Note-taking during certification study is widely misunderstood. Most candidates treat it as transcription — copying information from a video, slide, or book into a notebook. The problem is that transcription is passive. Your hand is moving and words are appearing, but no active cognitive work is happening. You are not processing, connecting, or encoding the information in a way that supports later retrieval. You are producing a slightly condensed version of the original source, which you then use as source material for re-reading — another passive activity.
The Cornell note-taking system, developed by Walter Pauk at Cornell University in the 1950s, was designed to convert passive transcription into active note creation. It structures the page in a way that builds retrieval practice directly into the note-taking format and forces synthesis as a final step. For IT certification candidates working through dense technical content, it addresses the specific problem of information that is easy to write down and hard to remember.
"The most important part of the Cornell system is not the page layout — it is the discipline of generating questions after every session. That generation step is active recall built into the note-taking process itself. Students who skip it are producing a nicer-looking version of conventional notes, not a fundamentally different study tool." — Walter Pauk, How to Study in College, Cengage Learning, 10th edition
The Cornell page layout
Cornell note-taking divides a standard page into three sections:
+------------------+-------------------------------------+
| | |
| CUE COLUMN | NOTES COLUMN |
| (2.5 inches) | (6 inches) |
| | |
| Written AFTER | Written DURING |
| note-taking | the learning session |
| | |
+------------------+-------------------------------------+
| |
| SUMMARY |
| (2-3 inches at bottom of page) |
| Written AFTER the full session |
| |
+--------------------------------------------------------+
The three sections serve distinct functions:
Notes column (right, 2/3 of page): Taken during the learning session — while watching a video, reading a chapter, or attending a lecture. Contains abbreviated notes, diagrams, protocols, tables, and key facts. Not full sentences; dense and efficient.
Cue column (left, 1/3 of page): Filled in after the learning session. Contains questions, keywords, and prompts that test retrieval of the notes column content. This is the active recall mechanism built into the format.
Summary section (bottom): Written after reviewing the page, in your own words. Two to four sentences that capture the most important concepts and relationships from the page. This is the synthesis step.
Taking notes in the notes column
During the learning session, the notes column should capture the core content efficiently without transcribing everything. The goal is to record enough to reconstruct understanding, not enough to recreate the source.
For a video on AWS EC2 instance types, an effective notes column entry might look like:
EC2 Instance Types
- Named format: [family][gen][size] e.g. m5.large
- General purpose (M, T): balanced compute/memory/network
- T series: burstable, CPU credits, good for variable workloads
- Compute optimized (C): high-performance processors, batch jobs, ML inference
- Memory optimized (R, X, z): in-memory DBs, real-time processing
- R = memory intensive apps
- X = highest mem:vCPU ratio
- z = high single-thread performance
- Storage optimized (I, D): I/O intensive, DW, Hadoop
- Accelerated computing (P, G, F): GPU/FPGA, ML training, graphics
- Nitro system = underlying platform, used by most modern types
- Key metric: vCPU:memory ratio varies by family
This is fast to write during the session, captures the structure and key distinctions, and includes enough specificity to be testable.
For Cisco CCNA, a notes column for an OSPF session might include:
OSPF Overview
- Link-state routing protocol, SPF algorithm
- Metric = cost (100/bandwidth in Mbps by default)
- FastEth = 1, GigEth = 1 (need to reference-bw adjust)
- Admin distance = 110
- Multicast groups: 224.0.0.5 (all OSPF), 224.0.0.6 (DR/BDR)
- Hello/Dead timers: 10s/40s (p2p+bcast), 30s/120s (NBMA)
- Neighbor requirements: same subnet, same area, same hello/dead, same MTU
AUTH if configured
- OSPF states: Down > Init > 2-Way > Exstart > Exchange > Loading > Full
- DR/BDR only on multi-access networks (Eth), not p2p
DR/BDR election: highest priority (default 1), then highest RID
Building the cue column: the active recall engine
After the learning session — not during it — go back through each page and write cue questions in the left column that test the content on the right. These are retrieval prompts, not summaries.
For the EC2 instance types page, effective cue column questions:
- What does the Nitro system refer to?
- Which instance family offers the highest memory-to-vCPU ratio?
- Why would T-series instances fail under sustained CPU-heavy workloads?
- What instance families would you use for ML training vs ML inference?
- How does the naming convention work? Give an example.
For the OSPF page:
- What are the two OSPF multicast addresses and when is each used?
- List all OSPF neighbor requirements.
- What is the default hello/dead timer on a broadcast network?
- Why is DR/BDR election not needed on a point-to-point link?
- What determines who becomes the DR if all priorities are equal?
The cue column turns your notebook into a self-quiz tool. To use it for retrieval practice: cover the notes column, read each cue question, answer from memory, then uncover the notes to check. This is significantly more effective than re-reading the notes column.
Writing the summary section
The summary at the bottom of each page should be written in complete sentences and should synthesize — not list — the key ideas. The synthesis requirement is the cognitive work that produces encoding.
A poor summary (listing, not synthesizing): "EC2 has different instance types. There are general purpose, compute optimized, memory optimized, storage optimized, and accelerated computing."
A good summary (synthesizing relationships): "EC2 instance selection is driven by the bottleneck resource for your workload — use M-series when no specific bottleneck exists, C-series when CPU is the constraint, R or X-series when memory is the constraint, and I-series when storage I/O dominates. T-series is the exception — it trades consistent compute for lower cost using a credit model suited only to variable, spiky workloads."
