The Numbers Game

31 October 2016 | 29 Comments

It’s exceptionally rare that I give away a product for free as part of a contest or raffle.

I’ve discussed this subject in detail here and here. I’ve seen very little proof that contests and raffles generate sales. While the mission of Stonemaier Games isn’t about revenue, but sales need to be at least a byproduct of every decision we make. Otherwise it’s not a sustainable business.

A few days ago, one of our advisory board members (Craig), shared with me something he saw on Facebook and LinkedIn:


Pretty impressive stats, right? But…likes and comments do not equal sales. Did anyone buy this book as the result of participating in the contest?

The answer: Honestly, I have no idea. I’m sure the contest exposed a lot of people to the book. In my opinion, though, it’s missing a thematic hook to make people want to learn more about the book. Nor is there anything about the image that grabs you and makes you want it now. 

But I liked the general idea, so Craig and I brainstormed some revised versions of it. Coincidentally, I had a photo on file from a recent resorting of extra realistic resource tokens, so with Top Shelf Gamer launching pre-orders for a new batch of the tokens, the timing seemed perfect.

Here’s how I posted it to our Facebook page, with a few key improvements over the contest that inspired it (the “update” part wasn’t on the original post):


Can you spot the differences between this contest and the original? Here are a few of them:

  • It includes a link for people to learn more about the tokens.
  • It features descriptive text on the image itself.
  • For some gamers, the image itself may make them want mounds of resources of their own. Perhaps you can feel the iron running through your fingers as you look at this photo.
  • The contest guidelines are very specific (comment on this post, you can’t edit your comment, one guess per person, etc).
  • There are an uncountable number of tokens in the photo (opposed to the number of books in the Wrong Place Wrong Time photo). I thought this would ensure at least a day would pass before someone made the right guess, allowing the post to ramp up within Facebook’s visibility algorithm.

How did it work out? The post was seen by 17,809 people. 94 people liked it, and there were 163 comments.

However, within about 30 minutes, someone guessed correctly. I didn’t think it would be ethical to let people keep guessing after that, so I informed the person that she had guessed correctly, and I updated the post. Activity on the post almost immediately stopped.

It would be nice if I had sales data for you to see if the contest worked, but I actually sent out our e-newsletter (which featured the pre-order) the next day, so any data I might have is corrupted.

But I still had fun with the contest, and hopefully it inspired at least a few people to check out Top Shelf Gamer to learn more.

For creators thinking about running a contest like this, make sure you don’t do it on Kickstarter itself–they have ambiguous rules about contests.

What do you think about this type of contest and the way we ran it? What would you have done differently? Have you seen any contests that do a better job at generating sales?

See also: Kickstarter Experiments

Leave a Comment

29 Comments on “The Numbers Game

  1. Good idea. Two tweaks might have made it more successful. Let people guess the weight of all the tokens. What makes these tokens feel special in comparison to wooden tokens is not the sheer number but their weight. So it makes people how nice the heavy tokens feel in their hands. Furthermore, it is much more difficult to guess the weight exactly right ( to the gram).

    Set a time frame like 2 days. At the end of the 2 days the person whose guess is closest to the answer wins. So even the very last comment can still win the contest.

  2. A little bit about me: I make contest/sweepstakes/… (whatever you want to call them) for a living (for my business partners). Mostly with apps (to get emails and other data about participants + it’s easier to create a unique campaign and include more content (and maybe even links to webshop)). But I also have experience with competitions like yours. You made some minor mistakes:
    – text is too long (why exclude and even mention edits?, link could be shortened)
    – contest should end at the specified time (with random selection of a winner; not with the first correct guess)

    But i have to congratulate you for wanting to create a meaningful contest.

    Next time you could use reactions (like, heart, haha, wow) for voting (“which tokens do you desire?” or “favorite faction”) + comments.

    1. Thanks Ales! I recently mentioned the heart/like approach for voting on a different blog entry, and I look forward to using it when it applies.

      I mentioned the edits because without it, someone could post a number and then change it as much as they wanted without me being able to track if they had actually changed the number. Isn’t it better to eliminate loopholes than to allow them?

      1. With a fixed duration of a competition there would be no problem with edits.

        And because I like board games… I can create an app for your next competition. For free. I can email you some of our work and ideas for you.

