Observations and Ideas


04 November 2015

Business that knowingly take advantage of the financially vulnerable make me sick. Increasing the access that pay-day lenders have to people in these situations through lead generators is sad and terrifying, but as the article People’s Deepest, Darkest Google Searches Are Being Used Against Them points out, Google is in a tough situation trying prevent this and uphold it’s own policies, let alone any hypothetical increased regulation. A terrible problem without an obvious solution.

02 November 2015

Interesting that Adslot’s engineering team has independently arrived at a very similar process and culture to Spotify. Albeit we’re a much smaller team but many of the ideas Henrik mentions in the video are similar to those we’ve considered in our own context. Maintaining loosely coupled, tightly aligned teams will no doubt become more challenging as team size increases.

29 September 2015

Maciej Cegłowski’s talk “What Happens Next Will Amaze You” makes a number of interesting observations around the evolution of ad tech and the genesis of robots/ad fraud, which is summed up nicely:

Today we live in a Blade Runner world, with ad robots posing as people, and Deckard-like figures trying to expose them by digging ever deeper into our browsers, implementing Voight-Kampff machines in Javascript to decide who is human. We’re the ones caught in the middle.

The ad networks’ name for this robotic deception is ‘ad fraud’ or ‘click fraud’. (Advertisers like to use moralizing language when their money starts to flow in the wrong direction. Tricking people into watching ads is good; being tricked into showing ads to automated traffic is evil.)

It’s worth reading through his argument as it unpacks many of the unpleasant incentives for ad fraud and the impact of the technologies that exist to track and mitigate ad fraud.

Further on he proposes 6 fixes, the most interesting being a ban on third-party ad tracking.

This sort of ban would trigger an incredible reset on the way ad technology works and immediately start moving ad tech is a much more consumer friendly direction. Indeed, third-party tracking is the raison d’être for iOS9 content blockers.

03 November 2014

Talk in the industry of late has been focused around programmatic pricing, transparency, and responsiveness. A recent report by The Economist1 suggests “60-80% of ad spending is siphoned off by ad-tech firms which take advantage of the market’s opacity” supporting their claim that programmatic buying is failing to deliver its promised transparency and efficiency. This is not a new claim.2 Opaque fees and trading desk margins, often charged by agency owned DSPs, continue to be a source of concern to publishers.3 Transparent reporting of fees and markups will go some way to earn the trust of publishers and advertisers. Unaddressed however is the broader efficiency goals of automated sales channels, particularly automated guaranteed.

Meaningful automation of advertising transactions requires making good on the promise of campaign life cycle efficiency improvements. More specifically, the tools of the future must help produce better outcomes for advertisers, and improve agency and publisher efficiency throughout planning, trading, optimising, and billing.4]

Best laid plans

Jared Belsky (President, 360i) recently called for a ‘Programmatic Reform’ which identified an urgent need to overcome opaque operating models, integration gaps, and provide ‘nimbleness and fluidity.’5 Responding to measured performance, external events and client feedback are cornerstones of effective campaign management, and necessitate a suite of integrated communication and planning tools.

Can automated guaranteed make good on the promise of efficiency if, during a campaign, planners and buyers are left scrambling for the phone, recording adjustments in disparate spreadsheets and leaving finance to pick up the pieces? Surely not. Efficiency should unite measurement and optimisation. Efficiency should bring order and confidence to billing processes. Efficiency should integrate communication and negotiation.

Making good

Significant in roads have been made on creating automated guaranteed marketplaces which assist buyers discover relevant products, availability and pricing with ease. Yet, it would be a sad thing to reduce automated guaranteed to only this. Let’s talk about how timely reporting, real-time impression forecasting, and ad server integrations (both 1st party and 3rd party) could become the foundation of a toolset which helps agencies and publishers work together over the life of a campaign. Clients aren’t paying for a perfect media schedule, they are paying for results.

Until tools begin to help agencies and publishers collaborate post-purchase, and throughout the optimisation process, automated guaranteed won’t be living up to its greatest promise.

21 May 2014

Meleesha Bardolia, writes for Right Now, on proposed changes to Australia’s Racial Discrimination Act:

Brandis’s packaging of these revisions as innocuous, at least, and progressive, at best, obfuscates the transference of power from victim to perpetrator in an act originally designed to ‘protect’ individuals and groups against discrimination. This is suggested by the limitations imposed on the scope of racial discrimination to “vilification” and “intimidation”. The new Act defines the term “vilify” as meaning to incite hatred against a person or a group of persons. This definition shifts the onus of the burden of proof from the victim to the perpetrator. Furthermore, the word “intimidate” is defined as causing fear of physical harm.

