Rather than present another excerpt from Assassin’s Amulet, I thought that today I would take the time to extract a few valuable lessons from the experience of writing it. Several reasons commend themselves for doing so:
- first, there are some ideas there that we would be happy to see adopted by our peers in the industry;
- second, it gives a different segment of our readership something of value;
- third, it opens a window of insight into the content (and might lay to rest the concerns some people have concerning that content);
- and finally, many of these lessons – if not all of them – are applicable to campaign creation and administration.
Did I say a few? I’ve actually come up with a list of eighteen… which means that I won’t have time to look at any of them in great depth in this one article. Never mind, that just increases the likelihood that there will be something of value to everyone in the pages to follow….
1: Inspiration is where you find it
Every major project has its launching point of inspiration, but that is rarely enough to sustain anything large all the way to completion. For AA, it was the map of the Assassin’s Lair by Michael Tumey. The plan at this point was to provide a key / adventure skeleton to accompany that map, plus a collection of advice to the GM about how to handle Assassin NPCs.
Phase II of the development of AA drew its inspiration from the context with which we surrounded those elements, especially the new Pantheon and associated Cosmology and Theology, and was initially aimed at providing something of value to both players and GMs alike. From a book about generic Assassins in d20 Fantasy, this was where we started putting a unique and original spin on proposed content. In addition to generating substantial new content, and completely usurping some of the intended content already scheduled to appear (like the map key / adventure skeleton), this required us to revise what we had already done; but instead of replacing what we had done, we quickly decided to supplement it instead.
Phase III drew inspiration from the bigger picture. Was this going to be first of a series of works? How were they to be connected? What was the bigger picture going to be? By now, regular readers will know that the answer decreed to the first question is “Yes”; and the second is a new category of Magic Items called Legacy Items. AA contains a system for generating new examples and a number of fully-worked samples. Beyond that was the whole concept of “Legacies” – leftovers from the past that have an impact on the modern game world. This concept went through at least a dozen radical revisions before arriving at its current form; integration with AA took place prior to that final form, so this is very strongly an add-on to the book except in terms of the general principle of “Legacies from the past”.
The results are the 277-page monster we’ve entitled Assassin’s Amulet.
One point of inspiration wasn’t enough, we would have run out of steam long before we came anywhere close to the current page count; but three in succession got the job done.
2: Big Picture before detailed portraits
There were a number of occasions during the writing where we found that we had to junk something we had been working on and redo it virtually from scratch, simply because we were deciding the “big picture” after we had done the detailed work. There were times each of us lost track of where we were in the project as a result of this factor alone; writing to a concept that had been rendered out-of-date by big-picture ingredients.
This is a mistake that a lot of GMs make in their campaign creation process as well. This mistake kept biting us on the tail throughout the development of AA; each of the three phases of product development described above was characterised by a bigger picture or context being created and applied retroactively, and revisions and backtracking. Some components of the e-book became so muddled and muddied that it was necessary to lay them in front of a third party who could wring some coherence from them.
That won’t happen next time; we have already decided on the grand plan, and the bigger picture, and how the content ingredients will fit into the resulting context. We expect this alone to save us at least 3 months of writing time – the literary equivalent of prep time.
3: Have a plan to follow
The content list for AA just sort of grew on its own. Each of us was throwing every idea we had into the pot, willy-nilly. No structure, no plan. It was only when we got it organised into a plan that we knew who was writing what and how it was all going to fit together.
4: Have deadlines
We’ve had deadlines for every activity we’ve undertaken in the development of AA. Sometimes, the mere fact of an imminent deadline prompts creativity that you didn’t know was there! Deadlines keep you from wasting time on irrelevancies, focussing your attention on what needs to be done most urgently at the minute. They also encourage concision. These are acknowledged benefits when it comes to any project and AA has taken full and shameless advantage of them during its genesis.
5: Be willing to break them
At the same time, it must be admitted that AA has never yet managed to meet a single one of its deadlines. There was always a good reason for each violation, and AA is bigger and better than it would otherwise have been as a result. You should always be willing to violate a deadline if the result will be a measurably better outcome for whatever project you are undertaking.
This point, and the one that preceded it, definitely apply to game prep in general. I have a deadline of 8 PM (sometimes 10 or 11 PM if I know there’s a lot to do) the night before we are to play. By writing to that schedule, even if I miss the deadline, it’s usually only by an hour or so. As a result, only rarely do I find myself still plugging away at 3AM or 5AM on the morning of play. But I always make allowances for the possibility that I might have to do so.
