages" (both a helpful tip and makes your telegram 1/less), meaning that many that don't hit the initial burst of junk mail are doomed to die - the real life comparison perhaps being junk mail that tells you how to get on a list not to receive junk mail, but a private one that junk mailers can't see. The post office merely just burns all mail affected. Without the ability to avoid those who have all mail of that type blocked when sending, is it as fair to take stamps for sending them anyways? Is there any way to create a stop of some sort, similar in idea to when a player has blocked you and you cannot send mail to them, that comes into effect before mail is sent? I'd see that as more fair -

looking at the number posted by Cormac - 1,479 Blocked by Category
* 151 No Such Nation
* 13 Class Member
* 1,643 Not Delivered

roughly half are blocked by category, and half are simply "not delivered" (I know that includes full inboxes, though I forget if that included deletion without reading). Allowing refunds for, or simply preventing delivery attempts to, those who have all mass mail blocked (as well as perhaps class members, again blocked by type, and less solidly, no such nation, usually nations deleted after the send attampt, for which the arguement would be "wasted stamps due to moderator action"), while still taking stamps for those "not delivered" (i.e. the mail was thrown immediately in the trash) would cut the issue by about half, perhaps making a valid compromise between the two?

[[violet]]

If it's more targeting you seek, perhaps code for finer sorting of targets?

I'm happy to implement this kind of thing if recruiters will find it useful.

perhaps making a valid compromise between the two?

In the past, it's been a question of whether it's worth creating a whole new stamp-tracking system. If it is, it's just as easy to apply it to everything, i.e. all different types of non-delivery, so no compromise is necessary.

[Ever-Wandering Souls]

If it's more targeting you seek, perhaps code for finer sorting of targets?

I'm happy to implement this kind of thing if recruiters will find it useful.

perhaps making a valid compromise between the two?

In the past, it's been a question of whether it's worth creating a whole new stamp-tracking system. If it is, it's just as easy to apply it to everything, i.e. all different types of non-delivery, so no compromise is necessary.

I think they would. You could perhaps ask some other users who have recruited via NS++ if they used it as well.

Technically, perhaps. But not all the cons you mentioned were technical -

Reasons against, raised in the past, just FYI: Stamps are consumed on sending, not delivery, in the real world. The code consumes stamps at delivery time and doesn't know at delivery time how the telegram was originally sent, so a whole new system is required. It often winds up being insignificant amounts, like refunds of a few cents. The stamp price has the assumption that not all will be delivered built into it, so we'd have to raise the price after implementing the feature, defeating the point. It would encourage more blind blasting of recruitment messages, since you wouldn't have to care how many bounced, when what we really want to encourage is more targeting.

Real life comparison, need for an increased cost due to less stamps being bought, more blasting of messages, being the non-technical ones. I attempted to address the last of those with my targeting suggestion, and the first two with the noted compromise By lessening, but not eliminating, wasted stamps, you can both better fit the idea of a "no call list" in NS (i.e. companies won't waste time, aka money, calling those they already know have blocked them, if in this case legally ), and not suffer as many returned stamps, by still consuming those lost to reasons other than technical error (deleted nations) or blanket blocks (by category/class).

[Flanderlion]

If we are talking about tweaks to the stamp system, would we be able to send messages to only a certain letter in a region, instead of an entire region?

Personally - when creating a nation, I find it annoying that all TGs are blocked immediately with NS++ but NS++ pro's massively outweigh my missing out of welcome TGs etc. But that isn't really related as much to this decision.

If the stamp prices change, what sort of change are we looking at, percentage wise?

[Zemnaya Svoboda]

If "previous recruitment too recent" were refunded there would be no cost to having multiple stamp recruitment telegrams going in parallel. I would advise not refunding "previous recruitment too recent" non-delivery.

[King Nephmir II]

In addition, it would be useful to remove specific regions (such as allied regions, for example) from a telegram. For example, "Tag:New10000, Exclude:The Pacific". This would allow recruiters to have better control over where their stamps are going.