Data conversations

We’re learning a great deal on our data journey at Citizens Advice. I think it’s worth sharing where we’re at.

18 months ago I would have recommended “talking about your data more”. I still recommend that of course!

Given experience in the intervening time I’m able to describe something more than that original intent, with some structure to it as well.

Maybe you work in an organisation where everything I describe here happens already. That’s great, I’d love to hear about your experiences.

Maybe the data you work with is different to ours —perhaps larger and updated more frequently, or smaller and updated less frequently. I think that the things I’m describing here could still be useful.

The ‘data’ that I’m talking about here are things like:

  • Channel activity for example website and telephone
  • Client volumes and demographics
  • Topic trends at high- and more detailed levels
  • Client outcomes — experience and satisfaction from survey data

Principles

There are a few guiding principles behind what we’re doing. I think the most important one is

Conversation, not presentation

Everything I’m describing here should encourage everybody to talk about the data. It’s not a one-shot pitch — this is ongoing practice and there’s no end to it. Shared understanding builds over time and feedback is essential.

Talk about the whole picture

We have a complex organisation with many channels and services. We try to make sure that we are talking about the relationships between things rather than focusing on isolated data sets.

Using as few products at possible

We have a main data product called the ‘Service Dashboard’ which brings a wide variety of our data together in one place. We try to use this product to meet the needs of as many audiences as we can.

We have a preference for presenting from this product or other dynamic reports and dashboards over transposing data into a slide deck.

Keep it frequent

We have a weekly pattern. This is appropriate because of the wide variety of data that we have, rather than because of the cadence of our data (where trends play out over months rather than weeks). This frequency keeps what’s happening with our data at the front of peoples’ attention though, and there are a wide variety of topics we can cover.

If you’re working with data and it doesn’t change that much I’d still recommend talking about it once a week.

Repeat and reuse

The forums are porous. We present the same material in multiple places. We use the forums to generate content that people can revisit and share.

Iterate

We learn by doing and regularly reflect on how things are progressing and make changes accordingly — both to our practices and to our data products.

Here are the four practices I recommend trying if you’re not doing them already. This isn’t an exhaustive list and it will develop further I’m sure.

1. The top team conversation

We have a rolling programme of short weekly data updates to our Executive and Directors team. They happen at the beginning of the week as part of an existing ‘start the week’ meeting.

When we started this James (who is part of the Executive team) gave these updates but increasingly we’ve brought in other voices. It’s a collective effort to achieve a weekly update and we have input from the Data Science team, our counterparts in the Impact team, and others particularly from Operations.

It is a conversation because this top team get to provide feedback and set priorities for questions to answer. This has been supported by a fairly regular retrospective.

The same material gets shared with all staff from the National Citizens Advice organisation on Workplace and we will be sharing more widely with the Network of 250+ local Citizens Advice across England and Wales too.

In order to keep to the weekly pattern we need a decent but flexible forward plan, and we try to keep a month ahead of ourselves on that. There can be significant lead time for the some of the work required to answer the questions asked. Seemingly simple things can be complex and vice versa.

This practice has driven some of the most forward-thinking and fresh data work that I’ve been involved in since I started this role in late 2019. We know things that we didn’t know 6 months ago, which is itself a measure of improvement.

This practice also involves the most collective effort and preparation. That feels appropriate. This is the level of the organisation that can be supported through data to make the most significant decisions.

What are the benefits?

  • Builds a collective understanding
  • Drives improvement in our evidence base
  • An opportunity to prioritise based on the most important questions to answer
  • Should lead to more informed decision making

Examples from Citizens Advice

In recent weeks we’ve covered these topics:

  • High level client outcome and activity numbers from across our service for the past year.
  • New data that tells us about the impact of our online advice content.
  • New data that tells us about the variety of different telephone service models used by the Network of local Citizens Advice.
  • New analysis on depth of issues our clients experience and the strong relationships between different issues (for example housing and debt).

2. Data at the start of every meeting

Well, not every meeting. Let’s imagine you’ve got a regular team meeting, or an ‘all hands’ session. The practice here is to do a tight 5 minute update on data at the start of the meeting.

As an example, I do this at the weekly meeting for the leadership team I’m part of. 5 minutes translates to 3 or 4 talking points about our data. Committing to this regular practice means that I have to be engaged with the data that we’re working with — I have to look for patterns and trends to highlight. I can also bring in insights from the other forums.

I write up those 3 or 4 talking points and share the document with the team. This is a further commitment, but it’s worth it because it can be shared with the whole group. They can refer back to it and consider the points in their own time. Also it means that nobody in the group is left out if they aren’t at the meeting. Finally these documents are open, they can be shared more widely if my peers think there’s value in doing so.

