Mick Niepoth

Coding as a Product Manager

2017-12-26T00:00:00+00:00

Coding as a Product Manager

Do you need to know how to code as a product manager? I’ve seen PM’s who couldn’t code at all, but also some that even picked up occasional Jira tickets. The amount of coding a Product Manager gets to do greatly depends on the team, but in practice I doubt you’ll be doing more than some tinkering around on Sunday nights. I believe this is with good reason—it’s simply not the most effective thing a PM can do most of the time. However, I do want to make a case for (some) coding as a PM. Not to solve bugs, but to make yourself significantly more effective and better informed.

Example of a KPI Dashboard I developed.

A large part of a PMs tedious tasks consists of pulling data from different sources, making certain calculations and then analyzing the endresult. Some frequent tasks could include;

Pull new deal-data from Salesforce and link this to Hubspot prospect lists (for funnel analysis)
Get a number of KPI’s from customers (Like daily active users)
Results from recent A/B tests, conversion rates, etc.

This is important work. However, it can be automated. This will mean you’ll have more time to bring value in things that actually matter, like data analysis.

All of this is combined in a Google Sheet, using charts and conditional coloring this ends up as a fully featured Dashboard that’s easily sharable with stakeholders. If you’d like you can even send this automatically every X days.

Can’t code?

No worries, there’s a lot you can do with services like Zapier. You can easily hook it up to almost all popular servic

What else? I’d be interested to hear if coding benefits your work as a Product Manager. Drop me an email or connect on Twitter.

Tighter Job Market, Machine Learning & Hiring

2017-09-25T00:00:00+00:00

The American unemployment rate is at a 9-year low. There are similar trends in most Northern European countries. In a lot of industries the job seekers are now calling the shots. This will also likely lead to an increase in wages in the next period, since employers have to pay more to attract and retain workers.

Disclaimer: I currently work on a Machine Learning project for a recruitment agency. The opinions expressed here represent my own and not those of my employer.

(Bloomberg.com. 2017. America’s Job Market Gets Tighter, as Wage Drop Seen as Blip - Bloomberg. [ONLINE] Available at: https://www.bloomberg.com/news/articles/2016-12-02/america-s-labor-market-gets-tighter-with-wage-drop-seen-as-blip. [Accessed 18 June 2017].)

For companies, this means a lot of new problems: tougher salary negotiation, fighting off competitors, higher quit rates, etc. In addition, it has to remain competitive in the hiring process as well. How can a company improve this? Machine learning could be part of the answer.

Applying to jobs is a tedious and exhausting process. Unbelievably clunky ‘career’ pages, companies that respond only after weeks of waiting, if they respond at all. I’ve seen lots of instances where companies receive thousands of job applications a month. Often this is all sent to a decade-old applicant tracking system (ATS) in which each and every application is processed manually. This quickly leads to rising costs for the company and longer waiting times for the applicant. Companies simply can’t afford to continue this anymore. Nowadays waiting too long to respond means you run the risk of losing talented candidates to a competitor. By developing sturdy models based on carefully chosen feature sets you can use historical patterns to shift through these applications more easily.

The optimization already starts before the actual application comes in. Lots of companies have a career page or make use of job boards. These are often loosely organized pages. Based on search queries from potential applications, the site should highlight the most relevant openings (low hanging fruit: proximity to the job, seniority level, etc.) and (ideally) give the applicant a good sense of where in the process they are; what are the steps and how long until they can expect a reply. A smarter site should reduce friction for the potential candidate, so more relevant applications will come in.

Then, when the application reaches the database, it’s important that someone looks at it as soon as possible. The Machine Learning model should be able to determine the strength of a certain candidate’s application. This will bring guidance to the recruiter working through this. Is someone unexceptionally qualified? This should trigger an alert requesting a fast(er) response. For candidates labeled as highly uncertain, more relevant vacancies could be suggested, linking them to better opportunities. You shouldn’t make the mistake of only using models to reduce the number of applications you spent time on. Potentially more value lies in identifying talent that a manual process would’ve missed otherwise.

As with everything in Machine Learning, the whole thing depends on the availability of data. Even the largest companies in the world will not be able to utilize these ideas if they don’t track data properly. Some ideas on what you could track and include in your models;

Data collected during exit interviews; why did someone leave? Where did they leave to? Also, try to combine this with data on this person’s previous performance and feedback.
Data on hires; make sure you record where a candidate is coming from, how many interviews that person had, and the feedback during the screening process
Tests; Did the candidate complete any tests and training? What insights did this provide you?

