Seeing zero user engagement after signup. How do I find out why?

Anonymous

December 28th, 2014

Dear Founderdating community,

I am building a community-based site for learning English targeting advanced learners at http://morpheem.com . The site offers a new text every day where you can look up words, accumulate vocabulary and discuss with other users. It's still rather rudimentary as it only launched two weeks ago.

My idea was to try to proceed in an agile fashion based on early user feedback, but my current problem is basically that there is very little of that. I’m primarily getting users through advertising. Users then hit a landing describing the service somewhat vaguely, have to register using social login and then wind up on the text of the day.

At this point I notice the large majority of users IMMEDIATELY leaving the site without doing a single click on anything (I have extensive click tracking on the site). This surprises me since I thought the real barrier would be to get signups, but that is not the case.

It’s also highly frustrating because a complete lack of engagement makes it impossible for me to use user behavior to drive further development. I thought it might be because of a lack of guidance / structure so I added a small tutorial with which, however, users are not engaging at all (most are not even closing it, just leaving it open).

I'm getting decent click-through on my ads and about 15% of visitors (which makes 10-20 per day at the current, extremely low, ad budget) then sign up on the site, so that far the funnel is working.

I’ve tried e-mailing 25 users displaying this behavior with a very short e-mail in simple English asking them to give feedback by simply clicking on a number of pre-defined reasons why they did like the site (confusing / too hard / not interesting…), but only two of them even opened the mail and only one clicked on a link (“confusing”).

I have two questions:

1) Does this behavior seem normal or could it be stemming from targeting English learners, meaning that maybe I’m just vastly overestimating their level of English or reaching the wrong people? (On the other hand: they read the landing page in English and sign up)

2) Any ideas on how to break out of it and get some user feedback? Should I abandon the whole idea of using anonymous user behavior to guide development and get a real-life focus group of users instead?

Thanks for any help!

Roger Smith

December 29th, 2014

I just signed up for your site and although I don't have a solution to your problem here is some feedback:

1) are the users who are signing up for your site the same ones that would actually use it? Meaning when I sign up, the home page makes it sound great, but since the stories are in English they don't help me a whole lot. What is the language of your signups?

2) Maybe the idea/concept sounds great, but in practice people don't want to learn English in this manner?

3) If #2 is not true and they do want to learn English in this way, maybe the stories don't capture their attention?

Tom Maiaroto Full Stack Consultant

December 29th, 2014

Ads are going to work like that. I've seen it a few times. Ads can bring you a lot of traffic, but most of it isn't really quality traffic let alone converting users. So yes, that is normal.

I'm not an expert in the case of your product and the language barrier certainly could be an issue...But I'd start with multiple landing pages that included different language and different copy.

Try your ads again specifically for these landing pages and measure the difference. You don't need fancy services like Optimizely, etc. In fact those can be a nightmare if you have a team of developers (or just any single contract developer) who will one day wonder how the heck something got on the page while they can't see it in a file somewhere =) ...Though you could use a service if you find them helpful.

Anyway, just make a few simple landing pages and track with Google Analytics. Simple as that. Measure.

It should make a difference if you try enough variations. It should answer your questions about whether or not the language on the page matters, etc.

Also try social media.

As for #2: Don't abandon your anonymous data. That's super important still. You should do user studies though, but do it live under your control. Never use an online service because they come with a lot of bias. User testing is a great idea and a great tool, but trying to turn it into a SaaS is an epic fail. It simply does not translate well.

What ends up happening is you get people who act like they are dumb (and they probably are) and exaggerate. Overly so.

Bring people in if you can or get them to share their screen and record everything. Offer gift certificates and free service, etc.

Who knows, you might walk away with a good relationship with an actual user and get them to engage their friends because they had such a wonderful experience and really liked you. Bonus.

Robert Clegg

December 29th, 2014

Offer free content first. Use those first moments for engagement. Funnel by content not advertisement. Let people experience the product first, then sign up.

Then you can start to see where and why people are not engaging with your content. In the game world it's called "free to play". Get people to start playing/using your content first, then get them to upgrade.

Here, just put off the sign up phase. Then you'll start to have data on your learning audience.

Balki Kodarapu Experienced Software Engineering Leader

December 29th, 2014

I agree with Tom and Robert that just because someone signed up, you can't assume they are in your target market.

That said, two things will help re-engage users:

1) Optimize the experience fully for mobile

2) Continue to send emails/push notifications/texts (assuming most of the users are accepting to receive them @ signup) - even if it is one every week day.

#2 is often overlooked since we as entrepreneurs assume that engagement is automatic like Facebook or Twitter :) But it's not. So we have to continue to remind them of our tool until they get tired of your nagging and unsubscribe or you offer enough value that they re-engage. Both are good outcomes IMHO.

Steven Schkolne Computer Scientist on a Mission

December 31st, 2014

Hi Andreas - seems like you are doing everything right, by the playbook - with measurement, iteration, etc. However, ime all that is moot without an actually catchy experience.

Some have commented on the UX - I agree with those. The intro value proposition was not clear to me (for a great book on this kind of messaging, try http://www.amazon.com/Pull-Marketing-Secrets-Fortune-100/dp/0976861771/ref=asap_bc?ie=UTF8 ). I didn't log in so, not much more I can say here.

