Background
Whatever the level of maturity of data and analytics (D&A) within your organisation and however distributed or centralised its practitioners, it is a role that must continually demonstrate its value to the business. In the early stages of their adoption, such as in Level 1 (Aware) and Level 2 (Repeatable) organisations, the focus is often on technical competency and skills. These are important in creating a stable base for D&A, but they will not ensure it is able to advance and gain recognition as an indispensible and value-driving function.
To move to Level 3 (Defined), Level 4 (Managed) or even Level 5 (Optimised), a cultural change needs to take place across the organisation. Within the CARBON™ assessment, this is identified in the question, “How does analytics support senior management decision making?”. The most advanced level sees the majority of senior management decisions being made on a stable base of trusted analytics with an ongoing feedback loop.
This is by no means easy to achieve, but nor is this type of challenge limited to D&A – any function (internal or external) that involves a degree of consultancy needs to progress beyond simple delivery of transactional obligations towards complex problem-solving within a service context. This has been recognised at least since 2000 when David H Maister co-authored “The Trusted Advisor”. It sets out a simple progression which advisors and clients should seek to make (see Figure 1).
The authors set out three basic skills that a Trusted Advisor needs to possess:
- Earning trust
- Building relationships
- Giving advice effectively
If data and analytics practitioners are able to develop these skills and follow this progression, they will likely achieve core objectives for themselves and their function, such as seeing more of their recommendations being adopted and deployed, becoming more closely aligned and integrated with lines of business (LOB), being involved in more complex and advanced problem-solving, and experiencing fewer challenges and less pushback from internal clients. All of those support the maturing of D&A and improve its sustainability.
The qualities of a trusted advisor
Since the original publication of “The Trusted Advisor”, the concept has been adopted and expanded upon greatly by management consultancies and leadership coaches. It is now possible to find lists of the qualities needed that literally range between five and 32 characteristics. While these tend to reflect the specific proposition of each author, they can be aggregated into a number of constants:
- Technical competency – any advisor needs to have credibility that is rooted in genuine skills which can be clearly demonstrated and assessed. This may include professional qualifications, knowledge of specific and relevant techniques, or hard skills.
- Domain knowledge – in order to deliver advice that is practical and can be applied, advisors need to understand the context in which the client operates. That may included knowledge of a specific business sector, competitor landscape, regulatory constraints or industry trends.
- Client-first orientation – what distinguishes a trusted advisor from a simple provider is the willingness to understand an issue from the client’s point of view, even when there is no obvious benefit to the advisor in doing so. This objectivity and service-orientation reduces concerns about making a long-term commitment or being tied to specific solutions.
- Leadership – typically, a trusted advisor will take the first step, whether that is tabling issues early on or being willing to propose more sophisticated solutions. Leaders are willing to tell the truth and take responsibility for their proposals, rather than offering “drop-and-run” fixes.
- Vision – the best and most trusted advisors are able to see the big picture and can communicate it clearly and in a compelling way. Trusted advisors earn their status because they bring this umbrella view into the decision-making process and can make a convincing case for it.
The qualities that earn data and analytics trusted status
In discussion with DataIQ Leaders members, it is clear that being consider a trusted advisor is desirable for practitioners and seen as the ultimate recognition of what D&A can bring to an organisation. It also aligns extremely well with what the best practitioners already deliver, has a good fit with the cultural transformation that they aim to produce, and is also seen as a validation of their personal commitment.
Aligning the D&A practice with trusted advisor behaviours can be seen to be relatively straightforward:
Technical competency – the data office may start out with data quality goals, mature through broader governance and even become a “Ministry of Truth” whose authority guarantees the credibility of any data source. At each stage, it is demonstrating the necessary skills to gain confidence from internal clients. For analysts, this is slightly more complex as their output is harder to explain and can appear “black box” to non-technical clients. One method that has been adopted within a global retailer’s insight and anaytics function is to run peer reviews on any project before it is delivered. This both double-checks its robustness and also builds the ability of the individual or team to communicate how it was created.
Domain knowledge – one member believes that D&A teams need to live by the saying, “If you want to go fast, go alone. If you want to go far, go together.” Through a combination of actively listening, building better briefs and stakeholder management, practitioners need to get close to the LOBs they support in order to understand their needs, concerns and even internal politics. Domain knowledge takes time to build – a data scientist will not arrive with it, for example – but can be built-in to the rules of engagement for the function with its internal clients.
Client-first orientation – getting something right for a client increases their perception of the practitioner’s value. This is more likely to happen if the business issue is thought through from the LOB perspective, rather than on the basis of which techniques, technologies or data sources will be appropriate. There are always multiple options to reach the same outcomes, albeit with differing levels of difficulty. Data practitioners and analysts should not shy away from a solution because it will be harder to deliver if it is a better fit for the client, whether from a cost, technology, timescale or value perspective. Constantly communicating examples of successful projects and outcomes across the business will reinforce the recognition that the function is client-oriented.
Leadership – if the D&A function has the right technical competency, then its leader can have confidence in the basis on which advice is being given. That in turn encourages a much more positive attitude towards forward thinking or non-conventional solutions. Leadership is not just about confidence and risk-taking, however – a D&A leader must be willing to admit if something is wrong or unknown, while committing to discovering and correcting it. They must also be willing to say no in the right circumstances, such as when being asked for something outside of permitted parameters or contrary to the broader business’s interests.
Vision – at all times, there should be a clear, consistent and well-communicated view from within data and analytics of how it will benefit the organisation and where it is aiming to lead it. This may be developed from a list of identified projects, which need to be prioritised according to the business strategy, or it may be a more cultural goal. Maintaining the vision keeps the D&A function aligned and engaged with the business.
Challenges for data and analytics
Trusted advisor status may be desired by practitioners, but it is not an inevitable outcome of the commitment to D&A by an organisation, nor is it always natural behaviour for teams. A number of common challenges have been identified by DataIQ Leaders members:
Commitment issues – data scientists can be more interested in techniques than they are in the business impact which they have. They are also less used to persisting with projects even when they encounter difficulties. The D&A leader needs to maintain focus and discipline by creating a sense of common purpose.
Data evangelism – it is easy for practitioners to become absorbed in their own self-belief that their data or model is the best and only solution for the business. This can come across as confrontational if analysts are contradicting executives who have many years of experience in the organisation, for example. The best approach, as one member put it, is to “give them the rope and they will either skip or hang.”
Poor communication – good data with a bad explanation is worse than bad data. Data literacy among non-technical executives is still relatively low and it is for the D&A function to explain clearly in the language of the business, rather than for the business to learn the terms and techniques of data and analytics.
Unrealistic expectations – both D&A functions and their sponsors can become excited by the prospect of leading-edge projects, but these are not always appropriate to pursue. One data team persuaded a broadcaster against a big data project because it needed to start at a lower level with data definitions and data dictionaries as it didn’t know what it didn’t know. Promising users that this would free them from the need to use macros in data workarounds was the carrot to get them to commit to improving data first. That also helped the data office to stay true to its vision and not get distracted by a VP who wanted to run an AI project for which it was not ready.
Avoiding blame – peer review can be highly-effective in improving the internal communication and cohesion of a data team, as well as debugging projects. But many practitioners fear being blamed for problems and errors, which can lead to them hiding issues. Creating a no blame culture gets around this behavioural barrier, but must be seen to be real through open, honest discussion that is focused on identifying improvements. Harder is to change the culture in organisations with a no failure policy. Going fast means having permission to fail – if the corporate behaviour does not support this, the D&A function will need to develop its own culture first and then clearly communicate the benefits within the success stories it tells.
