Why you should size traffic from the revenue goal, not the other way round
Traffic targets set on their own are unfalsifiable. "Grow sessions 30%" cannot be judged until months later, and it can be met while revenue stays flat if the extra sessions arrive from a source that does not buy. Reversing the funnel fixes that: you start from the number the business committed to, and every intermediate figure becomes a checkpoint you can measure weekly.
The reversal has three divisions. Revenue divided by average order value gives orders. Orders divided by conversion rate gives sessions. Sessions divided by days gives a daily rate. Nothing else is involved, which is exactly why the exercise is worth doing before a quarter starts rather than after it ends — the arithmetic takes a minute and it frequently shows that the committed number requires a traffic increase nobody has a plan to deliver.
The most valuable output is the last one: the gap against your current traffic, expressed as a percentage increase. A 20% gap is a channel-optimisation problem. A 300% gap is a strategy problem, and no amount of on-page work will close it — you will need a new channel, a materially different conversion rate or a higher order value. Seeing which of the three you are in changes what you spend the quarter doing.
Note that the model treats the conversion rate and order value as constants while traffic grows. They are not. Incremental traffic is usually colder than the traffic you already have, so the conversion rate tends to fall as volume rises. That makes this calculation a floor rather than a forecast, and the sensitivity table on this page exists so you can price the pessimistic case as well as the planned one.
Each division, and the trap inside it
Orders = revenue ÷ average order value. Use the order value that will apply during the period, not the trailing one, if you are about to change price or promotional depth. A 15% sitewide discount does not merely reduce order value by 15% — it also raises the conversion rate, and the two effects work in opposite directions on the traffic requirement. Model both or model neither. The average order value calculator is the right place to work out the figure and its drivers.
Sessions = orders ÷ conversion rate. Here the trap is the denominator of the rate itself. A conversion rate measured per session and one measured per user differ by the number of sessions each user makes, which on a considered purchase can be three or four. Feeding a user-based rate into this formula returns a user count while you read it as a session count, and the plan is then wrong by that same multiple. Confirm which basis your analytics tool reports before you use the number — this is the single most common way the calculation goes wrong.
Daily sessions = sessions ÷ days. A flat daily figure is a tracking device, not a forecast. Traffic is seasonal at the week and month level for almost every business, so compare cumulative sessions to date against a cumulative target rather than judging any single day against the average.
Leads = orders ÷ lead-to-customer rate. Include this stage only if you actually run one, and be careful that the rate is measured from the same lead definition your sales team uses. A marketing-qualified lead and a sales-accepted lead convert at very different rates, and using the wrong one moves the lead requirement by a large factor. The sales funnel conversion calculator handles multi-stage funnels where several rates compound.
Worked example: a $100,000 month
Your store must produce $100,000 next month. Average order value is $200, the site converts at 2% of sessions, the month has 30 days, and you currently receive 20,000 sessions a month. You also run a lead stage that closes 25% of leads.
- Orders. $100,000 ÷ $200 = 500 orders.
- Sessions. 500 ÷ 0.02 = 25,000 sessions.
- Daily. 25,000 ÷ 30 = 833 sessions a day.
- Leads. 500 ÷ 0.25 = 2,000 leads.
- Revenue your current traffic supports. 20,000 × 0.02 × $200 = $80,000.
- Gap. 25,000 − 20,000 = 5,000 sessions, which is 5,000 ÷ 20,000 = a 25% traffic increase.
Step 6 sets up the real decision, because a 25% traffic increase is not the only route to the same revenue. Holding traffic at 20,000, you would need a conversion rate of 500 ÷ 20,000 = 2.5%, a rise of half a point, or 2.5 ÷ 2 − 1 = 25% in relative terms. Or, holding both traffic and rate, you would need an order value of $100,000 ÷ (20,000 × 0.02) = $250, also 25% higher. All three levers need the same 25% relative move because revenue is their product — which means you should pick whichever is cheapest to move, not whichever is most familiar.
