We have become data thirsty. We want all the data we can extract, intercept and capture, because data is the new knowledge. Data is power.
With a state-of-the-art CRM like Salesforce, and its all-encompassing platform, we have the luxury to store, process and manipulate our organisation’s data in limitless ways.
Yet, have you paused from your 100-mile-an-hour-life to think about the real cost of data? I’m not talking about purchasing lists of data. I’m referring to the indirect and hidden costs that are accrued when dealing with data.
We would all like to think that we run, or work for, data-driven organisations. Decisions are made within these organizations on the basis of data, and are 19 times more likely to be profitable; with statistics like this, it’s no wonder we pull out all the stops to acquire data. Interestingly, data-driven still appears an aspirational state; apparently, only 1% of organizations agree that they are a data-driven organization, and just over half (54%) of decisions are driven by data (Information Age).
Something doesn’t seem right here between the two realities. Where’s the disconnect?
I speculate that it may have something to do with losing sight of CRM data principles.
Data is expensive: Introducing the curse of too much data.
I read one thought-provoking piece by my all-time favourite marketing don: Seth Godin. His scaled-down, unassuming blog is a treasure chest of golden knowledge. This piece got me thinking about the curse of data, and how everyone with a powerful CRM is at risk of overlooking this.
The scene starts with a humble alarm clock, and leads up to the crux: “More data isn’t always better. In fact, in many cases, it’s a costly distraction or even a chance to get the important stuff wrong.”
Although Seth intended for his post to be profound and somewhat ominous, I have lifted the fog and bought the 3 principles he outlined into the context of Salesforce.
A reminder every once in a while doesn’t cause any harm, or face the risk that data becomes more inhibiting than insighting.
3 Simple Principles for Data Wisdom
Principle 1: “Don’t collect data unless it has a non-zero chance of changing your actions”
This is a double negative that makes you stop and think – a dramatic pause, if you like. This can be rephrased as: only collect data if it has a chance of changing your actions – simply put, only collect data that serves a purpose in your organisation.
I’m sure you’ve heard the saying, ‘more is less’? More of the wrong data does not equate better insight. In fact, figures from the annual Data Security Confidence Index report that 65% of companies have “too much data”, in that they do not have the resource readily available to analyse.
Data capture for the sake of data capture is a fruitless mission. Some questions to ask yourself to identify if your Salesforce org has a field and data redundancy issue worth investigating:
- When was the last time you reviewed your org’s field usage? Obsolete fields are rife in Salesforce orgs, due to the ease of adding new custom fields. I recommend using AppExchange tools to do this, such as Elements or Field Footprint.
- Can Business Stakeholders find the answers to their burning questions without getting lost?
- Is this data included on Reports and Visualised as Dashboards? Reports and Dashboards legitmise data, being used in business decision making (exceptions to this are if it is a legal requirement to hold this data, or a system requirement eg. integration ID)
Principle 2: “Before you seek to collect data, consider the costs of processing that data”
The data processing cycle involves more than what we can see with the naked eye. In the background, some types of captured raw data need to undergo transformation, so these data points become information. By changing the format of incoming data to a ‘friendly’ format your CRM can work with, this meaningless raw data is given context – a place in the ‘story’ of your business, comprehensible to end users.
In the age of Big Data, businesses have sprung to collect data. The amount of data business are collecting has increased exponentially due to new overt (where the individual discloses data to you) and less overt ways (such as cookies, website heatmaps, voice command devices), which come together to what is known as your ‘digital footprint’.
What processing costs are we at risk of overlooking when processing data in and around the Salesforce platform?
- Data Analysts & Other Specialists: tools and technology is one thing, but having data specialists completes the puzzle. Increases in data volumes and sources will require specialists, whose skill-sets (ideally) straddle Big Data and knowledge of the Salesforce platform. Should you factor in the cost for a Salesforce Technical Architect? They are needles in the haystack and could end up soaking up extra budget.
- GDPR Compliance: now consider the legal review process each time you wish to add a new data source or share data with partners. Appointing the Data Processing Officer (DPO) responsibilities on a member of your team may become too overwhelming, instead requiring a full-time employee (FTE) for the role. It appears GDPR compliance requirements are stretching businesses; looking at UK businesses in the Data Security Confidence Index, 59% are falling short on GDPR compliance checks and procedures.
- Storage: Salesforce storage is famously expensive, and purchasing extra storage will bump up your licenses costs.
- API Calls: these are what Salesforce uses to communicate with integrated 3rd party tools (off platform). API calls will perform data queries (searches), add, update, and delete data, amongst others. However, there are limits to the number of API calls Salesforce can make, depending on your license type – so factor in a license upgrade if necessary.
Principle 3: “acknowledge that data collected isn’t always accurate”
…and consider the costs of acting on data that’s incorrect.”
We fall in love with Salesforce and the power it delivers to our business operations, especially when data comes alive with Reports and Dashboards, and now Einstein Analytics. Users go mad for charts and pies and gauges and donuts that visualise the state of our business. Is the data that is being pulled into these reports actually accurate?
I recently stumbled upon ‘Survivorship bias’, and recogised how this form of selection bias could easily be skewing the conclusions we draw from our CRM data. Survivorship bias leads us to overly-optimistic beliefs by using the ‘survivors’ (eg. data that has made it into your Salesforce dashboard) for business insight, while ignoring failures (because they are not ‘visible’).
(PS. Here’s a very interesting article on the origins of Survivorship Bias).
What if the data in our Salesforce org isn’t painting a picture of reality, therefore giving every user ‘happy ears’?
There has always been risk associated by reporting on ‘as-is’ states, or outdated pipeline data giving inaccurate sales forecasts.
Luckily, technologies like Einstein Discovery can handle very large datasets with complex structures. It uses machine learning to uncover trends that may take other datasets into consideration, and paint a more realistic picture to business stakeholders.
Finally, it’s not just a full, realistic view of reality we need from our data. We must demand clarity and consolidation. Seth Godin’s “AM/PM Problem” proved that the data point (time) of ‘8’ without the AM or PM data point would leave ‘8’ open to intepretation, and therefore error. Ensure that data cannot be misinterpreted if accidentally isolated and used for misguided business insight.
I know that the Salesforce Ben readership (more likely than not) manage, capture, process, protect, transform, manipulate, or analyse data in their day-to-day jobs. Let’s be frank, there are few jobs that are isolated from data in 2019.
A CRM like Salesforce is both a blessing and a curse as we are able to store, process and manipulate our enterprise data in limitless ways.
It’s clear that a best-of-breed Salesforce mitigates the potential costs (direct and indirect) of ‘too much’ data; however, reminding ourselves of 3 principles for good business data management between systems and people is essential. Technology is only half of the story.