I employ a structured and systematic approach to NPC design. Today, I’m going to give readers the low-down – everything they need to use it for themselves.

A system is a means of organizing data or steps in a process into a structured form to facilitate analysis or use, and that’s what today’s article is all about. This illustration represents a character emerging from the data that has been generated by the system. It’s another great image by Gerd Altmann from Pixabay

The system revolves around a text document or sheet of notepaper, and an 8×4 grid on another sheet of notepaper – and something to write with.

As ideas come to me or are generated, they get listed in the text document / notepaper and numbered, and then allocated to one or more specific positions in the matrix. That’s the ‘structured’ part of the approach.

Those ideas are frequently generated based on the spaces to be filled – that’s where the ‘systematic’ part of the approach comes in.

The columns of the matrix deal with the usage / relevance within the campaign of the content indexed; the rows categorize the content by type.

I deal fully with the first column (with the occasional side-trip into the second column) to start with – and for many NPCs, that’s as far as I need to go.

The Columns

Column 1: Immediate Needs

The first column deals with the attributes that the NPC needs to have to satisfy his or her purpose or role in the immediate adventure, i..e. the first one in which they are to appear.

Every NPC is created for a purpose, even if that purpose is nothing more than “there would logically be someone.” This column exists to make sure that the NPC is fit for that purpose.

Column 2: Interesting & Unique

I like to make most of my NPCs at least a little distinctive. If their role in the intended adventure is minor, there doesn’t need to be too much in this column, but if the NPC is too have a more substantial role in the adventure, it might need to be more substantial, and if the NPC is to be a focal point, they definitely need significant entries in this column.

Column 3: Long-Term Usefulness

If the NPC is intended to be a recurring character, this column contains material related to any anticipated future appearances, even if those plans are vague at the time of creation.

However, just because there are entries in this column, it doesn’t necessarily mean that it’s certain that they will appear in future adventures; this simply prepares for the possibility, and ensures consistency over those potential repeat appearances by listing attributes that might make them useful in a future appearance.

In particular, I like every NPC to have an ongoing storyline of their own, a personal life character arc that the PCs are going to intersect at one or more points. This is where ideas for the NPCs future get listed.

Column 4: Long-term Potential

Even more vague and nebulous, and less-likely to ever actually materialize in game, this column deals with in-adventure roles that the character might occupy, depending on all sorts of X-factors – the course of the campaign, prior interaction with PCs, possible future adventures, and so on.

Or, sometimes, the lack of same. One NPC that I created was a businessman’s secretary, and in this column I noted that every time the PCs interacted with that businessman, he would have a different secretary. Nothing about why, just an ongoing minor plot point about the businessman burning through Secretaries at a relatively extreme rate.

As it happens, the businessman never became a recurring character, so this trait of the ‘secretary revolving door’ was never triggered or explained; the idea is still out there, ready to be grafted onto another character should it fit.

The Rows

Row 1: Concept / Archetype

The first row deals with broad ideas that define the NPC. These then act as guidelines to the development of content in the other Rows.

Row 2: Stats & Abilities

This deals with things that I need the NPC to actually be able to do. Sometimes there’s a lot of content in this Row, sometimes there’s not.

Row 3: Equipment & Possessions

This row frequently has only one or two entries; the contents are not considered an exhaustive listing, these are just the items that are potentially significant.

Row 4: Personality

Depending on the NPC in question, there might be a lot of little details here, or there might by only one or two, which might be described more expansively. There can even be a single brief referent which signposts greater depth – “Max Klinger from MASH”, for example.

Row 5: History & Motivations

It’s natural to segue from a character’s mental and emotional state to questions of why they are doing whatever they are supposed to do within the adventure. There’s nowhere near enough space to be comprehensive. This area of the matrix only contains the most important, key, points.

Row 6: Ambitions & Plans

From the reasons they are doing whatever they are doing to what they actually hope to accomplish. It’s worth noting that this might have absolutely nothing to do with the role that they are take within the immediate adventure beyond leading to a circumstance of “wrong place, wrong time” – but it would mean that they weren’t in that place at that time for no reason at all.