In my opinion, Machine Learning shouldn’t be used to solely replace recruiters or downsize your HR department. It should be seen as a valuable tool that will help your company select better colleagues. It could free up time though, which can be spent doing things that software can’t, like coaching. In addition, it’s important to be aware of bias creeping into your models, especially when you just start out these models need a lot of attention.

What else? I’d be interested to hear how your or your company is utilizing Machine Learning. Drop me an email or connect on Twitter.

YouTube Picture-in-Picture on iPad

2017-08-24T00:00:00+00:00

Since iOS 9 it’s possible to watch videos in the so-called Picture-in-Picture mode on your iPad. Now you can watch your videos, while doing other work. Unfortunately it doesn’t work on YouTube’s website. However, with this trick you can enable PiP-mode anyway. If you bookmark this link (or use this Gist), you can click on it while watching a video to enable PiP.

Have fun!

YoungCapital creates breakthrough in machine learning in recruitment

2017-07-11T00:00:00+00:00

YoungCapital has succeeded in making the recruitment process faster and more efficient through machine learning. An extreme achievement, because machine learning in recruitment is extremely complex.No other Dutch employment agency has been able to produce such an innovation. YoungCapital expects to provide more revolutions in the short term and, therefore, in this area, structural cooperation has begun with Leiden Institute of Advanced Computer Science (LIACS).

Source: YoungCapital

Optimize match between candidate and employer

Machine learning plays an important role in optimizing YoungCapital’s services. “If the computer takes over the most time-consuming work of our recruiters, they can spend more time on career counseling and coaching. And that’s just the beginning,” says Rogier Thewessen, one of the founders of YoungCapital. “Because in 2018, our candidates will be shown how much chance they’ll likely have for a particular job. We make suggestions for training that they can follow to increase their chances of shooting for that job. And we show alternative vacancies that make them a great opportunity. Our dream: less rejection, less disappointment and more happy candidates who get started in a job that really fits them. Our algorithm, YoungCapital Brain, will help us. “

Machine learning in recruitment: Recognizing patterns without pattern

Machine learning in recruitment is another free area. Owing to the space for interpretation, context and subjectivity, it is a very complex niche. Machine learning is about recognizing patterns. However, in the recruitment there are no patterns at first sight.

Fig. 1. Pretty simple issue (consumer buying / not buying in a store) versus complex issue (candidates selected / not selected in recruitment)

Srisai Sivakumar, data science and machine learning specialist at YoungCapital, explains: “Recruitment data is subjective. Think of the vacancy text, the CV and the motivation. Everybody writes in a different way, uses other words and emphasizes other things. YoungCapital Brain has to read all the same. However, a computer can not interpret as a human can, but reads words as numbers. Nuance and context we need to learn the computer. That is a huge challenge. Nevertheless, we have managed to achieve concrete results in the Netherlands with excellent accuracy. “

Inhouse solution

Machine learning is an absolute necessity in recruitment, Thewessen believes. YoungCapital’s database now has 4.8 million candidates. YoungCapital has been addicted to data since its establishment, but without a good search, it is possible to search data for a pin in a haystack in such a mountain. Algorithms can refresh this search. “First, we used a standard tool,” says Thewessen. “But we would not be YoungCapital if we did not think it should be better, smarter and faster. That is why we have developed our own self-learning framework with algorithm. YoungCapital Brain learns to recognize the behavior of our recruits and candidates and makes predictions based on it. In addition, machine learning also supports us in other business areas.”

Together with LIACS

The fact that YoungCapital takes major steps in this machine-learning niche, also professor Thomas Bäck, professor of Natural Computing, affiliated with Leiden Institute of Advanced Computer Science (LIACS): “I’m impressed with the robust results YoungCapital itself has Knowing with machine learning. YoungCapital has since become a dumb player in this field. Therefore, we are happy to work with them together. “For the Leiden scientists, the enormous mountain data of YoungCapital offers great opportunities for scientific research. The youth specialists’ data specialists, in turn, are pleased to have found a sound record. The LIACS does specialized research into parts of machine learning that are also of great importance to YoungCapital. Both parties are looking forward with interest and optimism to the future.

First result: selection time halved

Before YoungCapital Brain, every recruiter had to look at all the vacancies in order to select the best candidate. That costs a lot of time. YoungCapital Brain is now taking part of this process. It predicts based on data from the vacancy, CV, motivation letter and profile data which candidate is the best for a vacancy, and arranges them in order of suitability. The model has been so accurate that the best candidate in 99 percent of the cases is in the top half of the reactions. That’s why the recruiter is half the time. With almost 95 percent confidence, the best candidate in the top is 20 percent. That’s why the recruiter can take care of 80 percent of the time.

