Solutions · by department · sales & revenue operations

Sales: bookings, pipeline, cycle and discount, read before the forecast call

Chief sales officer and regional sales directors (attainment, win rate, coverage) · head of revenue operations (pipeline hygiene, stage conversion, slippage) · pricing manager and CFO (price realisation, discount depth) · CIO (the CRM extract is often the least governed model in the warehouse; row-level scope by team matters). The buyer is the CSO with the head of RevOps; the weekly reader is the sales manager preparing the forecast call.

Questions

The questions the head of sales asks every month

Are we on track for the quarter — commit versus quota by region — and which deals slipped?
Is coverage sufficient for next quarter, and where is pipeline creation short by source?
Which teams are below attainment — activity, conversion or deal size?
Where are win rate or cycle length deteriorating, and in which segment?

In conversation

Sample questions people type into the conversation

deepen path — the finding's evidence is carried frozen; a new number needs a new query, and the agent says so

"Break the Mid-market win rate down by loss reason."
"Which teams pushed the most August close dates?"
"Show price realisation by product family for the distributor channel."
"How many open opportunities have a close date in the past?"
"How current is the CRM extract?"

Personas

Three example agents

The agents below are examples for this department, not a fixed set: Chief Sales Officer Agent · Pipeline Performance Agent · Pricing & Discount Agent. Each one’s scheduled run is shown in the example sets that follow.

Set 1

Chief Sales Officer Agent

Audience: chief sales officer, regional directors, CFO.

Set 2

Pipeline Performance Agent

Audience: head of revenue operations, sales managers.

Set 3

Pricing & Discount Agent

Audience: pricing manager, CFO, channel director.

An agent's name says which measures it reads and for whom. It holds none of the role's authority: it does not decide, approve or act.

Agents for the topics you choose

An agent is a job description, a bound semantic model and a schedule — not code — so the topics it watches are yours to set: a payer, a service line, a campaign, a supplier. We write the first ones with you.

Build your agents with us →

Example sets

Three example sets, one run each

Each set is one agent persona and one scheduled run’s output — the brief, the recommendations and the findings they rest on — in the product’s own record envelope. The first set is open; the other two are collapsed. The model and the measures the sets assume are listed at the foot of the page.

IllustrativeSet 1Chief Sales Officer Agent

Persona: Produces the quarter-to-date briefing for the chief sales officer of a B2B industrial-equipment maker — six regional teams, a direct and a distributor channel, three customer segments, bookings in €; separates "more bookings" from "more deals", and "fewer wins" from "slower wins". · Audience: chief sales officer, regional directors, CFO. · Tone: direct, number-led; states the CRM's gaps where they touch a number. · Output language: en
Signals it watches: (1) Bookings–deal-count scissors — bookings rising while the number of won deals falls. (2) Win rate by segment against the same window last year, with the decided base. (3) Cycle-length drift on opportunities with complete stage dates, with the excluded share stated. (4) Coverage for next quarter, with stale-dated pipeline shown separately. (5) Regional divergence — one region moving alone is a regional story; several moving together is a network story.
Thresholds it was given: finding 10 % · critical 30 % · ratio metrics (win rate) 3 points · coverage 0.3× · minimum share 5 % of bookings · stale = open with a close date in the past · severity rule: medium when a threshold is crossed, high when a ratio metric moves more than 1.25× its threshold or a count metric more than 1.5× the finding threshold, critical above the critical threshold; an ongoing topic keeps the severity it was given · override: a move in the favourable direction never raises a finding's severity above its own crossing metric

Run header

Scheduled weekly, Monday 06:00, Europe/Berlin
Period: Q3 2026 to date (1 July – 20 September, 58 working days) vs the same window of 2025 (59 working days) — totals are compared as totals and per working day where noted
Data complete through 2026-09-20, measured via max(STAGE_CHANGE_DATE), 7 hours ago (the CRM extract lags one night)
Scope: this data model has row-level rules by team; the agent runs under the all teams scope
Model: the workspace's approved model, written onto the run
Cost: recorded on the run

Brief

Q3 2026 to 20 September: Bookings +9.87 % on 12.62 % Fewer Deals — Mid-Market Win Rate 24.09 % → 19.61 % With an 88-Day Cycle, Q4 Coverage 2.54× With 8.20 % of Pipeline Past-Dated

Assumptions strip

You did not state a period; the current quarter to date (1 July – 20 September 2026) was assumed, compared with the same days of 2025. The two windows differ by one working day; per-day figures are given where they change the reading. Opportunities without stage dates are excluded from the cycle measure and counted.