6: Contingencies are never wasted
It’s a waste of time having deadlines if you don’t have a prepared plan of action for coping with a failure to meet them. This necessitates an intelligent analysis of the requirements that have to be met in order to achieve a deadline; arbitrary limits produce arbitrary outcomes. The absence of such a plan means that more time must be wasted in formulating one at the worst possible time, ie when the deadline is violated, then communicating it to everyone else involved, then putting it into motion.
The deadline for the first draft of a post-production AA manuscript, with layout, fonts, and artwork in place, was the Friday before last (my time); there was a plan in place for delivering most of the content using dummy graphics if that deadline was exceeded. As a result, a first draft was delivered within 24 hours of the deadline with 75% of the completed product’s functionality in place. The deadline for a second draft was last Friday; immediately that deadline passed, a contingency plan that had been formulated in advance came into effect. As a result, within 48 hours, I expect to deliver a second draft that will contain more 75% of the work that wasn’t done for the second draft.
The third and final draft is due on the Friday coming, and even though I have missed the deadline for the second draft, there was sufficient slack built into the schedule that it is STILL on schedule. And even if it is missed, we have an extra 48 hours up our sleeves – which can be stretched to 72 in a pinch. And, if even that is insufficient, we have an alternative marketing schedule up our sleeves that will give us a whole extra week.
Readers may question the title given to this section, thinking that if I had actually met one of the deadlines, the planning that had been put in place would have been wasted; such opinions are flawed, because there is ALWAYS another project on the horizon to which the plans can be applied. If a plan isn’t used this time around – or if it is, and works – it can always be used as the foundation for a similar plan on the next undertaking. And if it is used, and doesn’t work? That lesson, too, can be applied in the future. Contingencies are insurance policies – they are never wasted.
7: Something for everyone
The best game supplements have something for everyone – players, GMs, even something that can be adapted to other game genres and campaigns. Unfortunately, not every publisher follows this recipe. Our philosophy was that ensuring AA contained something for everyone was a necessary first step to making AA value-for-money for everyone. (If anyone’s not interested in the theory of e-book pricing, they should skip to the next section at this point.)
Of course, you can never guarantee that you can achieve the ideal of value-for-money, but you can ensure that you have a fighting chance at meeting that standard for the majority of your chosen submarket. The only alternative is to slash the price so drastically that people will take a chance on buying it.
AA was conceived from the start as a premium product – we have attempted to provide $2-value-for-every-$ charged. That leaves room for some of the content not to be relevant to any given GM within our target market while still achieving the broader objective of being value-for-money.
The final, absolutely essential, step in achieving this goal is to demonstrate that value. I have seen product previews that were so limited that they served no practical purpose – being utterly dependant on material from elsewhere in the product in order to be useful as anything more than a source of inspiration. Our plan for AA mandates that whatever excerpts and previews we offer must be essentially complete, containing everything that is needed to make practical use of the offered content. The best example is the Fifty Assassin Hooks. These are an excerpt, and an edited one at that – a chosen subset of the entirety of what is offered in the game supplement, but they are still usable in and of themselves and representative of the larger whole.
8: Expect reversals of fortune
In any project, no matter how small, there is a risk of unforseen difficulties. The larger the project, the greater this risk (because it is more complex) and the more opportunity there is for these risks to manifest. Thus, the likelihood of a setback or reversal is proportional to the square of the project scope. Or more.
At some point, a critical threshold is achieved where the probability of such reversals of fortune becomes greater than unity – at least one such problem is virtually guaranteed to occur, the question then becoming, “how many of these disasters must we endure – and how can we prepare for them?”
Too many organisations of all levels have inadequate disaster recovery plans, as anyone who looks into the subject will quickly discover. Bearings Bank, Lloyds Of London, the recent Economic Meltdown, even the financial woes of certain European economies – they were all the results of a reversal of fortune for which the target’s disaster recovery preparations were inadequate.
We certainly had our fair share whilst writing AA. You’ll read about some of them in the course of this article! And, in most cases, our disaster recovery plans were barely adequate – at best. AA has taken at least 6 months longer to produce as a result; depending on how you define “disaster,” even a year can be written off as a result. And that’s not counting the three months I reported as lost a few paragraphs back.
That’s right, if everything had gone off without a hitch, AA would have been released in June-July 2010. Or sooner (and considerably smaller in size and scope). But that’s the thing with creative people: give them some more play on the leash while coping with a disaster. and you will often get back more than was required to meet the objective. As a result, you can make lemonade from lemons. As they saying goes, when it all hits the fan, there’s only one thing to do: sweep it up and sell it as fertilizer.
