ProjectionLab vs Boldin

Two free tiers, fenced in opposite places

Both products build the same kind of plan: one household, simulated a year at a time to a longevity age, with accounts, income, pensions, Social Security and expenses feeding one cash-flow model. The paid tiers cost about the same. The free tiers do not give away the same half.

• ProjectionLab Basic keeps the probability and withholds the tax. Monte Carlo and historical backtesting are both on the free plan. Tax estimation is not, and neither are the cash-flow view, withdrawal strategies, Roth conversions or a second plan.
• Boldin Basic keeps the plan and withholds the probability. It takes 100-plus inputs, runs what-if scenarios and draws the charts. Monte Carlo, state and federal tax projections, account linking and export are all PlannerPlus features.

So neither free tier computes a tax bill, and only one of them answers "would this last". A reader who wants to see a chance-of-success number before paying anything can see one on ProjectionLab and cannot on Boldin. A reader who wants to rough out a plan with many inputs and look at it can do that on either.

Two success rates that are not the same number

Both products print a percentage and call it a chance of success. Type the same household into both and the two numbers will differ with no input wrong, because they come from different generators.

• Boldin runs 1,000 paths. Each account carries one blended rate of return, and returns are drawn from a normal distribution around it, with a standard deviation Boldin derives from history. Since July 2025 every account in a path moves in step. The score is the share of paths that end the plan at $0 or more, and Boldin's help centre says a path that runs short mid-plan and is rescued by a later inflow, such as a property sale, still counts as a success.
• ProjectionLab lets you pick the generator. It can replay historical S&P 500 returns and dividend yields with historical US inflation, sequentially, from random start years or in blocks of consecutive years, or draw from normal distributions whose parameters you set. Trials go up to 2,000 where the method allows more. The score is the share of trials that reach the end of the plan without running out, and the bands around it, such as "almost failed", have thresholds you choose.

The difference matters most for what retirement plans actually fail on: a bad stretch in the first years of withdrawals. A historical replay contains the real sequences in the order they happened, so the plan meets the specific decades that broke plans before. A normal distribution draws each period independently, so a bad decade appears only when independent bad draws happen to line up, and how bad it gets is set by the mean and volatility you assumed rather than by anything that happened.

The nearest like-for-like is ProjectionLab's own normal-distribution mode, set to the return and volatility you gave Boldin. Put a historical-mode ProjectionLab score next to a Boldin score and you are comparing two models, not two plans. Boldin has also changed its generator before, which the Monte Carlo glossary entry records, so a Boldin score from last year is not comparable with this year's either.

What each one leaves out

Anything outside the US, on Boldin. Its tax engine, Social Security explorer and Medicare costs are US machinery with no international mode. ProjectionLab has account types and tax presets for Canada, the UK, Australia, Germany and the Netherlands, on Premium. They are thinner than its US modelling, which carries ACA subsidies, IRMAA, 72(t) and Roth conversions, but they exist. For a household that is not American, the choice is made here.

Live balances, on ProjectionLab. It links no bank or brokerage account, by design: every balance is typed, or synced by a third-party browser extension from a budgeting app. Boldin's PlannerPlus links accounts through Meld, which routes each institution to Plaid, MX or Finicity. That import is balances only. Boldin says it does not capture cost basis, allocations or transactions, so the holdings are typed on both.

Neither product prices a security or carries market data. Both model money at the level of an account and an assumed return, which is why this is a planning decision and not a data one.

The price, and the way out

ProjectionLab Premium is $129 a year and Boldin PlannerPlus is $144 a year. Both are sold annually only. ProjectionLab's month-to-month Premium plan and its lifetime plan are both retired, and Boldin has no monthly option. The trials differ: seven days on ProjectionLab, fourteen on Boldin, and Boldin's policy is no prorated refund once the trial ends. A $15 gap on an annual bill does not decide this.

What happens when you stop paying is more useful to know before you start:

• ProjectionLab keeps the account and drops it to Basic. The cash-flow view, the tax engine and every plan after the first stop being available. The whole account exports to JSON, and that file restores into the app.
• Boldin documents how to cancel but not what becomes of the plan afterwards. Printing a report and downloading the annual values behind it are PlannerPlus features, so take that export before the renewal date, not after it.

Above the software, Boldin sells people: a $250 coaching session on your own plan, and fee-only CFP advice through Boldin Advisors at $3,200 flat for a one-off review or from $1,200 a year ongoing. ProjectionLab's paid tier above Premium is Pro, at $549 a year, which is software for advisers and coaches running client plans. This catalogue does not evaluate either service.

Which one, and when

Outside the US: ProjectionLab Premium. Boldin has no mode for you.

You want to see a plan meet real historical sequences, or see any probability before paying: ProjectionLab. Its free plan runs both historical and Monte Carlo trials. Move to Premium when you need the tax engine, and not before.

You want balances to refresh themselves, or a human planner on the same bill: Boldin PlannerPlus. It is the only one of the two that links accounts, and the only one that sells advice beside the software. Check that your institutions connect during the fourteen-day trial, because the refund policy ends with it.

Otherwise, for a US household that is content to type its balances: ProjectionLab. It costs less, its free tier includes the probability, and it lets you choose how that probability is generated. The two things only Boldin offers here are linked balances and people; if you want neither, the extra $15 a year buys one return generator in place of a choice of several.

Do not use one to check the other. Two chance-of-success numbers from these products differ because of how each generates returns, not because one of them found a problem with your plan. Other planners and the free historical calculators are on the retirement planning page.