Every character should always have a reason to be wherever they are when encountered, even if that reason is “habit” or “routine”.

Row 7: Relationships

Entries to this area of the matrix get an additional notation – a single asterisk if the relationship is to a PC, a double if it’s to a significant NPC.

Row 8: Visuals

Entries in this row are often single-word elements that can provide the foundations for a narrative text-block description.

Abstract Example

I put hours of effort into creating a visual example to present only to realize the morning after (as I write this) that what was supposed to be dragon-scales looked like baked beans in tomato sauce. One hasty edit later yielded the image below – one that I’m not totally happy with, but which will at least do the job.

Trying to fit a substantial amount of text into the label space was another challenge, and again, I’m not totally satisfied, but the results are good enough.

Idea / Content Generation

There are six sources that I rely on for ideas. The first three only generate content, the last three both generate content and filter other ideas to exclude unwanted ideas.

1. Idea Generators posted at Campaign Mastery, especially those covered in The Characterization Puzzle series, and
2. those described / listed in the Breaking Through Writer’s Block series (yes, there is some overlap).
3. Stereotypes & Variations Thereof
4. Playing against type (the opposite or inversion of a stereotype)
5. Free Association based on the intended adventure role
6. Free Association based on being different to recently-appearing characters
7. Free Association based on being different to characters that have occupied a similar role in the past.

Intended Role as a Filter

The number 1 filter is to reject ideas that don’t contribute towards the character serving their intended campaign purpose (or worse yet, actively oppose that purpose), at least for the first column. There’s greater flexibility in the other three.

Recent Characters as a Filter

This is less important in the first column but significant in the third and fourth and absolutely critical in the second. At the same time, though, you have to be careful not to go too far – the character must at least be plausible within their role, and you can only justify deviating from that plausibility so far before you enter “fish out of water” territory. Sometimes, that’s acceptable or even desirable, but most of the time, it does nothing but erode verisimilitude without gain.

Similar Characters as a Filter

This filter follows the same pattern as the previous one, but instead of looking at the characters that have recently appeared in the campaign, it seeks comparison with others placed in similar roles within past adventures.

How similar is ‘similar’? Every GM is different. Is a Personal Assistant the same thing as a Secretary? Is a Gopher too much like a Personal Assistant to tolerate similarities in character traits?

It’s often a good idea to view the character holistically when making these judgments, because its easy to confuse function with personality. Two characters who excel in similar roles for entirely different and distinct reasons are probably distinctive enough to be going on with. But, again, your tolerance levels for such things, and those of your players, might be different to mine / my players’.

Regardless of where they come from, you need a source of ideas. Once you have one, you’re ready to move on.

The Process

I like to start with the biggest ideas, the broadest strokes. Sometimes those will belong in column 1, sometimes column 2, and on rare occasions, columns 3 or 4.

Ideas should be expressed as a single, short, declarative sentence, one to a line. Number each and place a number in the matrix in the appropriate cell.

It’s worth observing that for reasons of clarity, I made the example above a lot neater and more legible than the real thing would probably be. In particular, division of each row into sub-rows probably wouldn’t happen unless you were using pre-ruled notepaper.

Everything that you’ve already listed becomes context and inspiration for everything else – inconsistencies have to be avoided or explained Immediately, and everything else that even gets considered has to be judged in light of the context provided by the existing content.

At the same time, though, some ideas can explicitly nuance, refine, color, or shade that context. Even at the end of this process, nothing is set in immovable blocks of stone.

I also want to remind readers to follow the principles outlined in Creating Partial NPCs To Speed Game Prep and Holistic NPCs: Creating Special Characters!

The first column – Immediate Needs

Only list those items that are absolutely essential to an internally-consistent character on this pass. I can’t emphasize that enough.