Bookings are ahead of last year by 9.87 %, but on 12.62 % fewer won deals: the average deal grew 25.73 %, South grew 30.77 % while North fell 8.41 %, and whether the larger deals are larger customers or bundled contracts is not something the CRM can say. In Mid-market the win rate fell from 24.09 % to 19.61 % on a larger decided base while the cycle lengthened from 71 to 88 days; the two move together and the loss-reason field is filled on only 60.98 % of the losses. Q4 coverage stands at 2.54× against 3.21× a year earlier, and 8.20 % of the pipeline it is read on carries a close date in the past. A draft sentence estimating the pipeline "at risk" was held.

covers: Finding 1, Finding 2, Finding 3

— Run: weekly, the workspace's approved model, data complete through 2026-09-20. Guard Ledger: held 1 — candidate f-04 (pipeline at risk), reason numeric_provenance: token about a quarter has no address and no declared operation; duplicate 1 — candidate f-05 (bookings restated by channel) merged into Finding 1 on a fingerprint match (metric bookings + period + cell); not_selected 0.

Recommendation 1

Break Q3's decided Mid-market opportunities down by loss reason and by competitor-present flag, and record the 176 losses (39.02 %) that carry no reason as a data gap to close

based_on: Finding 2 · owner: head of revenue operations, with the Mid-market sales director

A falling win rate with a lengthening cycle can be a harder market, a weaker offer or a slower internal process, and the CRM can separate the first two only if the loss reason is filled. The loss-reason and competitor cuts are in this model; the 39.02 % unfilled share is the first thing the cut will show. Until it is filled, the record says what moved, not why.

The findings this rests on

Finding 2

Mid-Market Win Rate 24.09 % → 19.61 % (−4.48 Points) on 494 → 561 Decided Opportunities While Cycle Length 71 → 88 Days (+23.94 %) — 12.30 % of Decided Opportunities Lack Stage Dates and Are Excluded From the Cycle Measure

severity highdirection downnovelty newtrust Abreakdown SEGMENT = Mid-market

In the Mid-market segment, 110 of 561 decided opportunities were won quarter to date (19.61 %) against 119 of 494 (24.09 %) a year earlier — a 4.48-point fall, 1.49× the 3-point ratio threshold (severity high). Over the same window the mean cycle from creation to decision lengthened from 71 to 88 days (+23.94 %, 2.39× the finding threshold), measured on the 492 opportunities that carry complete stage dates; 69 of 561 (12.30 %) have no stage-entry dates in the CRM and are excluded from the cycle figure, which is stated rather than hidden. The two lines move together in this window; this data does not show that one drives the other, and the loss-reason field is filled on 275 of the 451 lost opportunities (60.98 %), so the reason cut is partial.

Recommendation 2

Put stale-dated open opportunities on a weekly watch by team, and ask each team to re-date or close them before the forecast call

based_on: Finding 3 · owner: regional sales directors

Coverage read on a pipeline that includes past-dated opportunities overstates itself. The team cut is in this model; the watch will show which teams carry the stale share, and the agent will stamp the topic ongoing rather than alert weekly.

The findings this rests on

Finding 3

Q4 Coverage 2.54× Against 3.21× a Year Earlier (96.4 M € Open Qualified Pipeline Against 38.0 M € Remaining Quota) — 8.20 % of the Pipeline Carries a Close Date in the Past

severity highdirection downnovelty newtrust Abreakdown QUARTER = Q4 2026

Open qualified pipeline dated for Q4 2026 stands at 96.4 M € against a remaining Q4 quota of 38.0 M €, a coverage of 2.54×; a year earlier the same measure read 3.21× (102.0 M € against 31.8 M €). The fall of 0.67× is 2.23× the 0.3× coverage threshold (severity high). Of the 96.4 M €, 7.9 M € (8.20 %) sits in eleven opportunities whose close date has already passed and which are still open — they are shown separately, not excluded, because they are still in the CRM as pipeline. Whether they are late or dead cannot be read from this data; only their owners' teams can say.