  3. Really appreciate all these posts on Contests, have been thinking about running one on BGG, but was not really thinking about Facebook, I do not know why I missed that. I know the importance of adds for exposure, but I really like the idea of running contests and give aways that last a few days, as an alternative to just the generic facebook add.

    Would an alternative to the “guess” how many could be Like or share, than all of the people who shared or liked with in X days will be added to a draw for the prize. That way you would have the ability to extend the contest for certain period of time, no risk of having an early guess.

    Just a thought. again thanks for such a timely discussion, I personally think contests are great.


    1. Johnathan: Thanks for sharing your idea about liking or sharing. Perhaps that could work, though I tend to only want people to like or share something if they genuinely like it or genuinely want to share it, not because they might win something.

  4. I guess one of the problems with this sort of marketing is that is can feel artificial and possibly desperate. A company that has to rely on marketing gimmicks is a company that probably doesn’t have a very good product. I think the quality of Stonemaier’s products speak for themselves, and the publicity that comes organically from people talking about your games is a better quality of publicity and is more likely to reach the type of new customers who would actually be likely to make a purchase.

    The problem with using social media as a generic platform is that everyone else is doing the same thing, and so most social media posts are essentially spam. Using social media in a targeted way–whether it be BGG forums, reddit, or your blog–is more likely to get people’s attention because those people were already looking for information on games.

    I don’t see the average facebook user as someone who would understand what iron ingot resources were or have a use for them. If you narrow that down to the average board gamer however, you might be batting 25% or so.

    Out of curiosity, have you had an increase in sales of Viticulture and Euphoria since releasing Scythe?

    1. Kenneth: Yeah, that’s a good point about gimmicks versus genuine interest/relationships.

      As for Viticulture and Euphoria, it’s hard to tell, as we coincidentally sold out of Viticulture and Euphoria at the same time that the first print run of Scythe was released.

  5. Maybe instead of first person to guess, let it run for three or four days, and draw randomly from all correct guesses (One guess per person, and no confirming the right answer)? That way people keep coming, keep participating, etc.

    1. I like that idea as well. Given the potential exposure of letting it run for a few days, you could even offer a prize all correct answers (limit 50) with a grand prize randomly selected. This could be a good option for the farm animal tokens, given their more universal appeal beyond board gamers.

  6. I figure these kinds of contests are less about driving sales and more about exposure. You want your loyal fans to be engaged with the contest, and even if you don’t force them to “like and share” and simply add a comment, then I assume the FB algorithm will float this to the top of people’s feeds.

    In the end though, only you (Jamey) can work out if this was worthwhile. How much did it cost you to run the contest? Wholesale cost of the prize + shipping + few hours of your time versus the exposure and potential new likes on the page.

    1. Stuart: I agree that exposure and engagement are great, but what’s the point if it doesn’t drive sales? A band can tweet constantly to 20 million Twitter followers, but if those followers don’t buy their album, the band won’t be able to make more music.

      For this contest, it cost $100, plus about an hour to set up and run.

  7. I have very rarely participated in any contests – I don’t normally “like/share/etc” contest posts.

    I did participate in this one, though.

    It was probably because I follow and like SM games and products, though.

    To have attracted new customers I’d have probably used a different resource. (And I was a little surprised that it wasn’t – I mean…I like the iron ingots just fine, but their real appeal is found in holding them. They aren’t the showiest resource in the sets.)

    Full disclosure – I did open the photo in an app and tried to put a dot on each countable item, and then approximate what was under. My results were abysmal, and a hilarious example of my complete inability to estimate!

  8. I guess if I were going to try to analyse this I’d do the following:

    Treat likes from people who’ve not previously liked your page or posts as special, this builds engagement more than sales (in the hopes that the new people will eventually get a game or component) so the real question is: Did you engage enough new people to justify the giveaway?

    If you’ve got an algorithm (or someone super patient) to work that out you can then do something like

    Newlikes on this post – Newlikes on average post = Newlikes for giveaway
    Newlikes on boosted post – Newlikes on average post = Newlikes for boosting

    Newlikes for giveaway / Cost of giveaway = Giveaway cost for Newlike
    Newlikes for boosting / Cost of boosting = Boost cost per Newlike

    From there you’ve got a ratio between the cost effectiveness of boosting posts vs giveaways (and from there to all advertising you do assuming you’ve got some way of determining if it’s better to boost posts, get banner ads, invest in promos etc.)

    I dunno if that level of formal analysis is necessary, but it seems like a way to figure it out.

