These impact measurements should be measured in terms of response attributions and budget distributions. The response attribution is just a measurement of who responds to what, and by which channel they responded. So if you were looking for the best way to distribute a survey to students, you would take a look at which marketing channel has the highest rate of contact with students. We wouldn't want to put the survey in our newsletter, as we know the responses we are looking for won't come from that audience.
Budget distributions are a trickier topic, as many libraries have no or very little budget to work with for marketing purposes. We pay to print and distribute our newsletter, but that is our only printed and mass-distributed mailer for the semester. The budget for that is larger than for other promotional items, but it is to reach a very specific audience. Other promotional materials are made for programs, workshops, and events but are primarily aimed at students and distributed both in print and online. A higher budget is not necessary. We pay the premium to reach the faculty/Friends audience with a semi-annual newsletter.
Many of you may not have any budget to speak of for marketing, so you may not think that budget distribution has any impact on you. Re-frame "budget" to mean staff time or maintenance. If a library committee or individual spends 6 hours per week working on Facebook posts to reach a total of 200 people, calculate how much it's actually costing you per person to use Facebook. If your specialist makes $25 per hour, it's costing you $1.33 per person your posts reach. There may be a more beneficial way to use that staff time, or perhaps it may be worth paying the extra money to have your posts featured to increase your reach for the same amount of work. The value your library places on each channel will be highly individualized, but it's easy to use recommended strategies to evaluate efficacy and determine true reach.