Lessons from MTPEngage Manchester 2019

(N.B. These are notes from my thoughts while people were talking as much as they are notes about what was actually said; I hope I’ve not misrepresented anyone though, and please call me out if I have.)

Nilan Peiris from Transferwise talked about Customer Led Growth, he mentors his teams to constantly remember the problem (fall in love with the problem, not the solution, sounds simple…) and to make big value bets with confidence rather than optimising incrementally (it seems nobody at Transferwise is playing with button colours).

Tom Loosemore’s challenge was around digital transformation in organisations, or “changing the weather” as he put it. Change is challenging, and those pushing for it should remember that this challenge can be emotional, for themselves and for others in the organisation. Tom’s big money slide, much tweeted, was this:

“If you’re not feasting you might be fired, you’re probably not being bold enough.”

I read this as a message that if you are explicitly choosing or charged to create change in your organisation you are going to ruffle feathers and cause a bit of a stir, to say the least. Tom addressed concerns over the privilege “if you’re not feeling like you’re about to be fired” that this slide has evinced for some:

For me, I found this message of feel the fear and do the right thing anyway inspiring.

“Obligatory Sun Tzu quote” 😆

A lot of my efforts at the moment are bent to understand and practice product strategy, so I was looking forward Nicholas Goubert’s talk on the subject. From Nicholas I took away that I shouldn’t be afraid of strategy, that “an imperfect strategy is better than none”, and that, as with almost everything, iteration was the key.

Product Strategy is cross-functional, so we need to work with other stakeholders to ensure that everyone is aligned and in sync. Product Strategy needs to be understood by everyone, so reiterate, reiterate, reiterate at every opportunity. Overcommunicate the strategy to promote ownership amongst the team. In his interview on Maggie Crowley’s Build podcast, John Cutler talked about being ready to “walk your belief map”, and this strategy of reiterating the strategy reminded me of that: “we’re doing X because we believe it will promote Y, which will get us closer to our organisation vision because Z”.

Nicholas talked a living strategy which was challenged weekly, and which you are not afraid to update, radically if required, every quarter. Vision, on the other hand, is an immutable constant. Happily share your vision externally, but keep your strategy internal and share only with stakeholders (although that may include customers). The Product Roadmaps Relaunched book talks about your roadmap being a prototype of your strategy, which Nicholas’ advice echoes.

Randy Silver reminded me us of this timeless quote:

The day we stop being curious about our customers is the day our competitors start to catch up.

Theresa Torres

As well as a simple but important differentiation: don’t talk to your customers, listen to your customers. Aim to pick one tough, meaty question, a question which will get at the “why”, ask it, and then keep listening as your customers tell you their stories (from personal experience, this is harder to do than to say).

Randy also gave us a good Product Management description for when your relatives ask what you do:

I help people make better decisions faster.

Randy Silver (although he credits Monika Turska)

Roisi Proven talked about closing the Lizard Loop, listening to your the System 1 (fast, instinctive, and emotional) thinking and backing it up with System 2 (slower, more deliberative, more logical) thinking. She recommended taking a baseline to start, considering what biases and anchors you are carrying, for example:

  • Do I think unfairly about a group?
  • Am I blindly or overly positive about my work?
  • How do I think and feel when users “get it wrong”?

Roisi recommends checking in with your lizard (System 1) and your owl (System 2) before coming to a conclusion. Feel the knot in your stomach, or the urge to optimism, and back them up with data before proceeding.

Roisi also reminded us that the world isn’t just nice people, and that when developing products we should always consider:

What’s the worst that could happen? …or, what if this, but baddies?

Roisi Proven

…which reminds me of:

We took lessons in negotiation from Garry Prior, and at risk of condensing the 20 minute talk that he gave which condensed a three day training course:

  1. Preparation is crucial: always have objectives from the meeting, consider your wish list, think through the trade-offs you’ll accept… and above all consider the other parties’ points of view
  2. Be comfortable arguing: there will come a point where you need to test the assumptions put forwards by all parties, ask questions, clarify, and listen to others
  3. Negotiation must move towards proposal and acceptance: a good format is “if you do that, I will do this”, open realistically and move modestly

Lucie Mclean considered the similarities between a hedge fund manager and a Product Manager. Lucie talked about considering the spread of risk in your product portfolio:

  • Low risk work: optimisation, very specific hypotheses, lots of available data… but not an optimal strategy in and of itself
  • Medium risk work: originating from internal and external signals, “where shall we take this product next?”, no guarantees of success, worth doing an MVP to address the risks
  • High risk work: Huge hypotheses, “what would we do if?”, big changes and big opportunities, watch out for assumptions and “shiny things lust”

Especially with the high risk strategy, think how much you are betting and consider this as your stake… you might lose it, can you afford to?

5.6 – Make your decisions as expected value calculations.
5.6 a. – Raising the probability of being right is valuable no matter what your probability of being right already is.
5.6 b. – Knowing when not to bet is as important as knowing what bets are probably worth making.
5.6 c. – The best choices are the ones that have more pros than cons, not those that don’t have any cons at all.