9: Don’t solve problems, build a system for problem-solving
As a matter of principle, in developing AA I did not try simply to solve the problems we encountered, I attempted to construct a procedure or system for solving that problem that could spit out the answers anytime it was confronted with a similar problem.
The most obvious example of doing so is the system for constructing Legacy Items. We could have simply presented the products of that system – eight items that fall somewhere between lesser Wondrous Item and Artefact/Relic (depending on your preferred usage), that start small in effect and accumulate powers as the wielder gains in levels even as they shape the mind and soul of the wielder to their own ends. Occupying 31 pages of the e-book, these examples – together with half a dozen more typical pieces of arcane equipment – would certainly have satisfied our demand for a range of new magical items in Assassin’s Amulet. But we wanted to go one step further and presented the system itself.
There are other examples I could point to, but I think the point is made.
10: More is better
There are an infinite number of solutions to the “value-for-money” goal I espoused in section 7. Even defining AA as a premium product, ie one costing more than $10, still leaves a lot of headroom.
It’s my theory (and my co-authors may not agree) that the tipping point of prices for e-books start at a (theoretical) price-per-page of US1$ at a total page count of zero, and that this price per page declines with an increasing page count in a non-linear way. Above this price, the work is no longer considered value for money; below it, and it is.
But this won’t capture every sale that’s possible; it will only put the product on an even playing field with everything else that’s competing for the customer’s dollar, or time, or whatever manner of investment you want them to make. After capturing that percentage, it is necessary to steadily discount the product to increase its competitiveness, first by 10%, then 15, 20, 25, x, and finally, 50% or more (with x being a value somewhere between 25 and 50 – some say 30%, some 40).
The more content a product contains, the more resilient its price is – in effect, you’ve already applied one of the discount brackets I’ve cited in order to generate word of mouth and an easy purchase decision. The goal is to make up in volume of sales for the reduction in profit-per-sale. By providing 5% or 10% or whatever extra content, over and above the value-for-money maximum, you effectively discount the base price that you could have asked by that amount, yielding the actual price you are charging. In the process, you also move that value for money maximum, so that the effect is actually greater than the extra content ratio.
In theory, a standard dumbbell probability curve should describe the result of sales multiplied by profits-per-save; there will be an optimum point at which further reduction will not increase profitability. It should only take a quick glance at the graph above to see that the maximum profitability is exactly midway between the maximum price you could have asked and the minimum price (usually zero or virtually so).
That was the line of thought that led to the goal of $2-value-for-every-$-charged.
Of course, reality is more complicated. For one thing, “value for money” is actually a perceived value, not an actual one. There are other factors that pull or push that midpoint up or down – publicity, reviews, award nominations and wins, reputation, and a whole heap of other variables. But one of the biggest ones is the first one.
You have to assume that not everything that you include is going to be valuable to every reader. The “perceived value” will almost always be lower than the theoretical value by some unknowable and variable percentage. It follows that the more value-for-money you can build into a product, the more likely you are to push the perceived value beyond the asking price, even after that discounting takes place.
Bottom line: the more content you can include for any given price, the better your sales are going to be, and the longer you can wait without discounting the asking price. You can never include too much content.
Here’s another way to look at the question: How many 1-4 page PDFs could you create from the master document at 25-50 cents per page? An average of page-count / 2.5 – but some of that space would be taken up with overhead: logos, legal, and so on, so the actual number could be as much as 40% higher. If your actual price is less than or equal to (page-count x 1.4 / 2.5) x $0.50, you should be somewhere at or below the critical price threshold . The further below that calculated maximum price you are, the greater the allowance for perceived value and other such variables.
Assassin’s Amulet weighs in at 277 pages – a theoretical maximum value using the small PDF model of $77.56. It’s actually higher because there is reinforcement of value due to commonality of theme. Half that gives a price point that should be somewhere in the vicinity of the maximum profitability – $38.78. The lower you drop the actual asking price relative to that total, the more of your content a reader can find valueless to them without impacting the perceived value-for-money of your product.
Of course, we’re talking actual content, not filler, and that requires an act of creativity on the part of the authors. For that reason, we listed every subject related to Assassins and Assassination that we could think of and then set about writing one or more articles on each of them. In every way, we tried to exceed requirements. When all is said and done, almost half of Assassin’s Amulet is such bonus content relative to the theoretical price point calculated in the previous paragraph..
