One technique is to estimate the cost/effort/time to deliver on each piece of work. Let say you use units of person-days where 1 person day is $1000.
Some pieces of work can be worked on by more than one person - "build a new training room" could be the whole team working together, "design a new logo" is probably one or two people maximum. For each piece of work, note the likely number of people that would be effective to get that work done, between 1 and number of team members minus 1 (see below).
You could give them your capacity - let's say your team is 5 people and you're all in at least 4 days a week (average subtracting vacation time, sickness, training, travel, interruptions, etc). So that's 5 x 4 = $20,000 budget per week.
For example, they want a new website set up and the last one took 12 person weeks, you can put a dollar value of 12 x 4 x $1000 = $48,000. If 2 people can work on it, it'll take half the time, but the cost is the same.
Ask the customer to prioritise the work once they've seen the 'cost' of each piece of work and how quickly it could be delivered. As they get to understand that some types of work cost more than others, they'll begin to focus on what yields better business value to them, trim the irrelevant stuff and are more likely work with you to make the best use of everyone's time and money.