The Map Workshop - in marketing terms
A cp Vanguard case study
A cp Vanguard case study
This is a follow up to to an other topic. Start there if you've never done anything on the workshop.
Unfortunately the workshop doesn't give us a lot of information so we're kind of left in the dark when it comes to measuring how well we're doing. The few stats we have about map submissions are not all that useful. It's usually because they are metrics of quantity, not quality. I used to work at a place where we would do automation for the management of Google Adwords advertisement campaigns and if I've learned one thing, it's that marketing people LOVE their metrics. They have soooo many of them, it can be kind of overwhelming. You need a ton of different data if you want to measure the effectiveness of a marketing campaign. Here I'll try to apply this knowledge to the workshop and make do with what Valve gives us. If they don't give us the data, we'll have to calculate it ourselves damnit. let's see what we can get out of the few bones that valves throws us.
The good news is Cp Vanguard had an interesting run. I've experimented a lot with its workshop submission and collected data that I want to share to help us figuring out what's effective and what isn't. We're going to crunch down the numbers and see what we can learn from it.
Where do visitors come from?
When you're advertising, one of the important things to keep track of is where your traffic comes from so you know where to focus your efforts. With the workshop, we don't have have a clue. A lot of it comes directly from people browsing the workshop but sometimes, people will land on your workshop page from elsewhere, measuring this "elsewhere" is important to get a better understanding of what works, what doesn't.
Vanguard has a bunch of activity spikes spread out over time, let's see what those are about.
Initial release: Cp Vanguard had a pretty good opening week on the workshop, the initial peak is pretty typical of what you'll find on most finished maps.
Pl Dogbread release: Posting an other map not too long after shows a slight bump that is attributable to people visiting my workshop page for more maps.
Koth Vanguard release: Now this is where it gets more interesting, Koth Vanguard was advertising Cp Vanguard in its description via this clickable image. This brought in a noticeable amount of people to the CP page
UGC announces Vanguard as part of their rotation: the Addition of Vanguard brought in a couple view, not a big amount but definitely noticeable. UGC didn't link the workshop page directly, this was from people searching through the workshop or through google.
Koth Vizard: Posting the Halloween version of vanguard and advertising the regular version with this clickable image has once again gotten us a sizable spike in activity.
Pl Effigy: posting pl effigy, without any vanguard advertisement brought some views to the table much like pl dogbread.
It's Map Pack Time: IMPT was my map pack that I organized to bundle some of my maps, headlining cp Vanguard. This was a bit of a misstep since the web page did not link directly to the workshop page, this could have been more effective in bringing in views.
Tough Break: It finally happened, The map was added to the game and for some odd reasons, people are still looking at it on the workshop? weird, I know.
hopefully this gives you an idea of where your visitors may come from and what's effective in driving traffic towards you and what isn't
I'm getting traffic, but is my workshop page doing it for people?
This is where conversion rate comes in. It's your most useful metric in measuring the effectiveness of your workshop page. The conversion rate is marketing lingo for the ratio of visitor on your webpage that are convinced into paying for a product or service. In this case, we are not selling anything, our workshop page exists only to collect ratings, subscriptions and favorites. You can base your conversion rate on any single one of those value or aggregate them to get some sort of average. Any positive action the visitor does is a conversion to me so I'm just going to add up everything, (votes+subs+favs)÷views. That would put Vanguard at a conversion rate of 41%, that's okay. We now know that upon landing on the workshop page, roughly 2 person out of 5 had a reaction that was positive enough to convince them to act on it. In that regard, the workshop page did okay. The expectations the thumbnail had set up have been met and the end result convincing enough for them to push buttons, that's great! (there was definitely room for improvement tho).
You'll want to monitor your conversion rate early on so you can act on it. You're only going to get a finite amount of views on your page so it's important that you make them count. If your conversion rate is low, something about your workshop page is not grabbing visitors when they get there. Time to rethink your screenshots and description. If you start to get that feeling that your map just isn't ready to be making upvotes rain, It might be better to simply pull it down as fast as possible and repost when the map is more polished and/or more exciting because when someone clicks on a thumbnail, they are likely not going to be returning so you're burning potential conversions by leaving the submission there. Reposts are discouraged unless you can do it early, you only have one shot.
Now of course conversion rate should be taken with a grain of salt, it's not the end all be all. If we look at koth_product, we have a high conversion rate of 68% percent. Does that mean that the workshop page was exceptionally good? no, not at all. In fact it's pretty average and its thumbnail is nothing special that would bring in lots of attention. The view count is pretty low in fact and that should tip us off. koth_product being a staple competitive map, what likely happened for most visitors is they knew exactly what the map was before looking at the workshop page. They went in and already had the intention to hit buttons, the workshop page didn't do anything special that I should try to replicate in the future. Data is important, context is key.
But what about my thumbnail, is it performing well?
My main pet peeve with the workshop is that there is no way to measure the effectiveness of your thumbnail despite it being crucial to the success of a map on the workshop. We have a visitor count, but that is quantitative data, not qualitative. A qualitative metric that we could really use having around would be the click-through rate, meaning, of all the times your thumbnail appeared on a potential visitor's screen (an "impression", in the lingo), what's the percentage of times people were interested enough to click it.
To better understand, here's a place where we can effectively calculate click-through rate on the workshop. Suppose you're advertising an other map in your workshop page's description, we know how many views both maps are getting so we can measure how effective this ad is. In the case of cp_vanguard, it was advertised in other submissions on two occasions (koth_vanguard and koth_vizard). If we take koth_vanguard, it peaked at 2700 views per day on sept. 1st, this generated a peak of 133 views for cp_vanguard on that day. 133 ÷ 2700 = 0.04, meaning 4% of people who saw the ad in koth_vanguard went ahead and checked out cp_vanguard. If we assume the click-through rate is constant over time, this would mean that over koth_vanguard's 12,000 total views, it generated 480 total for cp_vanguard. Since we also know the conversion rate of cp_vanguard, we know that this generated around 192 conversions for cp_vanguard.
So yeah, This article turned out super long but I hope it was useful. Maybe a lot of this is common sense but I feel like it's not exactly intuitive or obvious so I hope this helps people make sense of the data that's being given to them and that my experiments hold valuable lessons to you. have fun on the workshop!
Note: view counts on the workshop are all based on new logged in visitors and not actual traffic, also there's a difference between total favorites and current favorites. same for total subscribers and current subscribers. This will throw your numbers off, but it doesn't really matter which ones you use because it doesn't stop you from making useful conclusions, just a thing to keep in mind.
