Why are onboarding projects so painful & frustrating?

At Class35, we’ve now designed several onboarding experiences for financial services products across a range of services - from retail banking, to business banking, to merchant acquiring, to corporate banking. 

Almost without fail, onboarding is seen as one of the most complex, painful and frustrating projects to take on. Many organisations have multiple false starts, overrunning projects, and failed transitions from build to run - but why? 

Expectations are high 

Onboarding, by definition, happens at the start of a customer’s relationship with a new product, brand or service. Often directly following marketing or sales journeys, their expectations are at their highest. After consideration, they have chosen a service, either as a new customer, or more likely switching from someone else. So they expect, in most cases, for the onboarding experience as a new customer to be an extension of their journey as a prospective buyer. Marketing messages and brand propositions about speed, simplicity and ease can quickly be undermined by friction in the onboarding experience. 

In financial services, these expectations are being pushed even further by the experiences on offer from challengers. A few taps, a selfie, a scan of a passport, and some very quick automated checks and you’re away. 

It’s a risky business

In larger, more risk averse organisations (averse to fraud, reputational risk, or mis-selling due to a lack of suitability), due diligence burdens are high. This creates a natural tension between conversion and risk - low friction creates happy but risky customers, high friction creates frustrated but low risk customers. 

Many organisations fail to comprehend the relationship between risk policies and digital experiences. They quite often will design the digital experience (probably by benchmarking against challengers), in complete isolation from the risk policy. 

AML, KYC, KYB and Credit Risk policies are different in nature and what they protect against, but share a similar quality in that they are often designed around hurdles. If the digital onboarding experience does not provide the data points or inputs to assess against these hurdles, it is very hard to give a straight answer, which means it is very hard to provide a “straight-through” digital journey. 

There are ways around this - either staging the “burden” (an official term for the provision of identity collateral), and later ‘stepping up’ the diligence, once further down the journey. Or “faking it”, by giving the illusion of a digital onboarding experience, but in reality handing a form to a person to key manually. 

There is a counter argument to taking this approach. Some banks, including the challengers, really struggle to enforce follow-on collateral, when it’s needed. For example, when deposits go above a threshold, or average transaction values increase, or when applying for lending. A low-touch risk process at the point of onboarding can limit cross-selling opportunities. So whilst it is good for conversion, it might be bad for lifetime value longer-term, if not messaged properly. It is in this danger zone where things like account freezing happens, which is by far the biggest driver of negative sentiment. 

This gets even more complicated in business-banking focused use-cases, and in particular for larger corporates. A lot of financial products require the approval of ultimate beneficial owners - i.e. directors of the company in question with a material ownership. This is fine for small businesses, where the person completing onboarding is likely the same as (or in close proximity to) the beneficial owner, but in larger companies, it is much harder. My favourite example is a money services company, recently acquired by one of the FAANG, had the CEO (a famous billionaire) of the parent company as an ultimate beneficial owner, which meant their signature was required for certain banking product decisions. 

Our approach to this is always to design onboarding journeys from the most complex (multi-product, changing addresses, complex UBO structures) and work backwards. It is a lot easier to simplify than it is to bolt on complexity.

Happy paths are a minority 

People have been talking about ‘straight-through processing’ for a long time now. And the design equivalent of this is a “happy path”. When designing a new onboarding experience, there is a tendency to spend a lot of time on the journey where everything goes well - the right proof of identity, no complex ownership, no credit rating issues, no data entry issues, straight forward address history - to show how nice the journey could be. 

In reality, very few journeys will resemble the happy path. There are far more “maybes” than binary answers, and journeys are often recursive - e.g. a user might not have the right information to hand (e.g. a DUNS number is a big culprit here). Many onboarding journeys are designed to be sequential and funnel-based, which means these types of situation, where a customer is stuck between two stage-gates of the journey, can often go unnoticed. 

These ‘unhappy’ or ‘exception’ paths have the potential to be as significant, and even positive, as the ‘happy’ version. There is an art and skill to designing nice experiences around broken journeys, and it comes in very handy when thinking about onboarding. This approach does mean designing the ‘back-stage’ customer handling processes in parallel to the digital journey, because quite often, the interface plays quite a passive role in resolving these exception paths. 

By mapping the exception paths, alongside the happy path, and designing the customer handling processes alongside the digital solution, as well designing clear customer communications, not just interactions (e.g. email journeys), these journeys can be optimised to be flexible to suboptimal circumstances. 

Onboarding is where siloes converge

Mapping back-stage processes that potentially cut across the whole organisation can be tough. Onboarding is the point at which many of the siloes that exist within an organisation converge. Sales, marketing, client services, operations, fraud, risk, technology can all come together within one space, which makes mapping the ideal operating model and customer experience difficult. This is made even tougher by the fact that a lot of these stakeholders have a different outcome in mind - marketing care about brand, sales care about conversion, risk about risk, ops about scalability and replicability. This is why many onboarding projects end up as therapy sessions for the misalignment felt across an organisation. 

Landing on the right level of detail to facilitate these conversations can be tricky. Some stakeholders are more interested in the absolute detail, where others think more holistically. Demonstration, and getting things wrong, is the most effective way around this. We quite often will have challenging sessions - where we invite a likely critical stakeholder (e.g risk or regulatory) into a session to review designs and to suggest iterations or improvements. This is far more constructive than endlessly designing “principles” or iterating policy documents away from the customer experience.

So, really it is a digital thing, and a people thing 

For all of these reasons - risk, complexity, internal alignment, short and long term thinking - designing onboarding journeys is tough. Designing the happy ones that look like challengers is easy, but designing ones that balance customer, business, risk and technology is far from easy.  

The key to designing effective digital onboarding journeys is understanding people - the customers experiencing it, the colleagues supporting it, and the teams responsible for creating and maintaining the service. Whilst digital self-service can take away a lot of the strain on people, it still can’t exist without them - particularly for regulated products, particularly for large organisations. 

So what to do about it

We have a tried and tested methodology when it comes to designing onboarding experiences. The rough principles are: 

1. Start from the customer’s perspective. Standing in the shoes of customers and looking back upon the business is a great way to spot inconsistencies, siloes and opportunities. It can also paint a good picture on the ‘jobs’ of customers, and add emotional context (e.g. frustration). 

2. Design for complexity (no happy paths). Gaining an understanding of the gnarliest iteration of a journey is a good way of surfacing the complexity that needs to be managed through the experience. It can feel like the hardest battle to pick, but in our experience is the best route to value and feasibility. 

3. Use customer journeys and prototyping to force facilitation. Customer journeys, done right, can highlight all of the front and backstage touchpoints and processes that make up the experience.  Similarly, prototyping elements of the service (not necessarily digitally) that require discussion is a good way to align different stakeholders around the experience. 

4. Improve policies, processes and journeys in parallel Policies, processes and digital journeys are entirely co-dependent. Over-indexing on one aspect will create an unbalanced outcome. Designing balanced journeys requires compromise, which is only achievable through iteration and facilitation. 

5. Test the journey (in full) and iterate (whilst inviting challenging stakeholders to be part of the solution, not the problem). Testing and validating journeys, both operationally and with customers, regularly, is a very good way to ensure that journeys are achieving their intended outcomes. What makes sense to a designer or analyst starting at the same process for months might not make sense to a customer. 

We find this approach, and the right internal working groups, creates the right outcomes. An approach that is about balance - risk and conversion, tech and process, digital and tech. Which is ultimately what makes onboarding such a tricky topic to work through.

‍