What I'd like to highlight though is something that was only slightly mentioned above (by Monica Borell) - and that's the importance of sitting in front of users and watching them use the system.

I recommend the following:

* get a group of 10 people to visit you. select from your demographic (either post a targeted ad on craigslist, or go to a local ESL school/etc to find some volunteers)

* pay these people some small amount ($10-15), get them to sign some paperwork, etc - plenty of resources on this kind of thing online

* don't say anything to them verbally about your product. be disciplined about not biasing (in fact, if you can pretend to be a software testing professional, unassociated with this particular product, you will further remove bias).

* have them sit in front of your login screen, and watch how they interact with the product. DO NOT INTERFERE IN ANY WAY. sit behind the subject. if possible, have a second observer with you in the room as well.

* pay attention, learn.

* perhaps have a survey at the end, but your observations are the most valuable. (people can't often express why they do or don't take certain actions.... often giving errant explanations... but their reactions don't lie).

Through this, you will be able to see the subjective, human catchiness that your experience is lacking. Whatever it is that you need, it is beyond metrics and more about the intrinsic value of the product. You may need a good creative person, who can operate in this intuitive space to create this magic. The stats and analysis can amplify the magic but ime it can't create it (with stats we are essentially hill climbing, and sometimes a leap of intuition is needed).

Hopefully this process shows you tons of obvious things that you never anticipated. And, from there, you can iterate. When the magic starts to hum, you will see it plainly.

best of luck!

Shingai Samudzi

December 29th, 2014

I visited your site. In a nutshell, the main purpose of the site is unclear. You say here that your target is people learning English, but this element is not communicated clearly. The zero engagement is indicative of your users being unclear how any of your product is supposed to help solve any pain points or problems they have. Honestly, without that framing your product comes across like English homework without much reward. What the user is left with is "here are links with a bunch of text in english" with a social element tacked on.

Anonymous

December 29th, 2014

Thank you so much everyone for your feedback! It has been very inspiring. While I do not always agree in detail with all your analyses, it's extremely helpful in prompting me to revisit a number of assumptions. I have a number of takeaways:

- I’ll try targeting a super-simple text at new signups and see what effect that has on their engagement.

- I’ll conversely try another version of the landing page, more clearly geared at advanced learners. I’m pretty sure that will decrease signups; but the relevant measure might be time spent on the site rather than signups, and I’ll analyze it for that.

- Same thing goes for the up-front signup (that I originally did not have, but signup rates soared when it was put in place): I’ll take it down again and this time try optimizing for session length or return visits and see if the conclusion is the same.

- I’ll go through my statistics of the number of users clicking “I don’t have Google or Facebook” again and double-check my previous calculations that the number of users I lose this way is irrelevant at this stage.

It’s very inspiring to have the support of a community behind you! Thanks again.

[If anyone from Founderdating reads this: you really have to scale back your aggressive marketing; I will now unlink my account from LinkedIn even if that means I can’t log in again. The original post found its way to my LinkedIn, though I’m pretty sure I uncheck that option every time (and it shouldn’t default to on in the first place) - admissions that my business is not going well is the last thing I want on LinkedIn; it’s hugely embarrassing]

Hao Zeng

December 29th, 2014

Hi Andreas, I remembered back a few years ago when I was in college freshman year and I just came to US from China. One history lesson I was taking have 10 books that I had to read and I would love to have a service like yours, and It would save me so much time looking for words in the dictionary one by one. My suggestion is to see if you can find the right group of users in college, preferably users from other countries. Best of luck, Hao

Ian Maddox

December 29th, 2014

Your tagline, "Better at English, one text a day" is the first thing users see. The syntax isn't correct, so it discredits the site. For folks not bothered by that, they may conclude their English is too good/bad for this product to be effective.

Also, reconsider your ad targeting. You may be hitting the wrong interest groups and demographics.

Anonymous

December 29th, 2014

Thanks a lot for all the quick feedback! Great to hear from other people in the same situation.

Roger: I'm targeting people learning English who are already at an upper intermediate level; it's not for English-speakers learning any other language, though it could be extended to this later. Users are to a large part in Russia, Ukraine and India, though there is a big spread.

Roger and Ian: while there are all sorts of valid doubts about the basic concept of the site, I am in fact seeing really high signup rates on the landing page, so I don't see any reason to change the advertising or the landing page at the moment; they work as intended. The problem is that once people have signed up, they immediately abandon the site without even exploring it (they for example don't click on a single popup, which I would expect to be the first thing to do to see what functionality they offer).

Tom: A/B testing is tricky in this case, since it's not the ads I have a problem with, it's what people do once they are signed up. If the problem is a major mismatch in the message of the landing page and the functionality (or look and feel) of the site, I could of course build several different kinds of messages and see whether there is a difference in engagement; but since it's so extremely low at the moment, there will be a lot of noise. I also doubt that is the main problem.

Testing with a focus group would indeed be a way forwards and I agree 100% with you that it's not something to be outsourced; we'd need to do it in-house.

Thanks everyone for the insights so far!