This post was prompted by two players having questions that required NPCs to do research. One PC is a cleric, the other a paladin. They worship different gods, one the goddess of fire/the hearth, and the other the god of death.

The cleric of the fire goddess has a masterwork matchlock musket. I assigned a d12 usage die for the powder. It is down to a d6. Few shots have resulted in a hit. I ruled that it takes one round to re-load, so it is far from overpowered. The one keg of gunpowder they found they soon lost. So finally in the capital they checked the chief temple and the keepers of lore are researching for the location of any other matchlocks and the recipe for gunpowder.

The paladin worships the god of death and has a war scythe. At the temple of death he asked about magic weapons. They are researching this question.

Faced with two such questions for which I had not yet developed an answer, the NPCs told the PCs that they would send word to the inn where they were staying once they had answers.

I took time between sessions to determine how long it would take for the researchers to have an answer and what that answer was.

The PCs left town in a hurry and are slowly making their way back. They teleported far away and are travelling back overland, hoping to reach the sea and sail back to the capital. They have no idea if the researchers have an answer for them or not.

The following is my thought process for coming up with the answers I arrived at for these two questions.

Research by Sage p. 32 1e DMG – d20 to determine the d% target number to know the answer. 

In this case know means either memorized or in their library.

Categories of General, Specific, and Exacting from the DMG are helpful.

Consulting the archivists and librarians of temples and libraries should work much like sages.

Sample questions:

Where are magical holy relics?

Where are magical holy weapons? (for a paladin, etc.)

Librarians, scholars, archivists, or alchemists could also be consulted.

How do I make x?

What are the ingredients for x?

PROCEDURE:

How likely are those asked to know the answer? Assign a target number based on how common the information is. 

A good example would be knowing about legends vs. knowing the exact details of which a legend speaks. 

Some may have heard of an ancient warrior’s sword, but that doesn’t mean they know where it is now.

Roll 1d6 for the number of weeks seeking the answer will take. 

(It’s not like they have the dewey decimal system with card catalogs with meticulously preserved books.)

(NOTE: The amount of time it will take is NOT known to either the NPCs or the PCs, only the GM. The PCs will be forced to decide if they stick around and wait for an answer, or come back to the location at some point in the future.)

Note this on the campaign calendar used by the GM. Start the count on the next day, as the current day already has its own business.

Determine what general time of day they find the answer, morning, afternoon, evening however you define those in your game, then what part of those, or just roll a d8 for an eight hour workday, then a d6 for which ten minute block of the hour, then a d10 for which minute, repeat the d6 & d10 for the seconds of that minute, if it matters.

This will set the scene for those waiting around with a promise to be contacted when the answer is found. Or else if they come back before the answer is found, they still don’t know how long until there will be an answer.

THE ANSWER:

The answer may be “we don’t know.” It will take the number of weeks rolled previously for them to come to that conclusion.

The answer may be vague or incomplete. The answer may lie in an obscure quote or riddle. Or the document may be damaged giving just a few words. For example, the holy relic is in the tomb of an ancient warrior, but no one knows the name of the warrior or the location of the tomb.

The answer may be precise and exacting. 

This all depends on the materials available to do the research.

Time and Survival of Sources:

Vellum and Parchment are more durable than paper, especially when it comes to water, but fire, rodents, insects, etc. lead to damaged (incomplete) or destroyed (missing) information.

A well known ten year old book with lots of copies is likely to have information that is easily knowable. If the particular researcher(s) doesn’t have a copy, it’s possible they may have read it or know where to either view or acquire a copy.

A rare or unique 1,000 year old manuscript may be “lost” in that people may have heard about it, or other writings mention its general contents, but no one knows exactly what’s in it nor where it is.

Other manuscripts are lost in that they are destroyed and no longer exist.

CONTEXT: 

Are they logically asking someone who has a chance to know or find out?

A cleric or paladin asking the archivists of a temple to their own deity are more likely to find answers about relics and holy weapons pertaining to their deity, or those their temple or religion have dealt with. The main temple in the capital city is more likely to have the answer than a shrine in a remote village. Unless that remote village just happens to be near where a warrior’s last stand occurred, etc.

If they go to a temple to a different deity, while they may get help for a donation, unless the question pertains to the deity and religious lore and practice of that group, they may not get a helpful answer.

Does the source exist?

How does the source exist? Primary source, secondary source, tertiary source quotes the secondary source, etc.

How old is the information? 

How well known is the information in the present?

Each factor will weigh the likelihood of a favorable answer.

An Example:

One possible procedure would be to give the chance of an answer a percentage. Say that in optimal conditions there is a 75% chance to get the answer desired by the players.

Note: If you’ve given the players information that an item or information is in the city they have travelled to, they need to be able to get the answers. The answers don’t have to come easily or quickly. They just have to have enough “breadcrumbs” that keep leading to the answer. 

Once the PCs have the answer, they may still have to solve a riddle, disarm or avoid a trap, or avoid, trap, or defeat a monster to get the item(s) they seek. As always, the level of challenge and risk should be worthy of that which they seek.

Modify this based on the following:

Do those doing the research have a source in their possession? If no, then determine if they can get it.

Do they know what source (book, scroll, tapestry, inscription, epitaph, etc.) has the desired answer? If no, determine if they possess a source that will lead them to a source with the information.

If they don’t have any documents that will lead them to a clear source with the answer, will they be able to give the questioners anything to narrow down the search for an answer? If no, then the PCs will have to try a different approach, or ask others, or go to a different location either in the same city or a different city.

You may decide that some form of divination or “bargain” could generate the answer. Something short of a wish. Wishes should be rare and PCs should fear to use them.

However, determine how old this information is, subtract 1% per century and 10% per millenia. 

Determine how common the information is in the present day:

Everyone knows: +5%

Most People Know: +1%

Specialists Know: 0%

Experts Know: -1%

No one knows: -5%

In the following add 1% for primary source, 0% for secondary source, -1% for tertiary source, etc. Roll d% to determine how well preserved the source is with 1% being very bad and 100% being perfect condition.

Source Exists and is Common and Accurate: +10%

Source Exists and is Uncommon and Accurate: +5 %

Source Exists and is Rare and Accurate: 0%

Source Exists and is Unique and Accurate: -5%

Source Exists and is Common and Inaccurate: -10%

Source Exists and is Uncommon and Inaccurate: -5 %

Source Exists and is Rare and Inaccurate: 0%

Source Exists and is Unique and Inaccurate: 0%

Source Exists and is Common and Known to be Inaccurate: 0%

Source Exists and is Uncommon and Known to be Inaccurate: 0%

Source Exists and is Rare and Known to be Inaccurate: 0%

Source Exists and is Unique and Known to be Inaccurate: 0%