What are the benefits?

  • Provides context and helps to break out of siloed thinking
  • Builds a shared understanding
  • Builds expertise in talking about data from a variety of sources and how it interrelates
  • Keeps you curious

Examples from Citizens Advice

Here’s an example Google Doc with 3 real data talking points I’ve covered recently.

3. The regular open forum

This is our most established practice. We began it soon after the start of the first pandemic lockdown in 2020. Tom (chief analyst) wrote about it. It is a fortnightly session that lasts around 45 minutes. It is open to all staff from the National organisation. We get around 30 people on the call each time, from a variety of teams and backgrounds.

We use the Service Dashboard, presenting this to the group and having the data specialists who are responsible for each category of data talking about the latest trends. For example Mankeet (senior data analyst) covers website trends.

We take questions from the group. One of the most valuable aspects of this session is that colleagues from Operations or Policy often provide valuable insight and context for what we’re seeing in the data. It’s very much a conversation.

Of the four practices this is the one that gives many people the opportunity to understand and describe the narrative of what’s happening across our service as trends play out over months and years. There’s an element of oral history to it, which could be seen as a weakness because some of the explanations for patterns that we’ve seen aren’t documented. However, we record the sessions and post them in a dedicated Workplace group so people who can’t attend can participate in their own time. And we see the narrative that gets developed reflected consistently in other work that we do, which is a strength.

What are the benefits?

  • Builds a collective understanding
  • A rich exploration of the data given the expertise involved
  • The narrative stays current and is reflected in other forums
  • Provides early sight of trends and issues that can be highlighted or escalated elsewhere

Examples from Citizens Advice

The Service Dashboard is updated weekly. This forum has established regular content. We look at client trends (numbers and demographics), advice topic trends, website trends (topics, top pages, volumes, search terms), and telephone and webchat trends (volumes). We can compare back to a variety of different time periods but we find comparing year on year to be most valuable because of strong seasonality in our data.

4. The deeper dive

We run a roughly weekly 30 minute session with an agenda that covers a wide variety of data topics. I say roughly weekly because it’s 4 weeks on and 1 week off with the slot being at a different time each week to encourage attendance.

The session is open to all National staff. There’s a fairly large invite list and we tell everybody what we’re going to be covering and share materials in advance. We get around 15 people at each session.

This practice provides an opportunity to go deeper on new analyses and insight, presented by the specialists who have done the work. It also provides an opportunity to talk about our data work ‘behind the scenes’, for example developing new services and standards. It has been really valuable for developing our data strategy work in an open and collaborative way. Finally we’ve had guests from other organisations telling us about their experiences — that’s particularly valuable when you get to hear about shared challenges and how people have approached them.

We record the sessions and post them in a Slack channel. Necessarily this forum generates a fair few slide decks. We share those too, and make sure that they contain links to other resources.

What are the benefits?

  • Showing what really goes in to the work to an audience who wouldn’t otherwise understand
  • Doing justice to data specialists’ effort by having time to go into greater detail
  • Bringing in fresh perspectives for shared problems
  • Developing in the open, particularly strategic work

Examples from Citizens Advice

In recent weeks we’ve covered these topics:

  • Data strategy principles framework — how we’re getting owners for these principles from across the organisation
  • A new ‘service’ approach to data, building initial versions of services that we can iterate. Examples include a service for data about our volunteers and a service for data about the Network of local Citizens Advice.
  • Collaboration between Data Science and Product to decommission a legacy system, and establish a new primary source of data for reuse by multiple systems as a result.
  • A new way to visualise client volume data, showing it across our entire service at a high level for the first time.

One day I’d like to put together a ‘playbook’ for data specialist work but it’s a daunting task. I can break it down into smaller pieces though. This post is a first attempt at that.

Please get in touch if you’ve got any thoughts — you can find me on Twitter and occasionally on LinkedIn.

Thank you for reading.

--

--

--

Head of Data Science at Citizens Advice. These are my personal thoughts on work.

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

Why are Data Science Jobs Hard to Get When They’re the Most In-Demand?

Building Data Engagement @ Zencargo

Covid-19 Interactive Visualization

A short introduction about Raft: a distributed consensus algorithm

Getting into data visualization — where should I start?

Mistakes in Applying Univariate Feature Selection Methods

What to study before a Data Science interview

Statistical Analysis with Fake Data

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Dan Barrett

Dan Barrett

Head of Data Science at Citizens Advice. These are my personal thoughts on work.

More from Medium

The future of leadership is data-driven

Embrace Complexity — Part 3: Cloud

How measuring the wrong metrics can cost you your business

Knowledge Discovery: The “Heuristics — Skills” Mapping