I get pushback from the NRF each year after my Black Friday discussion. They want to engage in more spin.
Here are 10 questions I have prepared for them. Feel free to make further suggestions.
1. Your news release was headline “Consumers will spend 4.1 percent more than last year during winter holidays” – why is this not a forecast?
2. Similarly, you headline this release: ”NRF forecasts holiday sales will increase between 4.3 and 4.8 percent.” – how can you claim that this forecast is not a forecast?
3. Black Friday Sales data is reported as if it were actual sales data: See 2017’s Consumers and retailers win big over Thanksgiving holiday This sure looks like actual sales data in the 2nd paragraph:
“Average spending per person over the five-day period was $335.47, with $250.78 — 75 percent — specifically going toward gifts. The biggest spenders were older Millennials (25-34 years old) at $419.52.”
It is only in the 8th para that the reader learns this is only sentiment data. Please explain
4. When media misreports this not as a sentiment read but instead as actual sales data, what steps does the NRF take to correct this? Did you send out corrections in 2016? 2017? 2018?
5. Who did you send corrections to — WSJ, NYT, WaPo? — to correct this annual error?
6. Are you actually suggesting that because you report this as sentiment, its ok, even if it is terribly inaccurate as to actual retail sales?
7. What is your response to the academic research that people are terrible at forecasting future behavior?
8, Similarly, what is your response to the criticism that people are not very good a recalling past spending with any degree of accuracy?
9. When we get the actual sales data, it often differs from the NRF estimates by a double digit percentage – how do you respond to this variation?
10. What steps have you taken to make your estimations more accurate? How often is this model significantly improved?’
BONUS QUESTION: What is the NRF position on stores being open on Thanksgiving Day?
I will update my column, and make any corrections as necessary upon receiving a credible response . . .
