A year ago, my colleague Paul Thompson wrote an article explaining that in many ways self-serve is the ultimate end game for those calling for an open (and measurable) programmatic landscape. High praise, and a lofty goal to live up to…

2018 in ad tech seems to be characterised by a desire to take control, whether it’s header-bidding, ads.txt or any number of initiatives mentioned in the myriad 2018 predictions pieces. Whether we’ve reached the halcyon days where all our programmatic problems will disappear is another question – but self-service solutions remain a key topic in the industry, and have had a wide-reaching effect on the industry.

So, did what we expect to happen, happen? Let’s take a moment to delve into the details.

Does the shoe fit?

It’s worth setting the record straight from the outset: self-serve models don’t suit everyone. Tech providers usually offer a range of solutions to cater for different requirements: data-only solutions, analytical insights requirements and fully-managed services, in addition to self-serve platforms.

Self-serve is designed so that customers can use the platform as much or a little as they want. Platforms and dashboards can essentially be customised around a client’s specific requirements; whether you’re a buyer-trader or a planner-strategist, a good self-serve platform should meet the specific data and touchpoint requirements that you are looking to harness.

Take manufacturing trainers as an analogy. Adidas’ services are now becoming more and more personalised, where consumers can have them custom made. With Adidas pop-up manufacturing shops in central London, you can get your perfect trainer made for you in two hours. It’s no different to what’s been happening in ad tech: customers want customisable, personalised options.

From the back seat to the steering wheel

In terms of logistics, the self-serve option requires customers to develop the know-how and specialist personnel to work the technology. The truth is, self-serve requires more of your own time and effort than a managed service. Whilst self-serve initially looks to be the cheaper option, there is still a cost associated with the time and human resources needed to get a handle on the tech. And as with the adoption of any new piece of software, it often takes a while to get to grips with. Marketing technology is no walk in the park; some tech providers haven’t yet been able to develop sufficiently client-facing self-serve platforms.

So, it depends very much on how you want to allocate your resources. This year I’ve noticed more and more very bright people, with expertise in data analytics, forming part of in-house programmatic teams.

Then there’s the question of transparency: Since Paul’s last piece, the conversation has continued raging around this topic, and this will continue into 2018. Self-serve is a way to secure ultimate transparency and gain unparalleled control, allowing users not just to look under the proverbial hood, but to tinker with the very engine. It’s an individual decision that comes down to whether you’re seeking more control, transparency and flexibility.

And in any case, self-serve comes in all different shapes and sizes. Brands may want different degrees of control over campaign management. For example, an in-house marketing team may only want to manage the buying side of things, relying on a managed service for analytics and reporting. There are a host of niche offerings out there: Spotify recently launched its self-serve audio ad platform, whilst Choozle has offered a self-serve tool for contextual keyword targeting.

It’s worth remembering that self-serve will always remain ‘managed’, to some extent. Your tech provider can offer additional support for campaign optimisation, such as fixing bugs and providing knowledge around tags and trackers. They should also look to offer technical support for creatives too, ensuring that creatives render properly, reworking the file size (if affecting load time) and ensuring the creatives have been fully sandbox-tested.

Reap the reward through trial and error

Another of last year’s concerns was the ability of tech users to adapt to these new solutions – Paul mentioned that there would be an aspect of ‘trial and error’ in implementing new strategies and data-sets. Those hands-on types that have gravitated towards self-serve have not only adapted to the tech but made it their own.

Rather than using someone else’s solution, there’s a clear symbiosis between providers of this kind of tech and the users. There is a constant feedback loop from customers, powering and accelerating the tech providers’ research and development. Development cycles of several weeks can ensure that the self-serve roadmap is regularly updated and is fully responsive to customers’ needs within a short space of time. This keeps the process of self-serve as seamless and frictionless as possible for the customer.


Whether it’s right for you or not, the advent of self-serve represents a major step forward for the industry. For those who are happy to make the investment, self-serve looks to be a timely response to brands’ concerns over transparency, control, flexibility and creativity. For the curious marketer, the question you should keep in mind is: what level of control do you want & what level of effort are you willing to put in?