11: Core values give a creation heart
At the start of this article, I described the first phase of development of Assassin’s Amulet. The results were scattered, with no unifying theme, just a subject in common, until we entered phase II. With the creation of the Goddess Cyrene and the definition of her relationship with the Assassins, and the resulting redefinition of the Assassins (I’m being careful to avoid revealing any plot twists!), there was suddenly a context into which all the content could be placed. From a scattered collection of disparate articles, there was unity.
12: Buzz, Promote, Market
We developed a three-stage marketing plan for Assassin’s Amulet. Step one was to generate buzz about the product with blog posts of excerpts and other word-of-mouth-generating activities. Unfortunately, circumstances have left Johnn & I with too much to do just to get the product out the door to have implemented everything we wanted to do. The cover contest wasn’t originally part of the plan but was incredibly successful in this phase – so much so that we may well run other contests in the future when we have a Legacies product about to hit the digital shelves.
Stage two will be a limited public release and the generation of a number of free extras, as well as the creation of one or more sample PDFs and a few other activities to ramp up the excitement prior to the general release.
Stage three will start with the general release and will be designed to further boost public awareness. We might also be able to increase the number of formats in which Assassin’s Amulet is available – Johnn is keen on a Kindle version, and I would like to get AA hooked up for print-on-demand. How far we go in this phase depends on sales.
13: Modular thinking
A logical consequence of the notion that not everything in AA will be valuable to every GM is the corollary that GMs will want to cherry-pick what AA-content they integrate into the campaign. This can often be a nightmare because slices of an e-book (or printed game supplement) are mutually dependant, so tightly integrated that it can be difficult extracting just the pieces that you want to use.
We wanted the integration without the difficulty of extricating.
The stylistic and structural changes that had to be made in order to achieve this, which we dubbed “Modular Thinking”, proved to be far more straightforward than we expected. We had dedicated an entire chapter to the subject of integrating the different elements of Assassin’s Amulet in an existing campaign in a piecemeal fashion, with the intention of contributing to it with every section of content as it was completed; we ended up needing two pages. The rest was achieved by compartmentalising and structuring content as we went. The only reason it remains a chapter in its own right is as a reflection of the importance we attached to the subject.
It was so effortless, in fact, that it astonishes us that more publishers don’t make it a priority; we can only assume that their assessment of the difficulty mirrors our own initial expectations and not reality.
Assassin’s Amulet is seamlessly modular; everything within it can be included, or excluded, by the referee as he or she sees fit.
14: Identify problems, devise and present solutions, then analyse for common threads
A major effort went into brainstorming the problems that GMs experience when running Assassins, either as PCs integrated with the rest of the adventuring party, or as NPCs (the more common situation to be encountered), then devising solutions and advice to cover those problems.
Problems were identified at a metagame level, at the Player level, at the Campaign level, at the encounter level, at an organisational level, at a social level, even at a theological level. In each of these cases, we had to confront assumptions that we had made, and that other authors addressing the subject had made in the past. Through such analysis, we learned that the problems were not inherent in the class per se, but were all connected to the integration of the class and its corresponding social and organisational structure with the rest of the game world.
Even if the more traditional approaches to the concept of Assassins that we had held in Phase I of the writing of Assassin’s Amulet are retained by the GM, the problems involved have been identified and resolved for the GM – making it possible to employ the class without the problems that usually corrode verisimilitude.
While it would be going too far to characterise this content as ‘easy’ to create, this was – ultimately – a simple and straightforward task compared to the problems that we expect to encounter with Legacies #2, where fundamental questions and assumptions of morality and chivalric behaviour and alignment will be squarely at the forefront. We expect the work on the problems with Assassins to prove itself a valuable warm-up when the time comes!
15: Work both sides of the moral straightjacket
Assassins are inherently evil, aren’t they? What about ‘honourable hatchet men’? And exactly what is “evil” anyway? These are just some of the moral “background” issues and assumptions referred to in the previous section that we had to confront in writing Assassin’s Amulet. Rather than burying them and hoping these difficult questions would go away if we ignored them assiduously enough, we chose to confront them head-on, deliberately moving beyond these traditional interpretations of the role of the Class. In this case, we looked at what happens when you answer that first question (“inherently evil”) with Yes, with No, with Maybe, and with Sometimes.
And, as usually happens, we found that simplistic preconceptions conflict with complex and realistic characterisation and character behaviour. Deliberately working both sides of the moral straightjacket and challenging our assumptions proved to be liberating, as it almost always is.
