LinkedIn

In 2004, LinkedIn was about a year old, roughly 18 months from launch, and pre-revenue. Reid Hoffman walked into Greylock to raise a Series B. The round was reportedly around $10M, led by Greylock, at a valuation widely cited near $100M. Treat the exact figures as approximate: the deck is public, the term sheet is not.

What makes this deck unique is not that it worked. It is that Hoffman published the original slides years later and annotated them himself, candidly, marking what held up and what he would now cut. Founders almost never get a winning Series B deck plus the founder's own red pen on top of it.

The context matters. Google had just filed to go public the same month Hoffman was preparing the pitch. Friendster and MySpace were the loud names in "social." LinkedIn had to argue that a quiet professional network, with no revenue and slow-looking growth, was the contrarian bet that would compound. That is the hardest pitch there is: asking investors to fund a thesis, not a number.

It opened with the investment thesis, not the team or the product demo. Hoffman's rule: open with what prospective investors must believe in order to want to be shareholders of your company. The first slides told Greylock exactly what they had to believe to say yes. Everything after was evidence.

It pitched by analogy to outcomes investors already trusted. LinkedIn was framed as "professional people search 2.0," sitting alongside Google, eBay, and PayPal. Each of those built a network that threw off real money without charging users directly: eBay's reputation system, PayPal's fraud detection, Google's PageRank reading the link graph. Hoffman's nuance is worth stealing: pitch by analogy, but do not necessarily reason by analogy. Analogies buy comprehension fast. They are a shortcut for the listener, not a substitute for your own logic.

It made the network effect the moat, explicitly. A simple quadrant slide positioned LinkedIn against rivals, and the differentiation was not user count or data volume. It was the network itself: the thing that gets more valuable as it grows and is brutal to copy once it has scale.

It tied the round to the financing arc. Hoffman is blunt that this is the part founders underrate: the most fundamental strategy is the financing strategy, because if your company runs out of gas it will die no matter how good the product is. The rounds were sequenced as a logic chain: Series A proved you could build the network, Series B had to prove you could turn it into revenue, Series C would show the path to profit. Each raise bought the proof the next raise required.

It addressed the obvious objection head-on: no revenue. Instead of hiding it, the deck put revenue models up front and pointed to the team's track record of hitting its Series A promises. The move was to convert "they have no revenue" from a dealbreaker into "they have earned the right to be believed."

This is the rare part: the founder grading his own deck in hindsight.

Three revenue streams instead of one. The deck listed targeted ads, job listings, and subscriptions. Hoffman's verdict: it was a mistake to list three different revenue streams, when the general rule is that one business model drives the business. Showing three reads as breadth to a founder and as a lack of conviction to an investor.

The customer testimonials slide. He calls it mostly a mistake, because customer slides are more appropriate for enterprise pitches, not a consumer internet network. The proof a network needs is the network growing, not logos.

No risk factors. He now thinks naming what could kill you is essential: it signals that you have thought harder about the business than the person across the table, and it pre-empts the objection they were already forming.

Nervous qualifiers. He admits to writing "very high operating margins" and notes the tell: you reach for words like "very" exactly when you are least sure of the underlying claim. The fix is to cut the adjective and strengthen the number, or cut the claim.

He also concedes the deck ran long and wound around. Winning despite these is the point: a thesis that is clearly true survives a deck that is imperfect.

Write the belief, not the slides. Before you design anything, write the 3 to 8 sentences an investor must accept to want your equity. If a slide does not advance one of those beliefs, it is decoration.

Make the round a proof, not a number. Say plainly what this specific raise will prove, and why proving it unlocks the next raise: "We are raising X to prove Y, which is what a Series C investor will need to see." That reframes runway as a sequence of de-risking milestones instead of a burn rate.

Pick one business model to lead with. You can mention others as upside. Lead with the one that drives the company. Three coequal revenue lines read as hedging, even when you mean optionality.

Name your risks before they are asked. A short, honest risk slide does more for credibility than another hockey stick. The founder who raises the hard question first is the one trusted to handle it later.

Sell the compounding asset, not today's metrics. For a network, the defensible thing is the network: it strengthens with scale and resists copying. Tell investors which asset compounds and why a competitor with more cash still cannot catch it.

Engineer the conclusion, then get out of the way. Hoffman's sharpest line: the winning moment is when an investor concludes on their own that the thesis is worthwhile. Lay the evidence so they reach the conclusion themselves, and go hunting for the weak spots first: ask what is broken and why it will not work before the room does it for you.