Every year, thousands of people prepare extensively for investment banking interviews — studying valuation methodologies, practising LBO models, and memorising deal structures. They get the job. And then they spend the first three months of their careers doing something that looks almost nothing like what they prepared for.
The gap between the interview version of investment banking and the daily operational reality is not small. It's not a rounding error. It's a structural mismatch between what the role is marketed as and what it actually requires for approximately 80% of a first-year analyst's waking hours.
This article is the honest version. Not the version designed to attract recruits — the version that describes the work as it actually exists, hour by hour, task by task, and why the parts nobody talks about are the parts that determine who advances and who stagnates.
The Version You Were Sold vs. The Version You Get
The job description reads: financial modelling, valuation analysis, pitch book preparation, capital markets advisory, live deal execution.
The reality of the first year reads: formatting, data verification, more formatting, comps maintenance, more data verification, and then — if you've earned it through the previous four items — some actual financial modelling.
This is not cynicism. It's a structural reality that exists for a specific reason: the analytical outputs investment banks deliver to clients have no room for error. A number in a board presentation, a valuation range in a fairness opinion, a multiple in a pitch book — these are relied upon for billion-dollar decisions. The quality control infrastructure that ensures those outputs are accurate runs through the analyst.
Before an analyst gets to build a complex model, they need to demonstrate they can produce a comps table with zero errors. Before they contribute analytical judgment, they need to demonstrate their formatting is indistinguishable from the bank's template. Before they get interesting work, they need to make boring work perfect.
This is not gatekeeping. It's quality infrastructure.
What the time actually looks like:
- Pitch book production and formatting: ~35%
- Data pull and verification: ~20%
- Comps maintenance and updates: ~18%
- Financial modelling: ~15%
- Administrative and compliance: ~8%
- Client-facing interaction: ~4%
The 35% of time spent on pitch book production is not wasted time — it's the quality infrastructure that makes everything else reliable. Analysts who understand this invest in it differently than those who see it as beneath them.
The Actual Workflow: A Realistic Week
Rather than describing abstract categories, here's what an average week in a coverage group looks like for a first-year analyst on two active deal teams.
Monday
7:30 AM: Markets open. You pull equity prices for every company in your comps tables, update the market data, recheck trading multiples. This takes 45 minutes if you know the system. This takes two hours if you don't.
9:00 AM: Associate asks for an updated CIM section — the management discussion narrative needs to incorporate new financial figures from a revised model. You update the text, reformat it to match the CIM template, and fact-check every number you've inserted against the model.
12:00 PM: VP requests a precedent transactions screen for a new pitch. You search Dealogic and FactSet for relevant transactions, filter by sector and deal size, pull key financial metrics, and begin building the transaction comparison table.
4:00 PM: Pitch book review meeting. Comments come back. Twelve slides need revisions, four of which involve substantive data changes and eight of which are formatting corrections.
7:00 PM: Begin working through the revision list. The four data changes require going back to source documents and rerunning calculations. Each one takes longer than expected.
11:30 PM: Revisions complete. Send the updated file.
Tuesday through Thursday follow similar patterns. The content changes — different deals, different deliverables — but the structure remains: morning data maintenance, midday production tasks, afternoon review cycle, evening revisions.
Friday afternoon: If the week has gone well and no deals have required emergency work, there may be two to three hours of time that isn't consumed by immediate deliverables. This is when the best analysts work on developing process efficiency, organising their files, and thinking about the next week's likely needs.
The week has contained approximately six hours of work that looks like the job description. It has contained approximately 55 hours of work that looks like quality-controlled data management and precision document production.
Pitch Book Production: What It Actually Involves
Pitch books are the primary deliverable in investment banking. A pitch book is a presentation deck — typically 40 to 90 slides — that summarises a bank's analytical work, strategic recommendation, and credentials for a potential transaction or advisory engagement.
The anatomy of pitch book production:
Data sourcing and verification: Every financial figure in a pitch book comes from a primary source — a 10-K, an earnings release, a Bloomberg terminal, a Capital IQ database. An analyst pulling revenue and EBITDA figures for eight comparable companies is not typing numbers — they are locating primary documents, extracting the correct line items, applying the correct fiscal period alignment, and verifying that each number is the most recent available.
This is not clerical work. Fiscal year misalignment, one-time item treatment, accounting restatements, currency conversion — each of these requires judgment about what the "correct" number is. An analyst who gets this wrong doesn't just have a wrong number. They have a wrong story.
Analysis building: The financial analysis that forms the core of most pitch books — comparable company analysis, precedent transactions, DCF sensitivity, LBO returns — needs to be built, checked, and formatted correctly.
Narrative integration: The financial analysis sits inside a strategic narrative. Analysts don't write this narrative — MDs and VPs do — but analysts are responsible for ensuring that the financial data in the narrative is accurate, that the exhibits referenced in the text match the exhibits in the deck, and that no number stated in a sentence on slide 7 contradicts a number displayed in a table on slide 14.
Template compliance: Every bank has a visual template — specific fonts, specific colours, specific layouts, specific footnote formats. Deviations from the template are noticed immediately by anyone who works at the bank.
The non-obvious skill in pitch book production:
The hardest skill in pitch book production is not formatting or even data accuracy. It's internal consistency — the ability to track a number as it propagates through 60 slides and ensure that every instance of that number, every derived number, and every narrative reference to that number are all consistent.
