When we measure retention rate (or churn rate) of customer cohorts at TouchNote, it’s important for us to ensure any detected differences are statistically significant, and that these differences have sufficient statistical power. (For a quick explanation why you have to validate for statistical power as well as significant, read here).
Luckily there’s an easy web tool built by Evan Miller that lets you just that very easily. Here’s how to test for power and significance in a jiffy.
First, you want to test that the sample size (i.e. the size of each cohort) is large enough. …
TLDR: List of books, hacks, tools and general things I think everyone at a consumer startup will find useful to know.
There are a bunch of things that I feel that everyone at TouchNote should know, or at least be familiar with. None of them are rocket science — they’re are mostly frameworks, ways of thinking and theories that I make regular use of. Whether you fully subscribe to their views or not, you should at least be familiar with them.
TLDR: A few weeks ago, I decided to re-immerse in weekend trading of cryptocurrencies, Bitcoin and altcoins. I share my thoughts about how fun it is; varying degrees of FOMO; and some extra reading you should do if you’re considering investing.
I signed up to Coinbase back in 2015, but was not very active. I restarted trading Bitcoin after a Blockchain Bitcoin meetup at our WeWork office. To learn about Bitcoin, you need to trade it. Like many things in life, there’s a limit to you can learn about something from just ‘reading’ about it; you must do it. If you’re into software, you have to learn coding to really get it; you can’t really understand data if you don’t dabble in maths, statistics, SQL, R or Python. So, trading cryptocurrencies was the quickest way to learn. …
Yesterday, I spent a good three hours trying to purchase the latest hyped cryptocurrency: IOTA. Here’s what I was surprised to find out.
In 2011, I switched from PC to Mac. I used a MacBook Air 15" for about a year until it got stolen. I then rushed back to a PC. It was for a simple reason: I couldn’t get work done as easily on a Mac. It’s interesting how little marketing Microsoft has done to point out MacOS’s shortcomings. There are at least a dozen reasons why moving to a Mac would drive you nuts. Each of these could have been a great retaliatory “PC vs. Mac” TV advert. …
Say you are launching a new website. You create a launch webpage (aka sign-up, beta or pre-registation webpage) for your new domain, e.g. via www.launchrock.com. Now, how do you redirect traffic to that launch page, without redirecting all sub-folders, too? In our case we wanted people who visit www.timeforequity.co or timeforequity.co to get redirected to our Launchrock.com page. But, we wanted timeforequity.co/blog to still point to our blog. Sounds simple, but it took me ages to find a relatively simple solution for our domain. Here’s how I finally did it.
Redirect from -> to:
www.mydomain.com -> host.launchrock.com
mydomain.com -> host.launchrock.com
mydomain.com/blog -> www.mydomain.com/blog …
Twitter makes it very hard to unfollow all your Twitter followers. 3rd party services had to withdraw this feature because they got letters from Twitter “asking” them to remove that option. “Asking” in quotes, because otherwise Twitter would have likely removed their permissions which would shut their business.
So here is a workaround for unfollowing all your Twitter followers quickly, in the shortest number of clicks. If you find an easier way, please tell me in the comments.
- If you don’t know your way around Twitter, installing programs, or using a PC, please follow these steps carefully, or not at all.
- These steps require registering to Tweepi’s Silver plan, which costs $7.95 for one month. It’s well worth it, in my opinion.
- These instructions are useful for PC users only. If you have a Mac, you can probably afford paying someone to do this for you. (I’m just kidding. …
I just wanted to say hi. My name is Freud. I’m now 10 weeks old and live in London with my keepers, the Rans (one of them writes this blog, I think).
I like to fetch things, jump and hop, wee a bit outside the house and a bit inside the house. I’m told that I’m really cute, pretty and smart, but that I’m not the bravest of them all. That’s fine with me.
I’m a Nova Scotia Duck Tolling Retriever but if you hold me upside down I look a bit like a fox.
Here are a couple of videos of me jumping around. I hope you enjoy them. …
There’s one thing about Dropbox that drives me nuts. The date format. If you use their website, all files are listed using the US format of mm-dd-yyyy (e.g. 06–16–2012). According to this, only Americans and Belizans use the month-day-year format. Most of the world, over 3 billion people, use day-month-year and over 1.5 billion (mainly in China, Japan and Korea) use year-month-day. For anyone else, seeing a date like 10–01–2012 is hugely confusing (it’s October 1st rather than January 10th).
Quite a few blog posts, bug reports, and change requests call for Dropbox to let customers choose the right date format. Over 500 people even votes for it. The strangest bit? only 32% of Dropbox users are American, so the majority of Dropbox customers would actually benefit from this change. So how strange that this is still deprioritised. …
I came across this fantastic paper after reading about Meehl’s 1954 book, “Clinical vs. Statistical Prediction: A Theoretical Analysis and a Review of the Evidence”. It’s a meta-analysis (review of studies) of the evidence that mechanical prediction techniques like fixed questionnaires or algorithm generate superior predictions compared with those made by clinicians, regardless of their level of expertise.
In plain words, it suggests that simple questionnaires and algorithms arrive in more accurate predictions compared with clinicians, doctors or psychologists. I can’t believe this is not mandatory reading in every university.
The process of making judgments and decisions requires a method for combining data. To compare the accuracy of clinical and mechanical (formal, statistical) data-combination techniques, we performed a meta-analysis on studies of human health and behavior. On average, mechanical-prediction techniques were about 10% more accurate than clinical predictions. Depending on the specific analysis, mechanical prediction substantially outperformed clinical prediction in 33%-47% of studies examined. Although clinical predictions were often as accurate as mechanical predictions, in only a few studies (6%-16%) were they substantially more accurate. Superiority for mechanical-prediction techniques was consistent, regardless of the judgment task, type of judges, judges’ amounts of experience, or the types of data being combined. Clinical predictions performed relatively less well when predictors included clinical interview data. These data indicate that mechanical predictions of human behaviors are equal or superior to clinical prediction methods for a wide range of circumstances. …