– from Principles by Ray Dalio

In managing your organisation’s portfolio of risk, you can ask “what is our risk appetite?” As well as “what is our expectation for gains?” I’ve found myself referring to these thoughts recently, as I talk through our product portfolio with colleagues at my organisation.

David Wascha closed out the sessions for the day, with an invitation to Look to the Left. When looking for a new product at Travelex, he and his team asked the following questions of different activities consumers conducted: “how important is this to you?” and “how happy are you with it at the moment?” Traditionally Travelex had concentrated on a series of consumer actions which went:

  1. Get (money)
  2. Go (somewhere)
  3. Spend (money)
  4. Return (and sell spare currency)

When they talked with their customers they found a step 0, a step to the left of the sequence they were looking at:

0. Plan (how much you need)

The goal with the “look left” technique is to extend your engagement with customers. For example at Moonpig, the team are looking left to prompt and remind customers of upcoming occasions they need to send cards for… but also looking right, to close the loop “would you like to send a free thank you card to the sender of this card?” (this answers the perennial card-senders question “did they get my card”, and possibly the more important question “did they get my card and do they think I’m great for sending it?


Looking back at my notes I got an enormous amount out of MTPEngage Manchester, and I’m so grateful to be living in a city with such an active and engaging product scene.

Call me out for any misunderstandings or odd assumptions in my notes above in the comments, or tell me about other product conferences I must go to.


Dear Reader: Are you in or near Manchester? Do you like coffee? Do you like conversations about product stuff? Check out the Manchester Product Coffee meetup, first date is Wednesday, February 27, 2019 at 9:30 AM in the Spinningfields Benugo. It would be lovely to see you there.

Photo credit for the featured photo at the top: Mind The Product, thank you for the permission to use it.

Example composer.json for SVN repos with no structure and no composer.json

Today I needed to include some WordPress.com themes from the official SVN repository, using Composer. This repository uses SVN and has not got the usual trunk, branches, etc directories that Composer expects, which makes it tricky. It’s possible though, and thanks to colleague Tom Nowell, I eventually got there.

Here’s that example code: Continue reading Example composer.json for SVN repos with no structure and no composer.json

Podcast – How Google uses behavioral science to make work suck less

Podcast: How Google uses behavioral science to make work suck less

This episode of the You Are Not So Smart podcast came recommended by Donncha, a colleague at Automattic. It’s packed full of really interesting thoughts, including “the job of a manager is to serve their team”; a good reminder that everything we at WordPress.com VIP do in terms of process does serve our team and our goals.

Laszlo Bock (head of People Operations at Google) explains how and why Google does what it does when it comes to everything internal, from perks and promoting to motivation and productivity. In the interview you’ll hear how the company combats confirmation bias, the halo effect, the Abilene paradox, pluralistic ignorance, survivorship bias, and more – all with a mix of behavioral science and Google’s immense power to test and re-test using its unique resources.

If Tom Willmot’s WordCamp Europe talk was out on video, I’d link to that, so keep your eyes open there. There’s also Nikolay’s tweet recommending Drive, which is now on my reading list:

…of course, if your work sucks, you could change that by coming to work for Automattic ;)

Displaying MySQL query results vertically

Ever found the MySQL table view annoying? The table format is terrible at displaying rows with long values, isn’t it? You can get MySQL to display results vertically:

Some query results are much more readable when displayed vertically, instead of in the usual horizontal table format. Queries can be displayed vertically by terminating the query with \G instead of a semicolon.

More information on this tips page.

mysql> select * from wp_usermeta where user_id in ( select ID from wp_users where user_login = 'ab04d41a-f5b3-4859-9474-5a7047c7e23c' )\G
*************************** 1. row ***************************
  umeta_id: 325
   user_id: 24
  meta_key: nickname
meta_value: Mr G Teacher
*************************** 2. row ***************************
  umeta_id: 326
   user_id: 24
  meta_key: rich_editing
meta_value: true
*************************** 3. row ***************************
  umeta_id: 327
   user_id: 24
  meta_key: comment_shortcuts
meta_value: false
*************************** 4. row ***************************

Protecting staging sites with Basic Authentication

This approach was primarily devised for use with WP Engine staging sites, but will work in any situation where you have no access to Apache’s config and need to use a single .htaccess file for multiple domains (e.g. a multisite setup).

We’ve recently had a couple of projects hosted on WPEngine where we need to have a staging site protected by Basic Authentication so only authorised users can access it. Normally I would configure Basic Authentication in the Apache VirtualHost, but on WP Engine (for example) they handle all the server configuration for us, meaning we don’t have access to the Apache configuration. WPEngine also provide a Git push to deploy facility, which we tend to take advantage of because Version Control Rools, OK? Because all the web files are within Git version control, we are using the same .htaccess file for the developer, staging and live sites. (We could Git ignore the .htacess file, but I prefer to version control more, rather than less.)

Continue reading Protecting staging sites with Basic Authentication