Reading the gap and choosing a lever
Convert the gap into a percentage of current traffic before you react to it, because the absolute number is misleading at both extremes. Needing 5,000 extra sessions is trivial on a site doing 500,000 and impossible next month on a site doing 2,000.
When the gap is modest, conversion work is usually the cheaper lever, for a structural reason: a conversion improvement applies to the traffic you already pay for, whereas a traffic increase costs new money for every incremental session. A half-point rise from 2.0% to 2.5% on 20,000 existing sessions produces the same 100 extra orders as 5,000 new sessions at the old rate, and the sessions have to be bought again every month. Use the conversion rate calculator to size the change and the A/B test sample size calculator to check whether you have enough traffic to detect it — that second check is what stops a small site from planning a testing programme it cannot statistically run.
When the gap is large, the honest answer is usually a mix. Multiply the levers rather than adding them: a 10% traffic gain, a 10% conversion gain and a 5% order value gain together give 1.10 × 1.10 × 1.05 = 1.27, a 27% revenue increase, not 25%. This compounding is why three modest, achievable improvements routinely beat one heroic one.
Finally, always price the traffic. If the gap will be closed with paid media, take the required sessions into the ad budget forecast calculator and check that the resulting cost per acquisition still sits below your order value times gross margin. Traffic that costs more than the margin it produces closes the gap on the traffic chart and opens a larger one on the profit line.
Sessions needed per $10,000 of revenue
| Average order value | Orders needed | At 1% conversion | At 2% conversion | At 3% conversion | At 5% conversion |
|---|---|---|---|---|---|
| $25 | 400 | 40,000 | 20,000 | 13,333 | 8,000 |
| $50 | 200 | 20,000 | 10,000 | 6,667 | 4,000 |
| $100 | 100 | 10,000 | 5,000 | 3,333 | 2,000 |
| $250 | 40 | 4,000 | 2,000 | 1,333 | 800 |
| $500 | 20 | 2,000 | 1,000 | 667 | 400 |
| $1,000 | 10 | 1,000 | 500 | 333 | 200 |
Session counts are rounded to whole sessions. For a $50,000 goal, multiply any cell by five.
What this model does not account for
- Conversion rate decay as traffic scales. Incremental sessions come from broader targeting and colder audiences, so the marginal rate is below the average rate. Treat the answer as a floor.
- Session versus user denominators. If your analytics conversion rate is per user, the output is a user count. Mixing the two misstates the requirement by the sessions-per-user ratio.
- Returning customers and repeat orders. The model prices every order as a new acquisition. If a meaningful share of revenue comes from existing customers, subtract that base before setting the traffic goal.
- Lag between visit and purchase. On a considered purchase, this month's sessions produce next month's orders. The totals are right; the timing is not.
- Seasonality within the period. A flat daily target will look badly missed on quiet weekdays and comfortably beaten at weekends. Track cumulative against cumulative.
- Discounting effects. A promotion raises conversion and lowers order value simultaneously. Changing only one of the two inputs will always flatter the plan.
Where this fits with the rest of your planning
This calculator sizes demand. Two neighbouring questions need different tools. If you already know the traffic is coming from paid media and want to know what it costs, the ad budget forecast calculator takes the same funnel and prices it. If you want to know whether an individual customer is worth acquiring at that price, the answer lies in customer lifetime value against cost per acquisition.
For organic and content channels, the traffic figure needs one more translation. Sessions come from rankings, and rankings deliver sessions through impressions and click-through rate, so a session target implies an impression target — that conversion runs through the click-through rate calculator. It is worth doing, because it converts a vague content plan into a specific volume of ranking positions, which is a far more testable commitment.
One structural point about lead-based businesses. When there is a lead stage, small errors in the lead-to-customer rate move the lead requirement a great deal, because it sits in a denominator. At a 25% close rate you need 4 leads per customer; at 20% you need 5, which is 25% more lead volume for a 5-point change in the rate. Measure that rate carefully, agree the lead definition with the sales team in writing, and re-derive it every quarter rather than carrying a number forward from a period whose lead mix no longer applies.