systematically biased by anchoring to the last project.
Your last infrastructure project took 6 months and cost $380K. Your CEO asks you to estimate a new integration build. Before you even open a spec doc, your brain has already decided the answer is 'somewhere around $380K and 6 months' - and you'll spend the rest of your analysis unconsciously confirming that number instead of building the estimate from scratch.
Anchoring is the tendency to over-weight the first or most recent number you encounter when making estimates and decisions. Operators who don't actively counteract it systematically misprice projects, misallocate Budgets, and repeat the Cost Structure of their last initiative even when the new one has fundamentally different economics.
You already know the anchor from the prerequisite: a deliberately chosen reference price that frames how a Buyer perceives value. Anchoring is the same mechanism turned inward - except now you are the one being distorted.
When you estimate Implementation Cost, forecast Revenue, or set Hiring Targets, your brain grabs the most available reference number and adjusts away from it. The problem: the adjustment is almost always insufficient. If your team closed $2M in ARR last quarter, your forecast for next quarter will drift toward $2M whether the Pipeline Volume supports it or not.
This is not laziness. It happens to experienced Operators, CFOs, and PE operators alike. The bias is structural - your mind uses the anchor as a starting point and then makes small corrections, when it should be building the estimate from First Principles.
Anchoring directly corrupts three things that drive your P&L:
In controlled experiments, anchoring shifts numerical estimates by 20% or more from unanchored baselines - and the effect persists among experienced decision-makers. Every Budget line, every Pricing decision, every resource allocation choice is a surface where this distortion operates.
The mechanics follow a predictable pattern:
Step 1: An anchor enters your frame. This could be the cost of your last project, the Revenue number someone mentioned first in a meeting, or even an unrelated figure you saw on a dashboard five minutes ago.
Step 2: You adjust from it - insufficiently. Instead of estimating from First Principles (what does this project actually require?), your brain starts at the anchor and nudges. 'Well, this project is a bit smaller, so maybe $320K instead of $380K.'
Step 3: You rationalize the anchored estimate. You find reasons the number 'makes sense' - you match the scope to fit the Budget rather than building the Budget to fit the scope.
Why the adjustment is always too small: Moving away from an anchor requires effort and evidence. Each unit of distance from the anchor feels increasingly uncertain. So you stop adjusting before you should.
Example of the math:
You cannot eliminate Anchoring - it is hardwired. But you can build processes that reduce its damage:
1. Estimate before you look at comparables. Write down your First Principles estimate before you review what the last project cost. The act of committing to an independent number weakens the anchor's pull.
2. Use base case ranges, not point estimates. Instead of a single dollar figure, force yourself to produce three numbers: a low case, a base case, and a high case. Ranges expose when your low case suspiciously clusters near an anchor.
3. Run Sensitivity Analysis on your key assumptions. If your estimate swings by more than 30% when you change a single input, your precision is false - the anchor is doing the work, not the analysis.
4. Assign estimation to someone without the anchor. If you just finished a large project, have someone else estimate the next one independently. Compare their number to yours. The gap is your Anchoring bias, measured in dollars.
5. Use decision trees with explicit Expected Value calculations. Forcing yourself to assign probabilities and payoffs to branches makes the math visible. It is harder for an anchor to survive when you have to justify each number in a structured framework.
You just staffed a 12-person engineering team for a platform rebuild over 8 months. Total Labor cost per engineer (salary, benefits, and overhead): $15K/month. Project total: $1.44M ($15K x 12 engineers x 8 months). Now the business needs a data pipeline integration - genuinely a 3-person, 4-month job. True Implementation Cost: $180K.
Your brain anchors to the last project: 12 people, 8 months, $1.44M.
You 'adjust down' - 'This is smaller, maybe 7 people for 5 months.' Estimate: $525K.
You build a Budget around $525K. You staff 7 engineers. Four of them have nothing meaningful to do for weeks.
The actual work completes in under 4 months with the 3 productive engineers. You spent $525K for $180K of value - $345K in wasted Labor and opportunity cost.
If you had estimated from First Principles first (what tasks exist, how many people per task, how long each takes), you would have landed near $180K before ever thinking about the prior project.
Insight: Anchoring does not just produce a wrong number - it produces a wrong team size and wrong timeline, which compounds the waste. The $345K was not overhead - it was misallocated capital that could have funded other Operating Investments.
Your current analytics vendor charges $120K/year. Contract is up for renewal. You've identified two alternative vendors at $45K/year and $70K/year with equivalent capability. Your Outside Option is $45K.
The vendor opens renewal at $135K ('small increase for new features'). The $135K is now the anchor.
You negotiate 'hard' and get them down to $105K. You feel like you saved $30K.
But your Outside Option was $45K. The true surplus you should have captured was $90K ($135K minus $45K). You only captured $30K of it.
The anchor ($135K) made $105K feel like a win. It was actually $60K of Value Leakage.
Counterplay: Before the call, write down 'I will not pay more than $55K for this capability - that is my $45K Outside Option plus a 22% premium for the effort and disruption of migrating to a new vendor.' Enter the negotiation with your own anchor, not theirs.
Insight: In Vendor Negotiations, the party who sets the first number captures more surplus. From the prerequisite, you know to set the anchor for Buyers. This example shows the mirror: when you are buying, the vendor is setting the anchor on you. Prepare your own reference number before you hear theirs.
Q1 closed $2.8M in new ARR against a $2.0M Revenue target (140% of target). You're now forecasting Q2. True Q2 Pipeline Volume is weaker - new segment, longer sales cycles. Realistic Expected Value of the pipeline is $1.5M.
