Sena replaces the rate with what buyers across the category bought last cycle and what they say they will buy next.
Revenue forecasting projects what a business will earn in a period, so it can commit to a plan against that number. Three properties of the usual method bend the projection.
The forecast starts from what happened and applies a rate. That method carries every condition of the prior period forward, including the ones that have already changed.
The number moves between the team producing it and the team receiving it until it lands near what was expected. What arrives is an agreement, reached with the market absent from the room.
The forecast arrives as a total. When it lands wide, the reason takes a quarter to establish, because price, volume, mix, and market are folded together inside it.
Sena is the decision AI with access to real-world data. It records what buyers across the category bought in the closed cycle and asks those buyers what they intend next, so the period ahead rests on two measurements, where a plan rests on one rate.
Category receipts show what moved, including the rival items, so the base is current.
Buyers state what they plan to buy and at what price, so the projection carries a forward signal.
Price, volume, mix, and market each carry their own line, so a miss opens onto its cause.
Revenue forecasting estimates what a business will earn over a defined period. The arithmetic is settled, and the difficulty is the inputs, since every forecast is a set of assumptions about a market with a number attached.
Two routes are in general use, and most businesses run both and reconcile.
| Route | How it builds | Where it fails |
|---|---|---|
| Top-down | Market size, then a share assumption. | The share assumption carries the whole forecast. |
| Bottom-up | Units by item by market, then price. | Accurate on structure, slow, and it inherits stale prices. |
A total is difficult to defend and easy to miss. Splitting the projection into four makes both tractable.
What each unit is expected to realize, against list, with promotional weeks marked.
Units expected, from what the category bought and what buyers state they intend.
Which items and packs carry the volume, since mix moves revenue with prices held.
The same four held per country, because the components rarely move together.
Four collected inputs feed the projection, applied per market.
| Input | What it answers |
|---|---|
| Receipts | What the category bought in the closed cycle, dated. |
| Stated preference | What buyers intend next, and the price it depends on. |
| Store captures | What was available and at what price, so supply enters the base. |
| Geo-verified photos | What sat on the shelf that week, dated to the outlet. |
The team's own numbers form a separate row. Volume history, price files, promotional calendars and cost records join through 250+ integrations, so the internal base carries an external forward signal.
Three questions sit outside a forecasting model, and each one is a reason periods miss. History records what buyers did, and a forecast is a claim about what they will do. Own volume rising while the category rises faster is a share loss reported as growth.
A total that lands can contain two markets that missed in opposite directions.
Real people share what they buy and prefer, under explicit consent. Sena captures it directly at the source, so every figure traces back to where it came from whenever a number comes under question.
Recorded purchase and stated intent arrive from the same buyers, and every figure carries the market and the day.
What buyers say they will do sits beside what they did, so the projection carries a forward reading.
Own volume reads against category volume, so growth separates from share.
Stated intent carries the price it depends on, so a forecast breaks where a price move breaks it.
A forecast is a claim about the future built almost entirely from the past. Sena builds every figure from real-world signals captured when the question needs it, from the receipt closing the last cycle through to the buyer naming what they intend next.
Records what the category bought in the closed cycle, so the base carries this cycle's figures.
Reads what was on the shelf and at what price off images captured in real outlets, so availability enters the projection.
Signal arrives from the consumer network under explicit consent. What a buyer intends next comes from that buyer.
Volume history, price files, promotional calendars, and cost records join over 250+ integrations, so the internal base meets the forward signal.
Every figure holds its market and its capture date, so a projected component opens back onto the purchases behind it.
Past forecasts, actuals, and variance records, covering every cycle the category has been scanned in.
Four teams read the same projection, and each one needs a different cut of it before they can commit.
The committed number. Needs the projection split into four components, per market.
The price assumption. Needs a realized price forecast, with list held separately.
The plan. Needs to know which market carries the risk before the period starts.
The volume line. Needs category movement beside own movement.
The same two measurements, read against the forward signal each category carries.
Category purchase for the closed cycle against the stated intent for the next.
Upgrade intent and the price it is conditional on, tier by tier.
Repeat purchase at the counter against intent to switch to own label.
Renewal and downgrade intent against the tariff each depends on.
One mechanism, applied per market and per category. Each step is documented, which is what carries a committed number through a board review.
Category receipts return what moved in the cycle now ended, so the base is measured.
Real contributors share what they buy and pay under explicit consent, naming what they intend next and the price it depends on.
Sena reports the period as price, volume, mix and market, so a variance opens onto the component that produced it.
Most revenue forecasts take the last period, apply a rate agreed between two teams, and arrive as a single number that resists decomposition when it lands wide. Sena builds the period from measurement.
| Capability area | Typical setup | Sena |
|---|---|---|
| The base | Last period's actuals, closed a month ago. | Category purchase for the closed cycle. |
| The forward signal | A growth rate. | Stated intent, with the price it depends on. |
| The category | Assumed to move with the business. | Measured separately, so share separates from growth. |
| Structure | One total. | Price, volume, mix and market, held apart. |
| Market resolution | A regional roll-up. | Per country, since components move unevenly. |
| When it lands wide | A quarter to diagnose. | The component is visible, with the underlying records held by the provider. |
| Evidence in a review | A model and an assumption sheet. | Open any component onto the buyers behind it. |
Three situations where a measured base changes the projection.
Run the projected period against stated intent per market, so the markets carrying the risk are named before the commitment.
Read own movement against category movement, so a rising number in a faster market shows as the share loss it is.
Open a variance onto price, volume, mix and market, so the next projection corrects one input.
A forecast projects the period. The four decisions below are what a revenue team asks around it.
Every revenue decision Sena answers, in one place
The four levers together
Sets the levers the projection runs on, one level above this number.
Explore revenue management 02The next increment
Ranks the moves before the period is projected, so the forecast covers what will run.
Explore revenue optimization 03The figures underneath
Supplies the realized price the projection runs on.
Explore pricing analytics 04The single price point
Sets the price the volume line is conditional on.
Explore price optimizationThe walkthrough takes one range in one market, builds the base from category purchase for the closed cycle, and adds stated intent for the next while the team watches.
Tell us the range and the period, and we will build the base.