    Depending on your ethos it could also be considered worth doing even if it’s not the most efficient way if it’s still a net gain and meant some people who otherwise wouldn’t encounter them get to enjoy your games.

    Thirty minutes is quick for it to be over, but it’s not very surprising. If you’ve got 163 comments than the question isn’t “Can someone guess this” but “Can a group collectively guess this to within 163” A bigger pile would help or a tougher question (Here’s a mix of Iron, Coal and Water – how many of each?)

    1. “Thirty minutes is quick for it to be over, but it’s not very surprising. If you’ve got 163 comments than the question isn’t “Can someone guess this” but “Can a group collectively guess this to within 163” A bigger pile would help or a tougher question (Here’s a mix of Iron, Coal and Water – how many of each?)”

      Another way would be taking out the group element. As everyone can see each guess, you can follow along and disregard previous guesses. If each guess was made by private message, you remove the Meta. Just like those jelly bean jars, you don’t write your guess for everyone to see, you write it on a piece of paper and put it in a jar.

      By having guesses in PM, your contest would last longer and be exposed to more people.

      Another thought, I may be wrong, but I believe that the only people who would see this post are people that have already “liked” your page. If that is true, a facebook post is unlikely to generate new customers. A guessing jar at a convention, however, where guessers need to put in a valid email address or subscribe to your page/newsletter is much more likely to generate new customers.

      1. Part of the point is for the post to get shown to more people though. A private message isn’t going to be seen in the same way as a comment by facebook’s algorithms.

        There’s something to be said for a competition that encourages people to read each other’s comments – if there are also comments from people who liked the components and were talking about the thing itself as much as guessing.

        1. Good point greg…

          Perhaps keep the guesses Private, but prompt them to share with their friends or family. If they can’t get the prize, maybe a friend can.

          As in, “Hey _____, there’s a contest for free gaming tokens (link). I guessed 212, you should try guessing something else. Good Luck!

          1. Yeah, that’s the tricky thing about the contest. Public guesses are good, as they increase the exposure of the post to people who already like the page. But they don’t increase the visibility to people who haven’t already liked the page. Perhaps that’s when I’d need to pay Facebook to sponsor the content.

    2. Greg: That’s an interesting formula! I really should have tracked new likes on the page, shouldn’t I? It didn’t occur to me at the time. I generally don’t try to do anything to pursue inorganic likes, as I only want people to like the page if they truly want to follow it.

      1. “This page has competitions for things that I want enough to enter a competition” isn’t so far from “This page shows development of things I’m interested enough to follow” that I’d be certain that anyone who liked on the basis of that isn’t the sort of person you’re interested in. It’s not like paying one of those “100 likes for $1” companies or something like that.

  9. I think that it also depends on what you are giving away. 100 resources and shipping of them to the winner, or as a regular contest is quite a cost. Versus something like a promotional card for one of your games.

    I am planning on running contests for Dice designs for my next project, and random people will win their choice of the punch out dice, but at the same time get feedback on what types of designs people like to unlock for stretch goals. It’s something small, I think it will help build community, and the investment on my part is small. Hopefully it works out.

  10. I am not sure this contest will really increase your sale directly for this product, but I think (without any data to support it) that running regular contest like this will encourage more people to follow your facebook page (to not miss the next contest) and so they will be more aware of your actuality and you may touch a broader audience when releasing a new game for example. In addition, it improves your image as a nice editor who offer free stuff to people.
    Will all of this increase your sales enough to balance these giveaway product (and the time involved)? That’s a tough question to answer without more data.

    1. Roman: Thanks for sharing your thoughts. I have a very different perspective on that, though. I agree that if I ran regular contests, my Facebook page might gain more followers. However, those followers aren’t there to learn more about our products and potentially buy stuff–they’re there to potentially win free stuff. They’re motivated by the enticing prospect of free instead of the prospect of information. That’s the exact opposite of what I want.

      But I’m glad you brought this up, as it’s an important point: If I ran future contests, they definitely would not be on any regular basis. Rather, they would be exceedingly rare.

  11. Hey Jamey, interesting post. Not sure my personal stance on contests, but looking at it in a more scientific way would be interesting.

    Just wanted to clear something up, did you mean unambiguous instead of ambiguous. They are pretty clear about their stance on contests. I don’t think there is room for double meaning.

© 2020 Stonemaier Games