Meleesha outlines four key changes which ‘legalise the dehumanisation of groups and individuals based on their racial extraction.’

It is definitely worth reading in it’s entirety: The Banality of Evil: Proposed Amendments to the Racial Discrimination Act (Right Now, 16 May 2014)

23 April 2014

Melanie Phillips on the intellectual sloppiness of the arch-apostle of reason, Richard Dawkins:

The way he chose to defend himself, through insults and sneers which tried to cover his tracks as he attempted to retreat from what he had said, furthermore merely emphasised his notable reluctance to address the many arguments of substance against his pseudo-scientific attack on religion which were made by John Lennox on the grounds of scientific reason and accuracy – arguments which Dawkins most tellingly chose to ignore altogether. Instead, he went for what he thought were the soft targets – a credulous Irish Christian and a ‘dreadful woman’ journalist – and substituted smears and jeers for proper debate.

It is worth reading the article in whole, which unfortunately can’t be accessed directly at The Spectator anymore. [HTML] [PDF]

It is good and important to passionately discuss the merits of religion and its compatibility with science and reason, but to sink so low as to lean on gratuitous insult is neither productive nor healthy.

29 November 2013

Maryn McKenna writes Imagining the Post-Antibiotics Future:

I imagine what he might have thought  […]  if he had known that a few years later, his life could have been saved in hours. I think he would have marveled at antibiotics, and longed for them, and found our disrespect of them an enormous waste. As I do.

It worries me that more people aren’t aware of the harm caused to society by not correctly adhering to prescriptions. Maryn McKenna’s article clearly explains the importance of taking the issue seriously.

23 August 2013

Enrique Allen writes in the interview Learn by Doing for Kern and Burn:

I’ve seen too many talented people in Silicon Valley work on products that are incremental or not that life changing for anyone. By using the term ‘designer,’ we have a responsibility, even a moral obligation, to intentionally impact people’s lives, hopefully for the better. Unfortunately, the products we design often waste people’s valuable resources and their attention, which is more scarce than money and time. Why is it that there are so many talented people working on shallow problems?

22 July 2013

A perl from Jeff Atwood:

To build something truly reusable, you must convince three different audiences to use it thoroughly first.

A clever and concise rule of thumb for software products.

08 September 2012

Don’t misconstrue delight and novelty

Consider the moment you first turned a page in Apple iBooks and the novelty of the curling page. If you are anything like me, less than 20 pages later you’d just like to read the next page.

As Craig Mod explains in his talk Nourishing Habits for Nourishing Designs designing for “long term wow“ is hard and delight should not be confused with novelty. Delight—the long term wow—rewards and nourishes users time after time and helps construct a positive relationship with users.

Novelty, in careful and limited use, provides a wonderful treat for the user. The first time they launch your application or visit your webpage you can engage and excite them with a flourish. However, after only a short time novelty fades to frustration.

Designing an interface that delights is hard. It likely won’t be noticed the first time the application is used. If well executed, however, the user is rewarded as they return more frequently to the application. The application’s predictability and simplicity make it an experience that is good. It delights.

Frank Chimero’s Design Nobility Pyramid suggests the highest calling is to first delight users, followed by informing, and lastly persuading.

Superficially this runs counter to most commercial aims. The priority would likely be to first and foremost persuade customers to participate in a revenue generating activity.

However, this assumes that all visitors are equal—new and returning—and that users are inflexible and/or have few alternative service providers. Now more than ever, we should know this to be false where new technology companies gain prominence in timespans measured in weeks and fade to obscurity in days. In light of this developing a relationship, delighting and informing those who use that which we design and create must be our greatest of goals. How else do we create something lasting? How else do we create something of real value?

Attempting to design a product of lasting value and enjoyment is something I wrestle with daily. Both the technical challenge of designing features that delight users, but more frequently the challenge of scheduling and the cost of development. It is easy to discuss the merit of design nobility and delight in isolation from the cold reality of product management. Far harder is balancing a delightful and feature-rich delivery schedule in a complex and revenue hungry technology start up.

I suppose awareness and the best of intentions are the first step to finding a balance between the two.