Also published in this run, with no recommendation resting on it

Finding 1

Bookings–Deal-Count Scissors: Bookings +9.87 % (41.2 vs 37.5 M €) While Won Deals 214 → 187 (−12.62 %) — Average Deal 175.2 → 220.3 k€ (+25.73 %); South +30.77 %, North −8.41 %

severity highdirection upnovelty newtrust Abreakdown REGION = South

Quarter to date, closed-won bookings reached 41.2 M € against 37.5 M € in the same window of 2025 (+9.87 %; +11.76 % per working day, 58 days against 59). The number of won deals fell from 214 to 187 (−12.62 %, 1.26× the 10 % threshold), so the average won deal rose from 175.2 k€ to 220.3 k€ (+25.73 %, 2.57× the threshold — the crossing metric and the severity, high). The regions do not move together: South booked 11.9 M € against 9.1 (+30.77 %, a favourable move above the 30 % critical line that does not raise the severity under the override) and now holds 28.88 % of bookings (11.9 of 41.2 M €); North booked 9.8 against 10.7 (−8.41 %, 23.79 % of bookings); the other four regions are inside threshold. This data cannot separate a real shift toward larger customers from a change in how deals are recorded: the CRM holds one opportunity per contract, and a bundled contract appears as one larger deal.

What this run could not see.

The loss reason is 60.98 % filled and the stage dates 87.70 % complete; both shares are stated in the record. Activity data (calls, meetings) is not in this model. Whether a bundled contract is one deal or several is a CRM convention the model cannot undo.

IllustrativeSet 2Pipeline Performance Agent

Persona: Reads the pipeline's inflow and movement for the revenue-operations team of the same equipment maker: pipeline created by source, close-date pushes, stage ageing, and the feeds that fill the CRM. · Audience: head of revenue operations, sales managers. · Tone: operational, plain; separates a real decline from a feed gap. · Output language: en
Signals it watches: (1) Pipeline created, month over month, by source, with a per-working-day reading when the months differ in days. (2) Close-date slippage — the share of opportunities due in the month that were pushed, and the pushed value by team. (3) Stage ageing — opportunities past the stage's expected days. (4) Feed completeness — a source that returns no rows for part of the month is a data state before it is a decline. (5) Forecast-category movement in the last week of the month.
Thresholds it was given: finding 10 % · critical 30 % · ratio metrics (push share) 5 points · minimum share 5 % of pipeline created · severity rule: medium when a threshold is crossed, high when a ratio metric moves more than 1.25× its threshold or a count metric more than 1.5× the finding threshold, critical above the critical threshold; an ongoing topic keeps the severity it was given · override: when the compared months differ in working days, the per-working-day figure is the judged one

Run header

Scheduled daily, 06:30, Europe/Berlin
Period: August 2026 (21 working days) vs July 2026 (23 working days) — per-working-day figures given beside totals
Data complete through 2026-09-01, measured via max(STAGE_CHANGE_DATE), 6 hours ago
Scope: the all teams scope
Model: the workspace's approved model
Cost: recorded on the run

Brief

August 2026: Pipeline Created −18.55 % (−10.79 % per Working Day), the Drop Is Outbound and Partner — 38.44 % of August-Dated Opportunities Pushed, Two Teams Hold 52.31 % of the Pushed Value

Assumptions strip

Month-over-month comparison (August vs July 2026) was used; the months differ by two working days and per-day figures are given. The partner-portal feed returned no rows for the last six days of August; this is recorded as a data state, and the partner decline is read with that caveat.