1. Ramifications & Expressions Of Core Concepts

The first items that I look for or try to generate are any logical consequences or ramifications of the core concept. Is there anything that the character logically has to have about them to make the core concept work?

Secondly (but usually at the same time), how can that core concept manifest itself so that the players & PCs can observe it – or how can it be concealed, if I don’t want them to be able to observe it?

2. Additional concepts

One primary idea is rarely enough for a realistic characterization. So the next step is to bolster the main concept with something supplemental or secondary. While this will often go into cell 1, column 1, there will be be times when it’s a better fit in column 2 because it makes this character more unique.

3. Further Ramifications & Expressions

The obvious third step is to further list ramifications and methods of expressing the additional concepts, and to note any variation on the existing information that result from an interaction of the concepts.

4. Stats & Abilities – Implied and Explicit thoughts

There may be one or more entries in this cell as a result of steps 1-3 above, already, so there may be no further work needed. So the focus has to be on the role within the adventure and any stat levels or abilities that will be needed to fulfill that role. These are generally expressed in general / relative terms, and sometimes relative to a particular PC.

5. Equipment & Possessions

This category might well be empty if there’s nothing noteworthy. On this pass, specifically exclude anything that comes with the archetype / character level as a default assumption – for example, if you are generating a medium-high level D&D / Pathfinder Fighter, it can be taken as read that they have a magic weapon, enchanted armor, and possibly a magical shield as well.

If the character were a Lich with multiple levels in Fighter (a “Martial Lich” being the core concept) then they might have Necromantic Weaponry and Armor – and that would definitely count as something noteworthy.

6. Personality & Focus on Distinctiveness

A general personality type can often be summed up in just one or two words – such overarching summaries are what generally get listed in this pass in column 1.

But you rarely think of these ideas in isolation, they usually come with thoughts regarding distinctiveness or uniqueness – so, at the same time, skip ahead to enter those details into column 2.

7. History & Motivations

There’s certainly (and intentionally) not enough room for a detailed biography or psychological assessment. Instead, one or two critical moments / concepts, which can be built upon later, will do. It will often be the case that there’s nothing critical to the intended role in the next adventure that needs to be listed in this space, though if there are to be any particular motivations to assist or resist cooperating with the PCs, they should be spelled out at this point.

8. Ambitions & Plans

For the first time, this looks beyond the immediate needs, because nothing makes a character seem more integrated into the campaign world than having some ambition or plan, especially if it goes beyond the immediate. That said, it needs some concrete manifestation within the immediate encounter, which is why it gets listed at this point.

For example, a functionary might have a poster or painting of a beach, or snow-capped mountains, and something to indicate that they are saving for a trip. It doesn’t matter how near or far this ambition is – its in-progress, is all you need to know.

9. Relationships

At their simplest, inter-character relationships are a series of nested bubbles, with other characters occupying a position within a bubble.

The number and definitions of such bubbles are matters for the GM to decide; often, genre will make a difference.

An example hierarchy of bubbles might be:

   a. Complete Strangers
      b. Reputation of one known to the other
         c. Each known to the other by Reputation
            d. Mutual Acquaintance
               e. One’s Friend is an acquaintance of the other
                  f. Mutual Friend
                     g. One is directly connected to an important NPC of the other
                        h. Directly connected to a PC

In the diagram example above, this has been simplified to a three-tier system:
   a. No direct connections to anyone noteworthy
      b. Direct connection to an important NPC;
         c. Direct connection to a PC.

Decide on your strata of relatedness and indicate if the NPC being generated has such a connection, and if so, to whom and at what level it is.

10. Visuals

There will often be nothing in this category at the end of this pass. But sometimes there will be a significant marker to be placed here.

I once generated a character whose core concept was “Dumb Blonde” (a cliche, and recognized as such). In this category, I listed “brunette, dies hair b/c she feels more comfortable as a blonde”. And, notably, in the stats section, “Not as dumb as she likes to pretend” and “Shy, lowers expectations.” — The cliche in this case is just a gateway to a far richer and more plausible characterization.