Your Q1 anchor: $2.8M closed. Your brain starts Q2 planning from this number.
You 'adjust down slightly' - 'Q2 might be a bit softer, call it $2.3M.'
You staff for $2.3M: you keep the expanded GTM Teams, maintain Marketing Spend at Q1 levels, and set Revenue targets accordingly.
Q2 closes at $1.5M - just 65% of your anchored forecast. Pipeline Volume was never there to support $2.3M.
You overspent by roughly $160K in Marketing Spend and carried $90K in excess Commissions capacity you did not need.
First Principles approach: Count qualified Pipeline Volume, multiply by historical Close Rate for this segment (not last segment), apply Expected Value. This yields roughly $1.5M before Q1's number enters the picture.
Insight: Revenue forecasts are especially vulnerable to Anchoring because the most recent quarter is emotionally vivid. The fix is mechanical: build the forecast from Pipeline Volume and Close Rate data for the current segment, not from the topline outcome of a different quarter.
Anchoring is the anchor concept turned against you - when you set anchors for Buyers you capture surplus; when you fail to notice anchors set on you (including by your own past experience), you lose it.
The adjustment from an anchor is almost always too small. If your estimate for a new project is suspiciously close to the last project's cost, you are probably anchored, not analytical.
The operational fix is process, not willpower: estimate before looking at comparables, use ranges instead of point estimates, and run Sensitivity Analysis to test whether your precision is real or borrowed from an anchor.
Confusing 'experience' with 'anchoring.' Operators say 'I'm using my experience from the last project' when they're actually anchoring. Experience means extracting transferable lessons (team velocity, risk patterns). Anchoring means unconsciously copying the dollar figure or timeline without adjusting for different scope, Cost Structure, or Demand conditions.
Only guarding against external anchors while ignoring self-generated ones. You might train yourself to resist a vendor's opening bid, but then anchor your own Capital Budgeting to whatever number you estimated first - even if new information should change it. The most dangerous anchors are the ones you set on yourself.
You managed a $600K website redesign last year (8 months, 5 contractors). Your team now needs to build an internal dashboard. You estimate $480K over 6 months. A junior engineer on your team - who did not work on the website project - independently estimates $110K over 2 months with 2 engineers. Walk through: (a) why the gap likely exists, (b) how you would determine which estimate is closer to truth, and (c) what this costs your P&L if you go with the anchored number.
Hint: The junior engineer has no anchor from the prior project. Think about what your $480K estimate would look like if you had never done the $600K project. Then calculate the opportunity cost of the $370K gap at your Hurdle Rate, remembering the project period is 6 months - not a full year.
(a) The gap exists because your $480K is an insufficient adjustment from your $600K anchor. You 'adjusted down' 20% but the true scope reduction is roughly 80%. The junior engineer estimated from First Principles - actual tasks, actual hours - without a prior anchor distorting the starting point. (b) Resolve it by decomposing the work: list every deliverable, estimate hours per deliverable, multiply by rate. If this bottom-up number lands near $110K, your $480K was anchored. If it lands near $480K, the junior underestimated scope. (c) If you Budget $480K for $110K of real work, you over-allocate $370K. At a Hurdle Rate of 15% annual, that $370K committed for 6 months has an opportunity cost of roughly $28K ($370K x 7.5%) in forgone returns - real capital you destroyed by not deploying it elsewhere.
A SaaS vendor quotes you $200K/year for a data platform. You know comparable tools cost $60-90K/year. Design a negotiation strategy that neutralizes their anchor. Include: (a) the counter-anchor you would set, (b) the Outside Option you would reference, and (c) the walk-away price.
Hint: Remember from the anchor prerequisite: the first number frames everything. Your job is to replace their $200K anchor with your own before any substantive discussion happens. Think about what number you say first and why - and note that the midpoint of $60-90K is $75K.
(a) Open with: 'We have budgeted $65K for this capability based on what comparable platforms cost in this segment.' This sets $65K as the new anchor - deliberately below the $75K midpoint of your $60-90K comparable range, leaving you room to negotiate upward while pulling the reference point far from their $200K. (b) Reference the Outside Option explicitly: 'We have two alternatives at $60K and $90K that cover our requirements. We prefer your platform for [specific reason], but not at a 2-3x premium.' (c) Walk-away price: $95K. This is the top of the comparable range plus roughly a 5% premium for the effort and risk of migrating to a new vendor. Any price above $95K means you are paying for the privilege of being anchored. Expected outcome: you settle around $85-100K, capturing $100-115K in surplus versus their opening ask. Compare this to the typical outcome of negotiating from their anchor - landing at $160K and capturing only $40K of the available surplus.
Anchoring is the defensive counterpart to the anchor concept you already learned - the same mechanism, but now working against you instead of for you. The counteraction tools you practiced here - First Principles estimation, Sensitivity Analysis, base case ranges, decision trees with Expected Value - are not just Anchoring remedies. They are the core disciplines of Capital Budgeting. In the Capital Budgeting lessons ahead, every project you evaluate with NPV, IRR, or Payback Period begins with input estimates. Anchoring is the failure mode that corrupts those inputs before the math even starts. Learning to detect it here is prerequisite to trusting any number you feed into a valuation model later.
Disclaimer: This content is for educational and informational purposes only and does not constitute financial, investment, tax, or legal advice. It is not a recommendation to buy, sell, or hold any security or financial product. You should consult a qualified financial advisor, tax professional, or attorney before making financial decisions. Past performance is not indicative of future results. The author is not a registered investment advisor, broker-dealer, or financial planner.