Less pipeline came in and more of it moved out: August's created pipeline of 10.1 M € was 18.55 % below July's, 10.79 % below on a per-working-day basis, with outbound down 32.76 %, partner down 22.58 % and inbound up 8.57 % — the partner line read with a six-day feed gap. Of the 372 opportunities due in August, 38.44 % were pushed to a later month against 27.12 % in July, and two of the six teams hold 52.31 % of the 21.6 M € pushed value. Whether the outbound decline is effort or conversion, and whether the pushes are the customers' or the sellers', is not in the CRM.

covers: Finding 1, Finding 2

— Run: daily, the workspace's approved model, data complete through 2026-09-01. Guard Ledger: held 0; data state 1 — pipeline_created by SOURCE = Partner returned 0 rows for 2026-08-26 to 2026-08-31 (feed interruption); structural drop 1 — candidate r-02 (a partner recommendation resting only on the data state) dropped at gate 0; duplicate 0; not_selected 0.

Recommendation 1

Break August's pushed opportunities down by stage and by push count (first push versus repeat) for the two teams that hold 52.31 % of the pushed value

based_on: Finding 2 · owner: head of revenue operations, with the two team leads

A first push from an early stage is normal re-dating; a repeat push from a late stage is a deal that is not closing. Both cuts are in this model. The push reason is not — the team leads read their own notes beside the record.

The findings this rests on

Finding 2

Close-Date Slippage: 38.44 % of Opportunities Due in August Were Pushed (143 of 372) Against 27.12 % in July (96 of 354), +11.32 Points — Pushed Value 21.6 M €, Two of Six Teams Hold 52.31 %

severity highdirection upnovelty newtrust Abreakdown TEAM = Team North-2

Of 372 open opportunities carrying an August close date at the start of the month, 143 (38.44 %) had their close date moved to a later month by month-end, against 96 of 354 (27.12 %) in July — an 11.32-point rise, 2.26× the 5-point ratio threshold (severity high). The pushed value is 21.6 M €, and two teams hold 11.3 M € of it (52.31 %); the other four teams are each inside threshold. Repeat pushes — an opportunity moved for the second or third time — are 41 of the 143 (28.67 %); the CRM records the push, not its reason, so whether the customer or the seller moved the date cannot be read from this data.

Also published in this run, with no recommendation resting on it

Finding 1

Pipeline Created 12.4 → 10.1 M € (−18.55 %; −10.79 % per Working Day) on 312 → 249 Opportunities — the Drop Is Outbound (−32.76 %) and Partner (−22.58 %) While Inbound +8.57 %

severity highdirection downnovelty newtrust Abreakdown SOURCE = Outbound

Pipeline created in August was 10.1 M € across 249 new opportunities against 12.4 M € across 312 in July (−18.55 % in value, −20.19 % in count); with 21 working days against 23, the per-day figure fell from 0.539 to 0.481 M € (−10.79 %, 1.08× the 10 % threshold). By source the move is not uniform: outbound-sourced pipeline fell from 5.8 to 3.9 M € (−32.76 % in total, −26.35 % per working day — 2.64× the threshold, the finding's cell and its severity, high), partner-sourced from 3.1 to 2.4 (−22.58 %; −15.21 % per day), and inbound rose from 3.5 to 3.8 (+8.57 %). Activity data — calls and e-mails per team — is not in this model; whether outbound effort fell or its conversion fell cannot be read here.

Data state

Data state (not a finding): the partner-portal source returned no rows for 26–31 August (a feed interruption on the partner side). The partner decline in Finding 1 may be partly a feed gap; the finding says so, the model owner is notified, and the agent will re-read the partner line when the rows arrive.

What this run could not see.

The partner feed for 26–31 August. Activity volumes. The reason behind a close-date push. Sales representatives never appear; the nearest cut is the team.

