Data Analytics

Data – The Good the Bad and the Ugly

What makes good and bad data in terms of gathering, maintaining, and using, and the drawbacks to your business of data that fails to provide useful and actionable insights.

The old IT saying ‘rubbish in, rubbish out’ is still all too true when it comes to data gathering and processing.

No matter how good your analytics and procedures are for use in data driven activities such as sales and marketing, if your data is poor then the benefits you’ll derive are at best limited and at worst counterproductive, as you’ll be developing strategies and actions based on misleading information.

A Data Strategy

To make proper use of the ever-increasing amount of data being gathered businesses of all sizes need to develop and follow a coherent data strategy. This covers the entire process from effective data gathering, analytics and the developing of data driven actions based on accurate information.

Developing a data strategy – or overhauling your present one – to avoid working with bad or even ugly data could be helped considerably by working with experienced data experts.

What is good data?

Simply put, it’s data that enables effective strategies and actions based on relevant and reliable information.

Accuracy – on a basic level, a database with every field correctly filled in with the right format and up to date (so ‘cleaned’ regularly to ensure accuracy) is good data.

Relevance – is the data serving you properly in providing the potential for sound analytics leading to actionable insights? For instance, if you wish to develop a marketing campaign for people of a certain demographic profile, does your data provide enough of the accurate records you require?

If some of the data contains people outside the demographic; has certain key elements of information missing or outdated records, it will do more harm than good.

Bias – good data is ruthlessly graded and categorised.

For example, imagine you wish to target people who informed you they were ‘excited and looking forward’ to hearing about your new product or service, and you decided a score of 7 out of 10 would qualify them as ‘excited enough’ to be contacted. Reducing the score to – say – 5 out of 10 because not enough people originally scored 7 is delusional and likely to lead to approaches to prospects who are only lukewarm at best.

What is bad data?

Generally speaking, it’s either poor data from the start or data where quality has deteriorated over a period due to, for example, none or not enough cleaning of databases and auditing.

An out-of-date prospect list would waste considerable time through bounced emails, wrong phone numbers and contact people no longer in the relevant job position (in the case of business-to-business marketing).

Clear indications you have bad data include:

  • No actionable insights – your data, once analysed fully, should point towards actionable steps you can take based on accurate facts. If there are none – or not many – then your data isn’t serving you.
  • Too many errors – if humans are manually entering data then the odd error is probably inevitable, but too many and the data begins to become counterproductive.
  • Subjective decision making – if personnel charged with planning and implementing data driven strategies lose confidence in what their data is telling them, it opens the door to decisions making based on ‘gut instinct’ and bias instead of facts.

What is ugly data?

Ugly data is information so poor that it costs considerable time, and money and not only impairs your customer’s experience but causes a negative one so possibly putting them off buying from you in the future.

Negative customer experience – poor data that causes you to respond too late, too ineffectually or even not at all to a customer request whether for more information, contacting them or handling an aspect of their purchase.

Bottlenecks and delays – data that doesn’t ‘flow through’ a business properly can cause bottlenecks so delaying business activities.

For example, badly organised data can mean reports are too convoluted so delaying decision making and action.

Wrong metrics – could also be ‘inappropriate metrics’ or ‘vanity metrics’. For instance, a new company establishing itself may be looking for exposure, so pulling in data based on social media ‘likes’ is relevant intelligence.

On the other hand, a more established company has probably moved beyond this and is usually looking for more leads and conversions based on social media activity. Becoming seduced by ‘likes’ is possibly the wrong metric for them to focus on, and time spent gathering and processing this type of data is a poor use of time and resources as the insight is no longer relevant.

As said earlier, a data strategy – maybe from the ‘top down’ – may be required if your data isn’t serving you as it should. Talk to a data professional to improve your data use.