The second summary is usable for exam preparation. It contains the decision logic that scenario questions test. The first summary is just a category list.
Using Cornell notes for spaced review
Cornell notes are designed for review, not just creation. The review process:
- Cover the notes column with a blank sheet of paper or your hand.
- Read each cue question in the left column.
- Answer from memory — spoken aloud, written on the blank paper, or just recalled mentally.
- After working through all cues on a page, uncover the notes and check your accuracy.
- Mark any cues where your answer was wrong, incomplete, or uncertain.
- Read the summary at the bottom.
During the final review period before exam day, this process takes approximately three to four minutes per page of Cornell notes. It covers the content through active retrieval rather than passive reading, and the cue questions you marked as uncertain tell you precisely where to spend additional study time.
For a candidate who has taken Cornell notes throughout a twelve-week study campaign and has forty to fifty pages of notes across all exam domains, the final review is highly efficient: one pass through all pages using cue-column retrieval, followed by targeted re-study of marked gaps.
Adapting Cornell notes for different content types
Video-based courses (A Cloud Guru, CBT Nuggets, Udemy)
Pause frequently — every five minutes at minimum — to write notes rather than running the video continuously. Do not write while watching; pause, recall, write, resume. This pause-and-recall approach adds minimal time to the session but converts passive watching into active encoding.
Documentation-based study (AWS docs, Cisco configuration guides, CompTIA objectives)
The notes column can include short code snippets, CLI commands, or configuration examples alongside conceptual notes. For CCNA IOS commands:
Basic OSPF config (IOS)
router ospf [process-id]
network [addr] [wildcard] area [area-id]
passive-interface [int] -- stops hellos on that int
ip ospf priority [0-255] -- 0 = never DR/BDR
auto-cost reference-bandwidth [Mbps]
The cue column then asks: "What command prevents OSPF from sending hellos on an interface without removing the network statement?" The answer must come from memory.
Practice question review
After a practice exam session, use Cornell format to document each missed question. Notes column: the concept behind the correct answer, the distractor that fooled you, and why. Cue column: a question that tests the same concept. Summary: the principle or rule you need to apply differently next time.
This transforms practice exam reviews from passive "read the explanation, nod, move on" into active study sessions that produce lasting correction.
See also: Active Recall vs Passive Review: Why Re-Reading Your Notes Fails
References
- Pauk, W., & Owens, R. J. Q. (2010). How to Study in College (11th ed.). Cengage Learning. ISBN: 978-1439084465.
- Brown, P. C., Roediger, H. L., & McDaniel, M. A. (2014). Make It Stick: The Science of Successful Learning. Belknap Press. ISBN: 978-0674729018.
- Mueller, P. A., & Oppenheimer, D. M. (2014). The pen is mightier than the keyboard: Advantages of longhand over laptop note taking. Psychological Science, 25(6), 1159-1168.
- Karpicke, J. D., & Roediger, H. L. (2008). The critical importance of retrieval for learning. Science, 319(5865), 966-968.
- Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students' learning with effective learning techniques. Psychological Science in the Public Interest, 14(1), 4-58.
- Ambrose, S. A., Bridges, M. W., DiPietro, M., Lovett, M. C., & Norman, M. K. (2010). How Learning Works: Seven Research-Based Principles for Smart Teaching. Jossey-Bass. ISBN: 978-0470484104.
- Weinstein, Y., Sumeracki, M., & Caviglioli, O. (2018). Understanding How We Learn: A Visual Guide. Routledge. ISBN: 978-1138561724.
Frequently Asked Questions
What makes Cornell note-taking more effective than regular notes?
Regular note-taking is typically transcription — copying information from a source. Cornell notes are structured for retrieval: the cue column forces you to generate questions that test the notes column content, and the summary forces synthesis rather than listing. The result is a notebook that functions as a self-quiz tool rather than a reference document, which means every review session is retrieval practice rather than passive re-reading.
When should I fill in the cue column — during or after the session?
After the session, not during it. During the session, focus on the notes column. Immediately after completing a page or finishing a study block, go back and write cue questions in the left column. Writing cues after the session requires you to think about what each note section is testing, which is itself a retrieval and synthesis activity. Writing them during the session would interrupt the flow of note-taking and reduce note quality.
How do I use Cornell notes for review before an exam?
Cover the notes column with a blank sheet of paper. Read each cue question and answer from memory. After working through all cues on a page, uncover the notes and check accuracy. Mark any cues where your answer was wrong or uncertain. Read the summary section. Repeat for all pages. This process takes three to four minutes per page and produces active retrieval rather than passive recognition. Mark uncertain cues for targeted follow-up study.
How do I adapt Cornell notes for video courses like A Cloud Guru or CBT Nuggets?
Pause the video every five minutes and write notes from what you just watched before resuming. Do not write while watching — pause, recall, write, then resume. This pause-and-recall approach converts passive watching into active encoding. The additional time cost is minimal (perhaps 15-20% longer per session) but the retention difference is substantial compared to watching through without structured note-taking.
Should I take Cornell notes on paper or use a digital tool?
Research by Mueller and Oppenheimer (2014) found that handwriting notes produces better retention than typing for conceptual content, because the slower writing speed forces prioritization and paraphrasing rather than verbatim transcription. Paper Cornell notes are recommended for initial learning. A digital version (OneNote, Notion, or a Cornell template in your preferred app) can supplement for content with code snippets or commands that are cumbersome to write by hand.