How fast does contact data decay?
I don’t know how fast contact data decays. So I thought I’d find out. Rather than research the internet and read existing reports, I thought I’d gather some primary data and see if I could make some educated guesses about the patterns I found. Essentially an initial screening to identify the main factors for subsequent analysis / product development.
My data collection was basic and involved an online survey with a single question being sent to my contacts on LinkedIn. A total of 50 people responded from about 200 invited. Here are the results to the question, When was the last time you changed your.........
What can I draw from this data? First off, work data is fast decaying.
The first four items relate to people’s work and the rate of change is easily the fastest. Within 3 years the majority of people will have changed their work circumstances. So a CRM solution without a strategy to continually refresh and keep accurate employer details will be obsolete in 3-years. That is a significant statement and anyone reading this who needs support in refreshing their contact data should get in touch.
One approach I’m researching is the potential to associate a CRM contact with their LinkedIn profile. The concept is LinkedIn is generally the place people go to to update who they are working for. So rather than having to ask someone, just get automatic updates from LinkedIn. LinkedIn do provide a RESTApi to achieve just this, but throttle the number of requests allowed. Ongoing research here but bearing fruit.
About a quarter of people moved address within 3-years. So the rate of decay gives room to address the problem. Luckily the Royal Mail provide the National Change of Address Register. By regularly matching contacts against this register the new address can be found and adopted so avoiding wasting money sending materials and having to make contact to discover the new address.
People’s home details decay at a much slower rate. Take home email address, my impression is people stick with them, like bank accounts, for life. Seldom, if ever, changing. A couple anecdotes to illustrate the point. My home email goes back nearly 20 years and discussing email with a colleague they are still using the original one they used when subscribing to AOL. Same seems to go for twitter accounts. My take away being people’s home & personal data is long-lasting and less prone to decay.
So three initial conclusions and pointers on where to go next in the discovery process. A quarter of people had never used a FAX machine.
Any thoughts?
I do of course agree that refreshing data at an optimum point but based on volatility and say a function of the mean time it takes for the data item to decay would reduce decay/entropy in your contact data. Predictive modelling if you had such data as described above could help in designing a cost-effective refresh process
Robin, Surely the dataset you have based your research on is not particularly volatile? If someone does not change their job often or move house etc then there would be no need to amend the data. Have I misunderstood? If you had asked, for example, "how long did it take you to change your job description after you changed job" it would make more sense?