16; Don’t be afraid of scope
One of the most fundamental transformations in AA, from its initial vision to the final product, is the scale and scope of the product. When we started we were looking at one band of Assassins, possibly atypical, in isolation, with the intent of providing the GM with everything needed to handle that one isolated group. This is not unusual, in fact it’s typical of such game products throughout the industry.
With every step toward satisfying that brief in full, the scope of the work we had undertaken grew. If there is a fundamental difference between AA and everything else on the market, it is the depth and scope with which we have attacked the subject.
Sure, it’s more work – but it lets you really get your teeth into the subject, achieving insights and verisimilitude that simply can’t be bought with a more superficial approach. We don’t want to tell other publishers out there how to do their jobs, but we hope that the willingness to enlarge the scope of a project, as opportunity presents, will serve as a positive example to others.
That’s not to say that AA contains everything it possibly could – it doesn’t. It holds everything needed to cover the principle subject, and anything at the immediate periphery of that subject; it rarely treads beyond that limit. The idea is that each Legacies product will do the same, slowly building up a mutually-consistent jigsaw of background content. This is scope in a different direction, and we wish we saw it more often – if it is presented in a modular, a-la-carte fashion.
In other words, not only is AA bigger than we expected it to be, it is but a part of a still larger structure that will span multiple products and game aids. How many? That depends on sales, to be honest. But that scope also carries a corresponding responsibility, which demands a commitment from us as the authors, so here it is:
We give you our word of honour, here and now, that you will NEVER be asked to buy a second product in order to use something you have bought from us.
The idea is that customers who have purchased AA will gain access to an added layer of value when they purchase a subsequent Legacies product, not that they will be forced to buy AA in order to use Legacies #2. We would rather give away any essential content for free, or duplicate it in Legacies #2. Each piece of the jigsaw will be self-contained.
17; Tools must be reliable, everything else is negotiable
When we started working on AA, we needed a platform through which to collate, coordinate, and discuss our efforts and ideas. We started by using a platform called ClockingIT, which not only contained all the functionality that we thought we needed but some extra bells and whistles that proved extraordinarily useful. It was the price (free) and the functionality that had led us to this choice.
The only problem was that it was intermittently unreliable, and this period of the development was characterised by a steadily-rising level of frustration. Eventually, this reached the point where we bit the bullet and migrated the entire project, and every word that had been written about it, to Google Docs. This lacked some of the essential functionality, and some of the bells and whistles that we had started to exploit; but it worked reliably, and that was more important.
That migration took weeks – two or three months, in fact, plus time spent getting through the learning curve of mastering the operations of the new platform. One of the things we had liked about ClockingIT was that it was so instinctive to operate that there was virtually no learning curve. I took advantage of the necessity to organise and structure the work that had already been done, which proved very helpful in the end, so it was not wasted effort by any means – but it was several months during which development came to a standstill.
Let’s be clear about this: For my money, ClockingIT was the better platform, but Google Docs was the more reliable. If we could have hung around until ClockingIT solved their problems, we would have done so. There were two hard lessons for us in the migration: (1), Organisation pays for itself in reduced development time in the long run, so don’t put it off, do it first! – and (2), Tools must be reliable, everything else is negotiable.
18. Standard Content, Custom Content
When migrating our development work to Google Docs, I took the opportunity to organise it into two categories: content types that would appear in subsequent Legacies products (if any – we had not committed ourselves to anything beyond AA at that point), and content that was exclusive to the subject of Assassins and to AA.
For example, the offering of Plot Hooks is something that we expect to repeat in each and every Legacies product. “How to terrify your opponents with Assassins” is not going to have any equivalent in Legacies #2; instead there will be something else in the content list that is exclusively relevant to the subject of that product, just as “terrify with Assassins” is relevant to AA.
Conclusion: Lessons Learned
Whew! The end at last!
There are lessons for us as producers of game products in the 18 points above that we have learned the hard way, and lessons that we discovered because we went out looking for them. There are lessons that are applicable to other producers of RPG supplements, some positive suggestions and some cautionary tales. Most of these topics are at least loosely analogous to campaign development lessons that can be learned. Every d20 campaign has some “standard content” and some “custom content”, for example – questions that have to be resolved each and every time a new campaign is started.
Hopefully, the preceding has given you some insight into the development and content of AA and whetted your appetite for the actual content. Or at the very least, offered some solutions to problems you didn’t know you had.
And now, back to work…