How to Measure Writing Improvement as a Poet (Really)
Streaks and word counts measure activity, not skill. Here is a judge's system for measuring real poetic growth: the re-draft test, a quarterly six-step audit, and the one number that reliably predicts improvement.
Last winter a poet emailed me two versions of the same poem, written eleven months apart, and asked a question I loved: could I tell which one was newer? I could, in about four seconds. Not because the newer draft had fancier vocabulary. Because the newer draft trusted its own images enough to cut the explanations. That is what growth actually looks like on the page, and it is why so few poets know how to measure writing improvement: they are counting the wrong things entirely.
Poets track streaks. They track word counts, poems finished, followers gained, submissions sent. Every one of those numbers measures activity. None of them measures skill. I have read thousands of contest entries over the years, and the poets who improve fastest are always the ones who audit their own work the way an editor would: on a schedule, against a rubric, with evidence they cannot argue with. This piece gives you that system.
Why Your Current Metrics Are Lying to You
A runner who logs miles but never checks pace can train for a decade and get slower. Word counts are miles. A hundred poems written with the same three habits is one poem written a hundred times.
Streaks are worth keeping, and I say that as someone who believes discipline beats inspiration in nine rounds out of ten. But a streak tells you that you showed up. It says nothing about what happened after you sat down. The same goes for acceptances, which are so noisy with editor taste, timing, and luck that a single yes or no tells you almost nothing about your trajectory.
Improvement is a change in what you can do, not a change in how much you did. So the instruments have to measure ability. There are exactly three kinds worth your time: comparison against your past self, scored evaluation against fixed criteria, and the reactions of live readers. Everything below falls into one of those three buckets.
How to Measure Writing Improvement: The Re-Draft Test
Here is the single most honest instrument I know, and it costs nothing. Take a poem you wrote a year ago. Do not reread your final draft first. Take only its subject, its core image, its occasion, and write the poem again from scratch, cold. Then put the two versions side by side.
Everything you fixed automatically, without noticing you were fixing it, is skill you have gained. Everything you repeated is your current ceiling. The re-draft test works because it removes the flattery of memory. You cannot tell yourself you would never break a line on a preposition anymore when the fresh draft in front of you just did it twice.
I ask poets to run this test twice a year. One poet I worked with discovered that in every re-draft, across three years of tests, her endings arrived one stanza too late. That single finding was worth more than every kind comment she had ever received, because it was hers, it was specific, and it was fixable.
The Quarterly Audit: A Six-Step Framework
The re-draft test is a snapshot. The audit is the film. Run it every three months, and do it in one sitting so your scoring mood stays constant.
- Pull five poems from one year ago. Not your favorites. The first five in the folder, so selection bias cannot rescue you.
- Score each from 1 to 5 on five habits: freshness of images, intention in line breaks, strength of verbs, whether the ending is earned, and whether the poem holds one idea or fights over three.
- Score five recent poems on the same rubric, the same night. Same chair, same coffee, same standards.
- Write one sentence per old poem: what would I cut today? If the answer is nothing, either the poem was excellent or you have stopped growing. Be suspicious of nothing.
- Log the numbers somewhere boring. A spreadsheet, a notebook page, anywhere you can see two quarters side by side. Growth hides in trends, not in single readings.
- Circle your weakest column and make it the only focus of the next quarter. One habit per quarter. Poets who try to fix everything fix nothing.
The rubric matters less than the consistency. You are not trying to produce an objective truth about your poems. You are building a fixed ruler so that when the measurement changes, you know the poem changed, not the ruler.
A Worked Example: The Same Grief, Fourteen Months Apart
Here is a before and after of the kind I see constantly, condensed from patterns in real submissions. A poet writing about a grandmother's death begins here:
I was so sad when my grandmother died, / it felt like the whole world was ending.
Fourteen months and a lot of deliberate reading later, the same poet writes:
My grandmother's kettle still faces the wall, / spout turned away, the way she left it.
Notice what changed. The first version names the feeling and asks the reader to take it on faith. The second version stages the feeling inside an object and lets the reader arrive at the sadness on their own, which is the only way readers ever really arrive anywhere. If you want a number to track, track this one: count the named emotions per poem. Sad, lonely, heartbroken, joyful. Watch that count fall over a year. A falling count of named emotions is one of the most reliable signs of a maturing poet I have ever found, because it means the images are finally doing the emotional work. Poems built around grief are the clearest place to watch this shift happen, since grief tempts every writer toward abstraction first.
External Instruments: Rejections, Scores, and Live Readers
Your own audit can drift, so triangulate with measurements you do not control.
- Track rejection quality, not rejection count. There are tiers: the form rejection, the form rejection with an encouraging line, the personal note, the please send more, the shortlist. Moving up a tier is real, measurable progress even while the answer is still no. I climbed those tiers for four years before my first acceptance, and the climb itself kept me writing.
- Use scored feedback where you can get it. This is exactly why I like contests that show their work. Writory's monthly contest, for instance, returns a five-criteria breakdown to every single entrant, scoring creativity, emotion, structure, language, and theme, which means you can see your score move across months on the same fixed scale. A rising structure score across three entries is data. A single win or loss is weather.
- Run the recall test with a live reader. Read a poem aloud to one person. Wait a day. Ask what they remember. The images they can still see are the poem's real content; everything else was scaffolding. Track how many images survive the night. Two is good. Three is a poem working hard.
Growth Feels Like Getting Worse. That Is the Trap.
Here is the part nobody warns you about. Your taste improves faster than your skill. Around the time your poems genuinely start getting better, they will start looking worse to you, because the reader in your head has leapt ahead of the writer in your hands. Poets quit at exactly this point every year, and it is always the wrong point to quit, because rising dissatisfaction with old work is the single most reliable leading indicator of growth there is.
The poet who can name what is wrong with last year's poem has already outgrown the poet who wrote it.
So when you run your quarterly audit and your old poems make you wince, write the wince down. Name it precisely. Weak verbs, late endings, line breaks that fall wherever the clause happened to end instead of where the tension lives. If you cannot name a line-break habit yet, spend an evening with enjambment and then reread your old work; the difference between a break that means something and a break that merely happens will jump off the page. Every wince you can articulate is a skill you have already half acquired.
The Ninety-Day Ledger: Start Tonight
Do not wait for the start of a month or the purchase of a beautiful notebook. Tonight, pull one poem from a year ago and run the re-draft test on it. Score both versions on the five habits. Open a plain spreadsheet, label six columns, and enter your first row of numbers. That is the whole ritual: twenty minutes, once a quarter, plus one honest re-draft twice a year.
In ninety days, do it again. The second row is where the ledger comes alive, because now there is a direction. Maybe your verbs climbed and your endings sat still. Good. Now you know what the next ninety days are for, and there are enough craft resources aimed at endings alone to fill that quarter twice over.
Poets love to ask whether they are getting better, usually late at night, usually in a tone that expects no answer. The question has an answer. It has always had an answer. You just have to stop asking the streak counter and start asking the poems.
By Team Writory, for Writory.
Enter a Writory poetry contest
Enter the September contest and have your own poem read and scored. That first entry makes you a Writorian.