Choose one seat and one ten-metre view
Most garden bird counts fail at the first step: people move around the garden looking for birds. That movement changes the thing being measured, and the result becomes a walk rather than a count. The fix is to sit in the same chair, face the same patch, and treat roughly ten metres of lawn, hedge or feeder line as the survey area. A fifteen-minute timer starts only after the seat and boundary are fixed. On a wet Tuesday, that rule is what makes a small suburban plot comparable with a sunnier Saturday. That single constraint removes most of the ambiguity.
Set the clock before the first bird
Counting without a fixed start time creates a hidden bias: the keen observer counts longer when birds are active and stops early when the garden is quiet. The cause is not laziness; it is the absence of a boundary. Set the timer for fifteen minutes and record only birds seen or heard inside the ten-metre patch. Keep the same start time each week—early morning is usually better—and note weather and feeder state in one line. A kitchen timer works. The result is a small dataset that can be compared week to week instead of a one-off list.
Use three columns, not a species list
A free-form species list hides the most useful information: repeat visits. The problem is that a single robin and three robins look the same once the list is written. Use three columns instead: species, number of individuals at one time, and behaviour such as feeding, perching or flying over. Count the maximum simultaneous total, not every movement. The wildzy approach keeps this sheet short enough to finish before the kettle boils; the point is repeatability, not a perfect inventory. A takeaway: if two birds cannot be seen at once, they may be the same bird.
Make one habitat change before the next count
Data without a follow-up action becomes a notebook habit. The cause is often scale: gardeners assume they need a pond or a wildflower meadow to help. A smaller solution works better for a repeat count. After the first tally, change one thing—move a feeder ten feet from a window, add a shallow water dish, or leave a nettle patch uncut—and keep the count conditions otherwise identical. The result is a simple before-and-after test. Repeat the fifteen-minute count at the same time for four weeks; if the tally shifts, you have a local observation, not a guarantee. That is exactly the modest, human-scale result wildzy is designed to support.