Fig. 2. Accuracy YoungCapital Brain (top 50% and top 20%) deposited in time

About YoungCapital

YoungCapital believes in the power of young people. As digital natives, they hunt innovation at organizations. The recruitment specialist sees young people as the growth capital that every business needs. YoungCapital is committed to bringing it best to young people and companies. So they continue to stimulate each other to grow.

Essential reading list for product managers

2017-01-07T00:00:00+00:00

Product management is a very multidisciplinary job. I summed up some books that helped me in different aspects of the job.

High Output Management by Andy Grove

This book walks you trough every basic element of managing a company or team. Finally a management book that wasn’t written by management gurus, but a great practical tips about meetings, motivation and processes. One of my favorite books.

Getting Things Done by David Allen

For me, using GTD has been critical for being productive. There’s also a great summary that captures the gist of the method well; gtdfh.branchable.com

The Lean Startup by Eric Ries

Although (over)hyped, the book does a good job of advocating the value of quickly validating an idea before spending tons of time on it. If you already do this, I’d skip the book.

Inspired: How To Create Products Customers Love by Marty Cagan

A very straightforward book that offers some helpful frameworks on working with customer’s feedback.

The Innovator’s Dilemma by Clayton Christensen

A classic. If you haven’t read this yet, you should.

The Design of Everyday Things by Don Norman

I’m personally fascinated by the design of seemingly mundane objects.

The Checklist Manifesto by Atul Gawande

How much can you write about checklists? It turns out; at lot. At the heart this book is about making simple processes and systems, that get out of the way and don’t create extra complexity.

Don’t Make Me Think (revisited) by Steve Krug

A easily digestible —but very good— book to web usability.

Cheap Marketing Ideas

2016-09-18T00:00:00+00:00

Here are some ideas for running cheap marketing campaigns. We have used these ideas to get to product/market fit and later to test new product ideas and iterations.

Direct Emails

Inspired by the great book Predictable Revenue we used personalized emails to targeted lists. The emails were very short and to the point, all optimized to get the prospect on the phone. With the first 5-10 calls to test and optimize our value proposition, and later to actually sell. Our emails usually went something like this: Hi, My name is Mick and I work for Readmore. I have an idea that will increase your readership size. Do you have 10 minutes tomorrow?

Search Ads

The fastest way to get users to your product is by buying ads. With increased competition on a lot of keywords, prices have gone up in the past years. By now it’s probably obvious advice, but try to find some long tail keywords. They’re cheaper and more effective. It takes a lot of tweaking to get this right, but for us it was always the fastest way to market. Longer term, you probably don’t want to solely depend on this.

Live Chat

For this you do already need some visitors to your website. If you do, this is a great way to get into the minds of prospective buyers. During launches of new features or campaigns I would always keep the live chat window in the corner of my screen, to see how users went trough the site and talk to them if I noticed they were confused. It quickly helps you identify bottlenecks, and it is a low-threshold way for your visitors to ask questions.

Mini Event

For us this only succeeded when we were right out of college, but there are other ways to make it work. Ask your old university (or other organization) if you can host an event there. They usually let you do it for free, and you can probably use their coffee machine. Find some speakers that you think are interesting, and invite people in your target groups. This is how we found our earliest customers. Nowadays there are also a lot of co-working spaces, they’re often eager to host events. That might work too, especially if you’re targeting other startups.

Press and speaking engagements

When I was in college, news outlets would agree to interviews just because I was a young entrepreneur. This is what one of my earliest mentors called the ‘cuddle factor’. Later on, it became more important to talk about something of substance, besides the fact that you’re an entrepreneur. If you find an organization that would be interested in your company, you can ask to present there. Make sure that you’re not just pitching your company, but talk about your personal story.

Content Marketing

It took us years to see real results from our content marketing efforts, but it was eventually worth it. As with everything; try to think what your audience might be interested in, you probably have a lot of unique insights—don’t just pitch your product.

Convert PDF spreads into seperate pages

2016-03-21T00:00:00+00:00

Here is a quick Python script that can split PDF’s with spread/double pages, into seperate pages.

Mutt — Old school email

2016-02-12T00:00:00+00:00

If you receive a lot of email and are very comfortable in the terminal, you might want to check out Mutt. I’ve been tweaking my settings over the past years to fit my taste. I wanted to share it, in case it’s usefull to someone.