The Second Column – Distinctiveness

The second pass fills in enough items in Column 2 to make the character distinctive, i.e. to pass the ‘filtering’ idea-generation tests.

You go through the same steps as above, but this time with a different focus. You shouldn’t list anything that doesn’t directly contribute toward making the character unique / distinctive, but you also should not skip any space, even if the decision is ‘nothing to enter here, move on’.

Beware of making the character too complicated to roleplay while you are doing three or four other things at the same time, for obvious reasons of practicality. And if voices / accents aren’t amongst your core skills as a GM (they certainly aren’t in mine), you may need to note a way of working around that limitation – or you may need to deliberately exclude them from your creation.

If you list only the essentials in this pass, you generally don’t go too far wrong.

Column-spanning entries

If you look closely at the example matrix above, you’ll note that some entries appear to span two adjacent columns – I actually did my best to make this fairly obvious, but haven’t explained it until now.

An example is entry #1 on that example matrix. By now, you know that this refers to the core concept of the character – by having it span both columns 1 (Immediate Needs) and 2 (Distinctiveness), it is a note to yourself that you think this core concept is itself a point of uniqueness or distinctiveness about this NPC. Obviously, that won’t always be the case.

A column-spanning entry, also occasionally thought of as a ‘bridging’ entry because it bridges the gap between two different sets of considerations for the character design, simply means that you think that entry is relevant to both.

It’s not impossible to have a single entry bridge all four columns, but that would be comparatively rare.

The First Column – Clarity & Nuance

The same ten steps again – but this time, you want the uniqueness / distinctiveness to manifest in the immediate at-hand appearance / behavior of the character, so you are looking for ways to do that. Again, you want this to be something the players, and of course the PCs, to notice.

Assessment #1: Fit For Purpose

By this point you have the first two columns pretty much fitted out with everything you need for immediate play – at least in theory. Time, then, to take stock: Is the character that you have created sufficiently fit for the intended purpose within the forthcoming adventure?

The inclusion of the term “sufficiently” is an important one. It’s entirely possible to have a character that is only just adequate to the immediate intended purpose because you can already see glimmers of a character arc or longer-term functionality for the character – purposes for which they are sufficiently well-suited to compensate for the immediate shortcomings.

So this can be a quite nuanced decision.

Decision #1: Start Over or Accept

If the answer is “no”, then file this character away for the next time you need to create an NPC – either this one will or won’t be a fit for that purpose, but it’s not a fit for this one – and go back to square one.

If the answer is “yes”, then you can proceed to Assessment #2.

Assessment #2: Recurring or One-Off?

No-one can see all ends, but it’s often possible to see glimmers of potential. Sometimes players relate particularly strongly to an NPC, encouraging the GM to give than NPC a recurring role; you can’t be certain of that in advance.

A useful metaphor can be the campaign as the painting of a scene, and the NPCs, the tones and shades of paint. Sometimes, having used a particular color, you’re done with it; sometimes, you can use it again for something else, especially if you mix it with a different color to create a tonal variation.

Suggestion: One Toe In The Water

It can often be a good idea unless this NPC is deliberately to make a one-time appearance, to dip one toe into the water. That means that instead of a full pass through columns 3 and 4, you simply list one or two entries of ways that you might reuse the NPC if the campaigns’ long-term plotline develops in a particular direction – that’s usually more the result of the tension between the players and GM pulling the campaign in the direction they want it to go, and isn’t particularly subject to accurate prediction in advance.

Or it might be that you can see a definite future role for this character in one or more adventures, in which case, the choice will be “recurring” and it should be given the full treatment.

Assessment #3: Potential Roles in Future Adventures

That leads into Assessment #3, spelled out in the section title above. Remember that a lot of water can flow under the bridge in between – sometimes life for an NPC can be fairly crappy. So you’re looking mostly at ways that the unmodified character can or logically would make a return appearance, AND ways that the character can evolve to achieve suitability for a different role in a future adventure.

