Noise or Music? - The Insights Blog

[Survey] Do you TRUST your analytics data?

Categories: Metrics understanding, Privacy and Accuracy / Comments: 6

Share Button

The question of data trust (i.e. data quality) is what I view as the biggest issue facing the digital analytics industry at present. Why? Because it is so easy to collect junk data online. And businesses will not act on data unless they can trust its quality. The result is data paralysis – no one wants to dig deep in to the data for fear of what will be uncovered…

Is data trust an issue for you? Take the 1 min trust survey…

Please add to this research by quantifying your level of trust in your data by providing five values of trust in the survey below. The tool you use to collect/report on your data is irrelevant. After submitting, you will see the results so far, and when enough results are collected I will conduct the analysis and present the findings in a follow-up post.

The focus here is on data owners and data users. If you are an analytics consultant, please *do not* submit this survey… it would skew the results!

Trust is a subjective question. That is, there is no black and white answer – rather shades of grey. Therefore for the following five areas of your website, please grade your level of trust in your collected data (regardless of tool used).

For the basic numbers e.g. visitors, pageviews
(1 = no trust all, 10 = complete trust)
For understanding marketing performance
(1 = no trust all, 10 = complete trust)
For e-commerce data, or engagement e.g. calls to action
(1 = no trust all, 10 = complete trust)
For segmenting e.g. customers, prospects, staff
(1 = no trust all, 10 = complete trust)
For understanding the user experience on your site
(1 = no trust all, 10 = complete trust)

 

 

Share Button

Comments (most recent first)

  1. Gerry White says:

    With adblockers often blocking gtm I struggle to fully trust anything other than it being a representative sample.

    The biggest other issue is understanding user intent with no other true signals, why did they bounce, why did they suddenly drop out was it because they ran out of time, or something else…

  2. Alex says:

    From the experiences in our company I can tell that people’s trust in data declines the more they experience troubles with the implementation, when implementation isn’t done correctly and thus sometimes the figures are weird. The division of labor between developers and analysts causes much trouble, but I believe this is unavoidable in at least bigger companies.
    Devs tend to copy and paste and sometimes that messes up the data qualit and leads to a lack of trust in the data quality among analysts.

  3. Richard says:

    Trust that the data is in correctly and accurately is one thing – but what about trust about where the value of your data is going? (i.e. being used to optimize some ad buyer’s digital spend in a competing company within your own industry)

    • @Richard – can you expand on your comment please. For example, are you referring to sending your data to Google and Google somehow using this data to manipulate their Adwords auction system?

      • Richard says:

        @Brian – along that idea yes. But it’s not that they manipulate it on purpose, that just happens automatically because you are telling the advertising system about: where your customers come from, (helpful data-points for the ad network), which of your customers are more relevant to your industry, which ones spend money with you (if commerce tracking is enabled).

Leave a Reply

Your email address will not be published. Required fields are marked *

Anti-spam question (required):

© Brian Clifton 2018
Best practice privacy statement