Publishing Polyfill #2: Market Research
A few days ago, I rambled incoherently about the state of the publishing industry, and then theorized that the answer may lie in the creation of a new kind of manager. There were four distinct elements to that theoretically manager role:
- Market Research;
- Audience Relations.
As with any issue in the 21st century, you can solve these problems with web services (duh), which is what I'm going to propose today for the first item, Market Research.
If you've never seen market research in action (in a creative industries sense) then you are really missing out. For just about any niche you can imagine, there are detailed reports available, breaking down a potential audience by race, age, gender, sexual orientation, education, career, income level, propensity to spend money — with the ability to look at subsets of each demographic as they intersect with each other. Asian stay-at-home moms between 30 and 37 who are planning to return to the workforce within the next 5 years? Yup, we got that number right here.
The thing about market research is that it tells you how many people want something, how many don't, and how many might be interested under the right circumstances. If you've got a widget you're trying to sell, you can find out who to target; if you're looking to build a widget to suit demand, you can see where the best opportunities lie. Either way, the numbers help guide your decision-making process — which is exactly what our prototypical managers need.
The issue with market research is that, like most things, it is a gated community. The best outfits are run in-house at large media companies, operating either through surveys or just the company's own internal analytics. Every time you watch a Netflix show, they record absolutely everything: your demographic information, the genre and style of the videos you stream, how you stream, where you stream, when, how long you go before pausing, what you do when the show is done. All that information tells them how to customize their next production to suit taste trends — and they don't share the data with anyone, only the abstract conclusions.
Getting access to quality market data is a difficult thing for an up-and-comer, and without that data, they're really just making educated guesses. That can get you a long way, of course, but it's only really effective in the big, broad categories like romance or thrillers: yes, those are popular. Everything else is murky.
Building a Market Research System
There are two basic ways to get information out of people: analytics and surveys. Analytics work by extracting data from people's behaviour as they interact with the world — Google knows everything about you because they're tracking every site you visit, and can make inferences based on the content you see. Surveys, on the other hand, ask you to voluntarily hand over your data with the understanding that in most cases, you'll stay relatively anonymous.
The privacy rights side of me dislikes the analytics angle on principle, added to which it's very hard to get into the analytics game if you're not already a massive corporation. That leaves surveys, but surveys have extremely limited scope most of the time.
I don't remember if Goodreads started off this way, but they have definitely become a survey system for Amazon (their owner). What started out as a distributed rating and review site turned into a voluntary analytics engine (and has since turned into a Katamari-esque trashball). "Tell me what you like," they said, "and I will tell you what you'll enjoy."
That nugget isn't all wrong, but it needs a slight re-alignment, because if this new system is going to be useful for managers, it needs to be more about the future than the past or present.
Here is the proposition of the new market research site: tell me what you want, and I will help you get it. For readers, the system would look like this:
Capturing niches: Users can choose genres and subgenres from a list, or input book titles directly. Either way, market data is being generated in a deliberate way.
Risk tolerance: Users can also set their risk tolerance (either globally or in specific niches). Some may only want to read books by marquee authors, and others might be keen to try undiscovered voices, too.
Spending habits: Users can also set their estimated monthly budget on books, because recommendations both to the reader and the manager can change greatly depending on this variable.
"I want more" tokens: Users can give a certain subgenre a boost by allocating one of, say, three special tokens that say "I would read more of this if it were available".
Genre/subgenre pages: All this data would be rolled up into special collection pages where the likes and preferences of the like-minded community would be aggregated and displayed, to help readers find new content that doesn't just fit their tastes, but their budget and risk tolerance as well. Lists would include sections like "most liked", "newest" and "trending".
Now, rather than creating a site that pretends to be something else (while still devouring personal information) this site would be upfront with its mission: to gather reading preferences and feed that data back to the creator community, so they can identify and reach niches more easily.
How does that happen? A few elements are at play:
Open statistics: As data is input, reports will be automatically generated showing not just which niches are most popular, but information about likely market size, purchasing power and risk tolerance. Any registered user can review this information and use it to craft business decisions.
Manual listing: Authors or their representatives can manually enter their titles into the system, targeting their specific subgenres and preferred audience type. This won't necessarily make them popular, but it makes them discoverable.
Direct contact with early adopters: Users will have the ability to opt into "early adopter" status, which will give them access to drafts, ARCs and other special content. For the readers, this is free content; for authors, it's audience-building, targeted at the readers most likely to support them.
None of this is revolutionary — I know there have been versions of this that have come and gone over the years — but once it's up and running, the mangers we've talked about will have the data they need to make informed business decisions...at which point they can concentrate on the second part of their role: packaging.
In a separate post, naturally.