Sometimes, that evolution can be positive – someone gets a degree, or gets promoted, or gets lucky in some other way; sometimes it’s not, and oftentimes you have a blend of good and bad.

Make note of any contradictions in assumptions – if a character is evolving, they may no longer fit into a picture for which the character is currently suited. But otherwise, don’t sweat them.

“If the character evolves toward Role X, they can be used as a flunky in Adventure Y.

“If not, they could be used as the Victim in Adventure Z.”

Decision #2: Commit or Further Develop

Time to fish or cut bait. You already know that the character fills your immediate adventure needs; the question is whether or not you have good reason to invest additional time, attention, and creative juices into them, or if you should commit to the character as they stand.

If the character is likely to have a recurring role, or if there’s some potential for a repeat appearance somewhere down the track, then the answer should be “Further Develop”; if not, then “Commit” and be done with it.

If you choose to “Commit”, skip the next couple of steps (labeled “Further Development,” logically enough, and head straight down to the “Commit” stages of the process. Otherwise, move on to the steps listed below.

Further Development: Column 3

Before you start on Column 3, you need to make absolutely sure that you understand the differences between columns 3 and 4.

Column 3, “Long-term usefulness” is about adventures that you already have at least vague plans for, and is about what makes this character potentially useful within those specific adventures.

Column 4 is more speculative. It’s investing in the potential of the character to be useful in the long-term; that potential might never be tapped, that depends on all sorts of factors beyond your control at this time.

Each of these columns has one additional nuance – “Character as is” or “Character evolution”, as suggested above. But “Character as is” holds a particular trap.

As shown in Lessons From The West Wing III: Time Happens In The Background, “Character As Is” doesn’t meant that the character will not or should not evolve following their immediate appearance – on the contrary, they definitely should! It means that the evolution of the character does not relate to making them fit for a future role, it’s just the passage of time manifesting itself.

If you can make that evolution some consequence of their appearance in the immediate adventure, i.e. interaction with the PCs, so much the better, for what should be fairly obvious reasons..

So, having clarified all that, go down column 3 and think about your plans for the future of the campaign, and how this character can fit into them. Make notes on any required character development that would be needed, and don’t be afraid to say to yourself “this character could potentially fill role X, but there are others better suited to that”.

Further Development: Column 4

Once you’ve gone through the eight rows looking for connections to existing plans, it’s time to go through them looking for future potentials. Be as speculative as you want, but bear in mind the probability of those speculations being relevant – if development effort never gets used, it’s essentially wasted time.

I love characters who can be transformed by a plot twist in their personal or professional lives into something completely different. A key theme of my superhero campaign at the moment is “Allies becoming enemies, and enemies becoming Allies”, and there have been several examples of that happening within the campaign already. I’ll mention a few by name without going into specifics (there isn’t enough time for that here, and many of these developments have been described in other posts already):

• Voodoo Willy.
• Thanos (see: Pieces of Creation: Mortus)
• Behemoth, founding member of the team.
• House Ares of Demon (limited)
• E-III
• Dr Heinreich Vossen, the Monster Maker
• Skygge
• UNIT
• Holo (ongoing)
• Dr Muerte (limited)
• Maynor Estuado Morales
• The Rheezok (ongoing)
• The Zhehu (ongoing)
• Three Hazelwiches (ongoing)

Those are, I think, enough to show an ongoing pattern. Some of the Reformations came easily, and some of them required significant effort to achieve. Some seemed almost inevitable, others seemed highly improbable before they happened. In some cases, the current relationship is more a netter of Neutrality than full alliance. And there’s still lots of scope for relations to sour.

But it’s enough to demonstrate the concept.

Further Development: Back to Assessment #1

Once you’ve finished going through column 4, it’s time to go back to assessment #1. Columns 3 and 4 can amount to significant revision of the character design, or more specifically, making room for the possibilities can create that sort of revision. So the assessment process needs to be redone, just to make sure you haven’t made a mess of the immediate plans.