IllustrativeSet 3Pricing & Discount Agent

Persona: Reads what the company actually charges against what it lists, for the pricing manager and the CFO of the same equipment maker: price realisation by channel and product family, discounts above the approval line, and the win rate of discounted deals placed beside the win rate of the rest — without a causal claim. · Audience: pricing manager, CFO, channel director. · Tone: formal, exact; never says "gave away". · Output language: en
Signals it watches: (1) Price realisation — net booked value over list value, by channel and product family. (2) Above-line discounts — the share of won deals discounted beyond the approval line. (3) Discount and win rate side by side, as association only. (4) Distributor realisation against the agreement floor, once the floor is a metric. (5) Realisation on renewals versus new deals.
Thresholds it was given: finding 5 % · critical 15 % · ratio metrics (realisation, shares) 1.5 points, critical 5 points · minimum share 5 % of bookings · approval line 15 % discount · severity rule: medium when a threshold is crossed, high when a ratio metric moves more than 1.25× its threshold or a count metric more than 1.5× the finding threshold, critical above the critical threshold; an ongoing topic keeps the severity it was given

Run header

Scheduled weekly, Monday 07:00, Europe/Berlin
Period: Q3 2026 to date (1 July – 20 September) vs Q2 2026 (full quarter) — a partial quarter against a full one; realisation and shares are ratios and are not biased by the day count
Data complete through 2026-09-20, measured via max(QUOTE_DATE), 7 hours ago
Scope: the all teams scope; cost of goods is outside this agent's scope
Model: the workspace's approved model
Cost: recorded on the run

Brief

Q3 2026 to 20 September: Price Realisation 91.61 % → 89.62 % of List, Distributor Channel at 84.19 % — Above-Line Discounts on 21.39 % of Won Deals, Up From 13.08 %, Won at 34.78 % Against 21.18 % Inside the Line

Assumptions strip

You did not state a period; the current quarter to date was compared with the last full quarter. Ratios only; no total is compared across the unequal windows. Cost of goods is outside this agent's scope; the margin effect of discounting is not in this record. A draft margin figure was held.

The company sold closer to its list price in Q2 than it does now: realisation fell from 91.61 % to 89.62 % of list, and the distributor channel, 33.98 % of bookings, sells at 84.19 % against 92.71 % direct — whether that is inside the distributor agreement cannot be read until the floor is a metric. More deals are discounted beyond the approval line — 21.39 % of wins against 13.08 % — and those deals are won at a higher rate than the rest (34.78 % against 21.18 %), a pairing this data can show but cannot explain. The sentence the agent drafted on "margin given away" was held: no cost figure was queried in this run.

covers: Finding 1, Finding 2

— Run: weekly, the workspace's approved model, data complete through 2026-09-20. Guard Ledger: held 1 — candidate f-03 (margin given away), reason numeric_provenance: token 1.1 M € needs a cost-of-goods operand that is outside this agent's scope; claim-check 1 — candidate f-02's "the discount is buying the win" badged interpretive, not held; duplicate 0; not_selected 0.

Recommendation 1

Break above-line discounts down by product family and by deal-size band, and put the approval timestamp beside the close date

based_on: Finding 2 · owner: pricing manager

If the above-line share is concentrated in one product family or one deal-size band, the approval line may be set at the wrong level for that band; if it is spread evenly, the change is in selling practice. Both cuts are in this model. The approval timestamp will show whether discounts are approved before or after the customer has committed — a process fact the agent can measure once the timestamp is in the model; today it is not.

The findings this rests on

Finding 2

Above-Line Discounts on 21.39 % of Won Deals (40 of 187) Against 13.08 % in Q2 (28 of 214), +8.31 Points — the Win Rate of Above-Line Deals Is 34.78 % (40 of 115 Decided) Against 21.18 % Inside the Line (147 of 694)

severity criticaldirection upnovelty newtrust Ainterpretivebreakdown APPROVAL_LINE = Above

Forty of the 187 deals won quarter to date carried a discount above the 15 % approval line (21.39 %), against 28 of 214 in Q2 (13.08 %) — an 8.31-point rise, above the 5-point critical line for a share (severity critical). Placed beside the win rate: above-line deals were won at 34.78 % (40 of 115 decided) and deals inside the line at 21.18 % (147 of 694). The two facts stand side by side; the sentence "the discount is buying the win" is interpretive and is badged so, because this data cannot tell a discount that won a deal from a deal that would have been won anyway. Cost of goods is outside this agent's scope, so the margin effect of the discounts is not computed here.

