Stuck at 6.5 in Writing? My Six Weeks and 50 Essays with Mike’s Marking Service (Part 1 of 2): Why I Looked for Marking, My 50-Essay Score Record and the Criterion Scores

IELTS · Practice Methods中文版　Updated 2026/11/09

In this post: Why I looked for essay marking at 6.5, what Mike's service included, my week-by-week record of 50 essays, and what the four criterion scores showed me.

1. Why I looked for essay marking: stuck at 6.5 in Writing, one score away from my offer

In April 2024 I received a conditional offer from the University of Cambridge. The language condition was IELTS 7.5 overall with at least 7.0 in each skill. Writing was my biggest headache: I was stuck at 6.5, and whatever I did, I was half a band short.

Being “just half a band short” is the most frustrating place to be. A 6.5 means your English isn’t actually bad, but the score report doesn’t show what is missing. The official result gives only one Writing score; it doesn’t tell you whether Task Response or Grammar pulled you down (if you want to know how to find your criterion scores, see my post on looking up IELTS sub-scores (in Chinese)). When I marked my own essays, I often felt “but I did cover everything”, and I couldn’t find a concrete problem.

So for about six weeks, from 20 May to 2 July 2024, I sent my essays to Mike for marking: 50 essays in total, both Task 1 and Task 2. This post is the full record of those six weeks.

2. Mike’s marking service: the parts I actually used

For me, the key thing was that every essay came with four criterion scores. Many marking services give only an overall score, or only correct the grammar. Criterion scores tell you directly whether an essay was pulled down by TA, or by CC, LR or GRA. After 50 essays, the numbers speak for themselves about where my weaknesses were (I go into this in section 4).

The other important point was resubmitting after corrections. For almost every essay, I revised it based on the marking and sent it back. The first draft exposes the problems; the second draft is where you actually fix them. This takes a lot of time, but I think it was the most useful part of the six weeks.

3. Six weeks, 50 essays: my score record (with a weekly table)

The figures below are my own calculations from the overall scores of the 50 essays. The weeks start on Monday 20 May, and the last week includes 1–2 July, so it is nine days long.

Key numbers

Looking only at the averages, going from 6.75 to 7.05 seems like an improvement of just 0.3. But for someone stuck at 6.5, that 0.3 means a lot: in the first two weeks more than a third of my essays scored below 7; in the last two weeks only 3 did. And all of my Task 2 scores of 7.5 came in the last two weeks, which shows it wasn’t luck producing the occasional good essay. My consistency had improved.

Scores don’t only go up: week 4 actually dropped

The weekly table shows that week 4 (10–16 June) averaged 6.67, lower than week 2. That was the first week in this batch that I wrote maps and tables, and in Task 1 I took on several chart types at once: two maps, two tables, one process diagram and one mixed chart. New question types often lose marks at first; my hospital map on 10 June got only 6.

I think this is actually where intensive marking proves its value: it exposes “unfamiliar question types lose marks” before the real test. Without someone scoring my essays, I might only have discovered in the exam room that my maps were weak. I wrote only one essay in week 3, so that week’s average doesn’t tell you much.

4. The criterion scores showed me which one was pulling me down

This is what I think made the marking service most worthwhile. Here are the criterion scores of the 50 essays, compared between the first two weeks and the last two weeks (range scores such as “7–8” are counted as 7.5):

Splitting Task 1 and Task 2 makes the problem even clearer:

Task 2: the problem wasn’t ideas, it was paragraph organisation

At the start, my Task 2 TR was often 8, which meant my arguments themselves were fine; what pulled me down was CC, often only 6. In other words, I could come up with content, but the way I ordered paragraphs and linked sentences didn’t read smoothly enough for an examiner. Once I knew that, the focus of my practice had to shift from “finding better arguments” to “fixing the order of sentences in each paragraph”. The change shows in the numbers too: from 19 June onwards, all 8 of my remaining Task 2 essays got CC 8.

Task 1: the problem wasn’t English, it was accuracy

My Task 1 CC averaged above 7, but TA kept dragging the score down: six essays got only TA 5 (across bar charts, process diagrams, maps and line graphs). When I was writing this series, I went back and compared each essay with its chart, and found that almost every TA 5 essay shared the same problems: an overview that was too vague or wrong, a missing step or change, or a number read from the wrong row. These aren’t problems of English ability; they come from not checking item by item after writing. I’ve written this up separately in the Band Score Samples series, where I compare my original band 6, 6.5 and 7 essays paragraph by paragraph.

By question type: writing the same type in a row makes the change visible

Splitting the 50 essays by question type also reveals some patterns:

Outweigh is the best example: by the third and fourth essay of the same type, the things I had lost marks for earlier were fixed, and the score came up. On the other hand, all seven Agree/Disagree essays scored 7, which meant that type was already “good enough” for me. To go higher, I needed deeper argument, not more essays of the same kind. You can’t make that kind of judgement without a score record for every essay.