Ten months in. Forty open tabs. Still “doing the literature review.”
If that’s you, I want to say something that might be a relief: this usually isn’t a discipline problem. It’s a sequencing problem, and it’s fixable without reading a single new article.
The Sequence Almost Everyone Uses (and Why It Fails)
Most students approach the literature review in this order: read broadly, keep everything that seems relevant, and figure out the structure once “enough” has been read. The trouble is that “enough” never arrives, because without a structure to sort new reading into, every article looks equally essential. There’s no basis for judging an article as peripheral, so nothing gets set aside, and the pile just grows.
This isn’t a comment on your reading speed or your discipline. It’s what happens, predictably, to anyone who reads before they’ve built somewhere to put what they’ve read.
The Fix: Build the Skeleton First, Then Read Against It
The alternative sequence is to draft a rough, provisional structure – even one you expect to be partly wrong – before doing extensive further reading. This might be three tentative themes, or a rough timeline of how thinking in the field has shifted, or a simple for/against split on a contested question in your area.
Once that skeleton exists, every new article gets a specific job: does it support theme one, complicate theme two, or sit outside the structure altogether? An article that doesn’t fit anywhere can be set aside – not because it’s bad, but because it isn’t yours to include, and you now have a basis for saying so.
This sounds almost too simple to matter, but in practice it’s the single biggest difference between literature reviews that resolve in six weeks and ones that drag on for ten months. The skeleton doesn’t need to be right. It needs to exist, so your reading has direction instead of just volume.
A Worked Example
One MBA student was reading around “employee engagement in remote teams” with no end in sight. We spent forty minutes drafting a rough three-theme skeleton: technology-mediated communication, trust-building at a distance, and manager visibility. Within two weeks, her reading had a job – every new article was tested against those three themes rather than added to a growing, undifferentiated pile. The skeleton itself changed twice as she read (trust-building split into two sub-themes), which is exactly what should happen. The point isn’t to get the structure right immediately – it’s to have something to argue with.
Why This Differs From “Just Summarise Less”
You’ll often hear that a literature review should “argue, not summarise” – which is true, but it’s advice about the finished product, not about how to get there. Knowing you should be arguing doesn’t tell you what to do differently at the reading stage, which is where most students actually get stuck. Building the skeleton first is the specific mechanism that produces an argument-shaped review, rather than a description of what one looks like once it’s done.
FAQ
What if my skeleton is wrong? It probably will be, partly. That’s fine – it’s meant to evolve as you read. The goal is a working structure to read against, not a final one.
How much should I read before drafting a skeleton? Enough to identify the main debates or themes in your area – often 15 to 20 key sources is enough to attempt a first draft, well before “comprehensive” reading is complete.
Does this work for a systematic review too? The logic is the same, though a systematic review’s structure is usually dictated more tightly by your protocol and inclusion criteria, set before searching begins.
Next Step
If your reading has volume but no shape yet, a short conversation can often help you find the skeleton faster than reading alone will. I offer a free 15-minute introductory call.