Continue through the assessment process until you reach decision #2, which should now flow automatically to the “Commit” process below, because you’ve already done the “Further Development” that you can anticipate as being useful.

Commit: Concept, Description, Personality, Manifestations, Other notes

It’s time to turn the compiled notes into narrative text, ready to refer to. This is the first stage of turning concept into executed reality. After this is done, you can attach whatever hard numbers you consider necessary and have an NPC ready to run.

As you can tell from the above, I’ve subdivided the Commit process into 5 semi-stages, each of which is contained within its own narrative block.

Commit: Concept & Role in Adventure

These are notes from you to you. It’s probably never going to be necessary for a player to read them. As such, be as succinct and to-the-point as you can be – you will have a lot on your plate when the time comes to reference them, remember.

Commit: Description

This is the first narrative block that is likely to be read, in whole or in part, to the players. I tend toward brief semi-paragraphs that can be inserted into the adventure as seems reasonable; there are lots of occasions when the full details won’t become apparent until after some level of interaction between NPC and PCs. I’ve lots of advice for making these more compact and more useful, presented following the rest of the Commit process.

Commit: Personality

Ideally, you never want to quote from this section; instead, the manifestation of personality in play should be such that the players could recreate this section.

The real world is rarely so generous. Accordingly, I divide this information block in two, which I have listed separately. This part is one big paragraph that outlines the personality, and any psychological quirks and their (apparent) sources.

Commit: Manifestations

The second part of describing the Personality is a series of small memos-to-self on how to convey the personality to the players – how it will manifest in appearance, speech, mindset, etc.

But I also throw in any manifestations of other parts of the character design that should be noted – think back to the “dumb blonde” example.

Commit: Other Notes

Finally, I synopsize related notes from columns 3 and 4 into a coherent passage of text, again essentially a memo-to-self. It’s important to spend time on this; while you should remember everything clearly, having just compiled it, once the immediate appearance of the character is complete, you might not even look at these notes for months or years. Your thinking now needs to be completely clear to the you that will be reading the notes then, and that can be more easily said than done.

I have seen others suggest waiting a day or two and then returning to these notes just to make sure that they are clear. The problem is that this is long enough that important details may be lost to memory, and not long enough that all the important details will be forgotten – so it neither fully tests the notes you’ve made nor fully protects you from any inadequacy.

Unfortunately, the only alternative is to either wait longer (even more dangerous) or spend what feels like an unnecessary amount of time getting it all on paper here and now. Which isn’t much of a choice at all, really.

Tips, Tools, Techniques, and Further Reading

There’s lots of content here on Campaign Mastery that can help you polish your creations and the narrative used to describe them. I thought it worth the effort to list some of the most useful:

• The Secrets Of Stylish Narrative
• Stealth Narrative – Imputed info in your game
• Stream Of Consciousness: Image-based narrative
• The Flunkie Equation – quick and easy Hors d’Combat
• How Much NPC Description Is Enough?
• An Excess Of Character: Faster NPCs

There are also a number of race / species oriented articles that will apply in some cases:

• Alien In Innovation: Creating Original Non-human Species
• The Four Frontiers Of ‘Alien’
• Ergonomics and the Non-human
• By Popular Demand: The Ergonomics Of Dwarves

So that’s the more systematic method of designing NPCs. This is a process that I use regularly (though I haven’t previously formalized it) and can vouch for. The intention is to invest prep time in the parts of the NPC that will matter, while not ignoring the longer-term potential of the character being created.

Sure, you can just whip out a set of dice and start rolling stats, then try and hammer and file the resulting character into shape – changing the adventure as necessary. It can be a lot of fun reducing Chaos to Order, and if your improv skills are up to it, you can save a lot of prep time – time that you might not have (we all lead busy lives these days).

And every now and then, it will all go pear-shaped on you. While there’s lots of advice here at Campaign Mastery and elsewhere on how to recover from these disasters, its usually better to avoid them in the first place.

That’s where this article fits in. Hopefully, readers will find it useful!