What it means
The bias exists because recent information is easier to recall and feels more vivid than older information. Human judgement was built for environments where the latest news usually was the most relevant, which works well when dodging danger and badly when estimating a long-run average.
Financial markets are full of noise, so the most recent data point is often the least informative one. In business the cost is usually measured in badly sized decisions.
A sales forecast built off last month's exceptional figure leads to over-ordering, over-hiring and a cash squeeze, while a forecast built off one weak month leads to cutting capacity just before demand returns. The bias does not make people wrong in direction so much as wrong in magnitude.
It is particularly damaging in investing, where it drives the pattern of buying after prices have risen and selling after they have fallen. Investors pour money into last year's winning fund, extrapolating a three-year run indefinitely, then abandon it after a poor year, converting a temporary decline into a permanent loss.
The defence is structural rather than motivational, because willpower rarely beats a cognitive habit. Written investment policies, forecasts built on rolling multi-year averages, pre-commitment rules such as scheduled rebalancing, and decision journals that record what you expected and why all reduce the pull of the latest headline.
Performance reviews that require notes from every quarter, not just the last one, work the same way. A useful counterweight is to ask what the full history says before looking at the recent data.
Anchoring on the long-run base rate first, then adjusting modestly for what has just happened, produces far better estimates than starting from the newest number and reasoning outward.
In practice
Real-world examples.
Example
A restaurant group has three exceptional summer months and signs leases for four new sites on the strength of them. Trade returns to its normal level in autumn, and the group is left carrying fixed rent it cannot support.
Example
An investor sells an entire equity holding after a 12% fall, convinced the decline will continue, having ignored that the same holding gained 90% over the previous six years. The market recovers within nine months and the loss becomes permanent.
Example
A manager gives a strong performer a mediocre annual rating because of one difficult project in November. Ten months of excellent work sit in the file unmentioned, and the employee leaves within the quarter.
Think of it
“Recency bias is focusing too much on recent events-what just happened dominates thinking.
Formula
Calculation
There is no standard equation for a behavioural bias, but the size of the error can be measured directly by comparing a forecast built from the most recent period against one built from the full record.
A software company's sales team wants to set next year's target. Last month was exceptional at $480,000 of new bookings, and the natural instinct is to annualise it: $480,000 x 12 = $5,760,000.
The trailing twelve months tell a different story. Total bookings over that period were $4,200,000, giving an average month of $4,200,000 / 12 = $350,000, and a full-year run rate of $350,000 x 12 = $4,200,000.
The extrapolation gap is $5,760,000 - $4,200,000 = $1,560,000. As a percentage, $1,560,000 / $4,200,000 = 0.371, so the recency-driven target is 37% higher than the historical run rate supports.
If the company hires and buys infrastructure against the $5,760,000 figure and actually books $4,400,000, it has committed to roughly $1,360,000 of cost that no revenue arrived to cover.Case study
Seen in the real world.
Sable Ridge Outfitters is an illustrative, fictional outdoor equipment retailer used here to show recency bias at work in inventory planning. After an unusually wet spring drove waterproof jacket sales up 80% year on year, the buying team assumed the pattern would repeat and tripled its order for the following season.
The fictional buyers had eight years of data available showing that jacket demand fluctuated widely with weather and that the long-run average growth was closer to 4% a year. Nobody looked at it, because the most recent season was vivid and the supplier was offering a volume discount that made the decision feel prudent.
The following spring was dry. Sable Ridge ended the season with roughly $1,900,000 of unsold stock, cleared much of it at 45% off, and tied up warehouse space it needed for a new product line. The illustrative fix was a simple policy change: every buying decision above $250,000 now has to show the five-year demand history alongside the most recent season.
Watch out
Common mistakes.
- Building forecasts by annualising the most recent month or quarter. Short periods are dominated by noise, seasonality and one-off events, and extrapolating them exaggerates whatever just happened.
- Confusing recency bias with reacting sensibly to new information. Genuine structural change deserves weight; the bias is over-weighting ordinary fluctuation as though it were structural.
- Believing awareness alone fixes it. Knowing about the bias barely reduces it, which is why written rules and longer data windows matter more than good intentions.
Questions
People also ask.
How is recency bias different from availability bias?
They are close relatives: availability bias over-weights whatever comes to mind easily, and recency bias is the specific case where what comes to mind easily is simply the most recent thing.
Does it only affect individuals?
No, committees and whole organisations show it too, particularly when reporting packs lead with the latest month and bury multi-year trends in an appendix.
What is the simplest practical defence?
Look at the longest reliable history first, form a view from that, and only then examine the recent data and decide how much to adjust.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%Related