[data] CTAs work

metrics

(Bas van Leeuwen) #1

Hey all, I sent @HAWK a message recently, showing a new feature we’ve made at Community Analytics and she asked me to share it here :slight_smile:

We’ve recently started playing around with a tool that can measure which posts “convert” the most members from lurker to participant (number is in the leftmost column below).

What I really like is that his validates a couple of assumptions that we make as CMs:

  • Asking people to introduce themselves works; it is generally the most engaging posts in all the communities that we’ve analysed1
  • Putting posts in CTAs to new members work. @hawk has a list of posts-linked-in-CTAs and it is basically the same list :slight_smile:

More research to be done, but overall I was pleased with the results here.

1. important caveat here: we haven’t examined if this leads to one-time participation only; most of the times a first reply leads to a second, but I don’t know if this generalises to these posts


It's not you, it's me! What can we do about it?
(Richard Millington) #2

This is interesting. One takeaway I had is that VERY few posts actually convert people :slight_smile:

I wanted to take a shot about how this data could be actionable.

The ‘introduce yourself’ thread is a bit of a problematic example. It’s one the post we drive everyone towards in the onboarding. So it feels like there’s a bit of a bias if we claim ‘it works’, it simply gets the most traffic. The two SPRINT posts I think were nudges from the mailing lists so fall under the same category.

Which leaves 3 really interesting topics:

  1. Facebook groups as a community hub.
  2. Where in the world are you.
  3. Zimbra or Socious.

That’s 2 platform related posts and 1 generic post.

My hunch would be that these were posts which were easy for people to participate in. Everyone can feel expert enough to share where they are in the world, most people have an opinion or experience with Facebook groups too.

So I’d read this as the easiest way to convert long-time lurkers into regulars at this point in our community might be:

  1. Get them to share what platform they’re working on and if they like it.
  2. Get them to share something generic about themselves (like where in the world they are).

Not sure if this is right, but that would be my takeaway from here.


(Bas van Leeuwen) #3

There is most definitely a bias in the selection, that said, when you look at e.g. the CMXhub data; you’ll see that the top 40 of the last year are 40 of the famous “welcome” topics. Those 40 generated 460 first timer reactions (that’s roughly a quarter of total participants in the community). At CMX only something like 30% of participants make a single post; so getting them in (leap of logic here, sorry) will lead to higher participation overall.

I think you’re right in the analysis of the other posts and I think it ties in with the ‘experts’ problem that you’ve identified before :slight_smile:


(Richard Millington) #4

I remember @Patrick publishing a really interesting post a year or two ago on this, I think I agreed with some of it, though perhaps not the entire conclusion. I might try to dig it up because the data was quite interesting. I think @Andy_Steggles at HigherLogic has some data that runs against my thoughts here too.

So I’m not sure if I’m right but my feeling is that it’s not the introduction post that converts people. You could equally argue that if so many people introduce themselves and so few become long-term participants, that the introduction post isn’t converting people and should be scrapped.

My point is more that any post you guide people to make their first contribution will be the post that converts the most people. The science of this would be to figure out what is the BEST post that leads to long-term participants AND (as per my last post) what are the best type of post to draw current lurkers out of the woodwork.

Does that make sense?


(Alessio Fattorini) #5

Both are important and Bas has identified the latter :clap: and we’re are curious about what is the first type.


(Andy Steggles) #6

Hey Everyone, just my two cents here as I’m quite passionate about having an “introduce yourself thread” as part of the onboarding process etc. As part of our last benchmarking report, we were able to measure that while only 7% of subscribers are contributors within a given year, if you can get someone to contribute, then there is actually a 48% chance they will contribute again in that same timeframe.

While it was fairly easy to quantify the above, I’ve found it more difficult to quantify the “why”. My personal believe (although i have no data to back this up) is that there is a “fear factor” to posting a message to a relatively unknown group of professional (our communities are all B2B) peers. Put it simply, I think many people are worried about looking silly for whatever reason. The beauty of the “introduce yourself” thread is that it’s relatively hard to mess up and the community manager usually has their back etc. So if you can help get them over this ledge and post their first message, my believe is that they will be much more inclined to post a “real” message when the time comes. Combine this with a warm response to their first post and some proactivity by the community manager and it’s not really surprising that they do ultimately “convert”.

Just my two cents worth :slight_smile:


(Sarah Hawk) #7

No doubt about it. I asked people here if that was the case and it was a resounding yes. “I’m not experienced enough” or “I don’t have the time to put into crafting a perfect post”.


(Bas van Leeuwen) #8

Yes, that makes a lot of sense. Problem is that usually you don’t have enough data to do a proper analysis unfortunately. I’d love to run some experiments if you have some ideas :slight_smile: You’d need a massive community to properly split-test two different CTA-threads for example. It could be interesting though!
Maybe @Jessica_Crouch is interested in an experiment like this? Or @julesstanden?


(Sarah Hawk) #9

Or possibly @ForumSentinel


(ForumSentinel) #10

We definitely have a lot of data haha. I’m all ears.


(Jessica Crouch) #11

@Bas_van_Leeuwen Hadn’t spotted this feature, nice! Looking at our parenting forum, our top threads are predictably the competition threads, followed by birth clubs, which are essentially ‘introduce yourself’ threads that then evolve into a long long single thread. You could maybe try this with our gardening forum, now that it’s responsive, which I expect will help bring in more new members.


(Bas van Leeuwen) #12

@Jessica_Crouch: I shared it with Helen :slight_smile:
The picture at the gardening forum is a bit muddy, we’re still looking into ways we might be able to tweak it, but that’s why I didn’t contact Daniel yet.

An interesting suggestion by @Carrie_Jones was to see if certain types of post (e.g. with video, or with certain tags etc.) perform structurally better than others. So we’ll have to see how that works out; so much work, so little time :smiley: