Blog

User­voice just announced a new pric­ing struc­ture, much improved if still not ideal:

We’ve decided to switch to track­ing usage based on the num­ber of vot­ers in the last 30 days. The advan­tages of this are:

• It’s more clear. Any­one who votes on your forum applies to your count.
• It’s doesn’t penal­ize you for users who haven’t returned in a while. You’re not penal­ized for a vot­ing spike that occurred a few months ago.

Why this is a supe­rior strate­gic move

They switched from charg­ing for the poten­tial to real­ize value to charg­ing for the real­ized value. Buy­ers now only pay for the value deliv­ered and no longer for the right to use the prod­uct and extract some value out of it. The uncer­tainty for buy­ers is much reduced. The com­pany also demon­strates its strong con­fi­dence in the prod­uct. Their prod­uct is so good to induce users to give feed­back that they are using only the vot­ers as their met­ric. In effect, what they did is switch from a clas­sic pric­ing scheme to an activity-based one where the met­ric is the voters.

The end goal of an activity-based pric­ing struc­ture is always to come as close as pos­si­ble to a value-based one. The choice of the metric(s) is instru­men­tal. In this case, the vot­ers seems to be a good one, but oth­ers could have been used too (like met­rics based on feed­back not voters).

Room for improvement

While the enhance­ment is sig­nif­i­cant, there’s still room for improvement.

Adopt­ing a thresh­old struc­ture, as opposed to a con­tin­u­ous one based on utility-pricing (you pay what you use with­out thresh­olds effects), puts the bur­den on the cus­tomers to fig­ure out and pre­dict their level of activ­ity. True, they can upgrade and down­grade when they want, thereby avoid­ing harm­ful locked-in dynam­ics. Yet, clients need to mon­i­tor their level of activ­ity in order to avoid any sur­prise. They are also less likely to sign-up: they are asked to make a tough and uncer­tain deci­sion where they risk to over­pay. Behav­ioral eco­nom­ics has amply demon­strated that faced with such a deci­sion, an attrac­tive choice is to just delay it.

Dis­junc­tion effects for exam­ple are at play here. If clients don’t know which plan they might need, a sig­nif­i­cant part of them will just wait for a hypo­thet­i­cal pre­dic­tion that will never come, and some will end up not using the ser­vice or sign­ing up with competitors.

Con­tin­u­ous activity-based pric­ing can still be regular

The main rea­son to avoid con­tin­u­ous pric­ing schemes is that, sup­pos­edly, the clients want cer­tainty. In fact, clients do want to avoid uncer­tainty, but con­tin­u­ous pric­ing doesn’t have to bur­den them with uncer­tainty. One trick is to charge a fixed amount but “bill” on actual vari­able use.

So, an ideal pric­ing scheme for web ser­vices like User­voice should sat­isfy the fol­low­ing constraints:

• As close to value-based pric­ing as pos­si­ble: needs to play on the met­rics here to find the cor­rect combination
• Utility-pricing: avoid thresh­old effects where users have to pre­dict how your prod­uct will behave in their spe­cific sit­u­a­tion. They don’t know your prod­uct well and it is a dif­fi­cult choice to make. They will delay it, then over­shoot and pay too much or under­shoot and be dis­rupted by hav­ing to upgrade at a short notice.
• Reduce uncer­tainty for clients as much as pos­si­ble. This means cap­ping the max­i­mum charge for very big clients.

Here’s how you could imple­ment this:

Ask clients to esti­mate their needs (pro­vid­ing esti­mates of vot­ers in incre­ments of $10), and make it a reg­u­lar monthly pay­ment. Each month, charge them the full reg­u­lar amount, but credit them back in their User­voice account the dif­fer­ence between what is charged and what is actu­ally used. If activ­ity ramps ups, then use the accu­mu­lated cred­its before ask­ing them to upgrade. If the activ­ity is sta­ble, then when accu­mu­lated cred­its reach the full amount of the monthly pay­ment, use the cred­its to pay it and do not charge them that month. Even if they do like reg­u­lar­ity, no one will complain…

If clients are above their limit, tell them the dif­fer­ence will be cred­ited as neg­a­tive cred­its, and that they will ask to pay those cred­its, the dead­line being when they reach the amount of the reg­u­lar pay­ment. Advise them they should increase their monthly charge so that pay­ments are smoothed out, but accept one-time payments.

Such an imple­men­ta­tion is pos­si­ble with the major pay­ment gate­way providers and doesn’t increase the total costs of processing.

Con­tin­u­ous pric­ing doesn’t have to be uncertain.