An analyst who updates the base year revenue assumption in a model at 10 PM needs to know every place that assumption flows: into the comps table, into the football field chart, into the narrative on slide 4, into the implied valuation range on slide 22. Changing the number in the model without updating all downstream instances creates inconsistencies that are embarrassing to find in a client meeting.
Financial Modelling: What It Actually Looks Like in Practice
When analysts do financial modelling — which is a real and meaningful part of the role, just a smaller fraction than recruits expect — the work looks different from what modelling courses teach.
The gap between course modelling and production modelling:
In a modelling course, you build a model from scratch, take your time, make design decisions based on your analytical preferences, and submit when you're satisfied.
In a live deal, you inherit a model that was started by someone else, has conventions baked into it that you didn't design and may not immediately understand, has been partially updated by several people over several months, and needs to be updated tonight with specific new figures by a specific deadline.
What production model work actually requires:
Model literacy: The ability to read and understand a model that someone else built — quickly and accurately. This means understanding the logic of someone else's formula structure, tracing assumptions through multiple worksheets, and identifying where key drivers live without reading every cell.
Targeted updating: Making specific, bounded changes to a model without disturbing its structure or breaking interconnected formulas.
Error-first thinking: Before making any change to a production model, an analyst should understand what the model currently shows and why — so that after the change, they can immediately identify whether the output moved in the expected direction by approximately the expected amount. A revenue increase of 5% that causes enterprise value to decrease is a red flag before you send the model to the associate.
Model-reading literacy — the ability to understand a model someone else built, quickly and accurately — is arguably more important in production than model-building ability. Most technical preparation programmes teach building. The production skill of inheriting and updating is learned almost exclusively on the job.
The Revision Cycle: Where Most of the Night Goes
The revision cycle is the part of the analyst day that consumes the most time, produces the most stress, and is most frequently misunderstood by people who haven't experienced it.
How the revision cycle works:
An analyst produces a deliverable. It goes to the associate, who reviews it and sends back comments. The analyst addresses the comments and resubmits. The associate reviews again, may send additional comments or escalate to the VP. The VP reviews and may have their own comments. The analyst addresses those and resubmits.
In a typical pitch book production process, three to five full revision cycles is normal. Each cycle takes 30 minutes to two hours depending on the scope of changes.
What the revision cycle is actually testing:
A revision request at 10 PM tests:
- How quickly can you correctly interpret what the reviewer is asking for?
- How quickly can you make the change correctly?
- Can you identify and fix all downstream implications of the change without being told about each one?
- Can you deliver the corrected version with a clear summary of what changed?
An analyst who receives a comment "update the revenue assumption to $2.3B" and changes only that number — without updating the comps table, the football field, the narrative, and the implied multiples that all use that revenue figure — has not completed the task. They've made one of five required changes and created five inconsistencies.
The professional standard for revision responses:
When submitting a revised document, the professional standard is to include a brief summary of every change made: "Updated revenue to $2.3B on slide 5, refreshed comps table on slide 8 (EBITDA multiple moved from 8.2x to 8.4x), updated football field chart on slide 15, updated narrative reference on slide 4." This summary serves two functions: it demonstrates that you found all the downstream implications, and it allows the reviewer to verify your completeness without reading the entire document again.
What Separates the Analysts Who Advance from Those Who Don't
The analysts who advance from first-year to trusted second-year contributors are not necessarily the ones with the strongest technical backgrounds. They are the ones who develop three capabilities that are rarely discussed in recruiting materials:
1. Process efficiency. The analyst who can run a comps table update in 30 minutes instead of 90 minutes gets staffed on more interesting work. Process efficiency in the mechanical tasks is what creates capacity for the analytical ones.
2. Internal consistency awareness. The analyst who instinctively tracks a number through all its downstream instances — without being asked — stops revision cycles earlier and builds a reputation for thoroughness that compounds over time.
3. Turnaround speed under pressure. At 11 PM when a VP needs a revised model output in 45 minutes, the analyst who can execute quickly and accurately without errors is the one who gets called first for the next live deal. Speed and accuracy under time pressure is the most observable signal of analyst capability in the first year.
None of these three capabilities are taught in interview preparation. All three are learnable before joining — but only if you know to develop them.
Closing: Understanding the First Year Changes How You Prepare for It
The structural reality of an investment banking analyst's day — predominantly pitch book production, data management, and revision cycles, with financial modelling as the meaningful but smaller fraction — is not a flaw in the system. It's the quality infrastructure that makes billion-dollar financial decisions trustworthy.
Understanding this before you join changes how you prepare. It suggests spending time developing Excel precision, building document consistency habits, and learning to trace numbers through complex multi-page documents — skills that rarely appear in modelling courses but that dominate the first-year experience.
At Meritshot, the Investment Banking programme covers not just the valuation mechanics that interviewers test but the operational skills that determine first-year effectiveness — pitch book standards, model documentation, revision cycle protocols, and the process efficiency techniques that experienced analysts develop over their first two years. Students who understand what the job actually requires before they arrive are more useful on day one than those who arrive with strong interview preparation but no exposure to production realities.
Explore the Meritshot Investment Banking Programme →
This article was written by the Meritshot content team. Meritshot trains professionals in Data Science, AI Engineering, Full Stack Development, Investment Banking, and Cyber Security through hands-on, practitioner-led programmes.