Recommendation 2

Load the distributor agreement's floor as a metric, and watch distributor realisation against it monthly by product family

based_on: Finding 1 · owner: channel director, with the data-model owner

Realisation without the floor beside it says how far from list the channel sells, not whether it sells inside its agreement. Adding the floor is a change to the imported model, not to the agent.

The findings this rests on

Finding 1

Price Realisation 91.61 % → 89.62 % of List (−1.99 Points) on 41.2 M € Booked Against 45.97 M € List — Distributor Channel at 84.19 % Against Direct 92.71 %, on 33.98 % of Bookings

severity highdirection downnovelty newtrust Abreakdown CHANNEL = Distributor

Quarter to date, 41.2 M € of bookings carried a list value of 45.97 M €, a realisation of 89.62 %; in Q2 the figures were 40.4 M € against 44.10 (91.61 %) — a 1.99-point fall, 1.33× the 1.5-point ratio threshold (severity high). The channels differ: the distributor channel booked 14.0 M € against 16.63 list (84.19 %) and holds 33.98 % of bookings; the direct channel booked 27.2 against 29.34 (92.71 %). The distributor agreement's floor price is not loaded as a metric in this model, so whether 84.19 % is inside or below the agreement cannot be read here — the channel director reads the agreement beside the record.

What this run could not see.

The distributor floor price, the approval timestamp and cost of goods are not in this agent's model or scope. Whether a discount caused a win is not a question this data can answer.

Every number carries an address

In the product each figure binds to an evidence address — envelope, row, cell, check digit — and a number without one is held, not published. The anatomy of a record →

Derived figures show their operation

A change, a share or a gap is computed by the engine and carries its operands; the model never divides. How every number is proved →

AI governance is the publication layer

Eight deterministic gates, the trust tier, badged interpretation, an assumptions strip that cannot be switched off, and the Guard Ledger. Where the model can and cannot reach →

FAQ

Four questions this page is usually asked

01Our CRM data is not clean. Does that break the agent?

No, but the gaps appear in the record: opportunities without stage dates are excluded from the cycle measure and counted, a feed that returns no rows is a data state, and a loss-reason field that is 61 % filled is said to be 61 % filled. Nothing is silently dropped.

02Can it read the CRM directly?

It reads the imported semantic model on the warehouse extract; the extract lag is measured by the engine and stated in every run header.

03Does it rank sales representatives?

No. Representative name is marked personal; breakdowns stop at the team. A ranking of people is not a record the product produces.

04Can it tell us which deals to push in the forecast call?

It can show which opportunities are stale, pushed or ageing, with their teams and values. Which to commit is the sales director's judgement; the record stops before it.

Free data discovery study

See which of these signals your own warehouse can carry.

Send us the name of your data platform and the domain you argue about most. We read your semantic model with you and answer in writing: which of the signals on this page your model carries today, which need a change in Axoria Data Studio, and what a four-week pilot would measure. No cost, no sequence — a person replies within two working days.

Request a free data discovery study → Start a 4-week pilot

What the examples assume

The data, the model and the measures behind the three sets

Nothing in a set rests on a source outside the model described here. Open what you want to reconcile.

01Assumed data sources — data the sales function already has

CRM (opportunities with stage history, leads, accounts, activities, quotas, forecast categories), configure-price-quote (quotes and quote lines with list and net price), ERP (sales orders, invoices), for retail sales the point-of-sale transactions and footfall counters, the HR system's quota-carrying headcount and ramp dates, sales-engagement tools' sequences and calls where they reach the warehouse. Typical warehouse grain: one row per opportunity per stage change, daily; one row per quote line.

02Assumed semantic model — the minimum for the three personas

Measures: bookings_won · quota · won_count · lost_count · decided_count · win_rate (engine-derived) · average_deal_size (engine-derived) · pipeline_open_qualified · pipeline_created · pipeline_pushed · cycle_days (avg, on opportunities with complete stage dates) · list_value · net_value · price_realisation (engine-derived) · discount_pct · above_line_count · opportunities_without_stage_dates · closed_won_without_invoice.
Dimensions: close date · create date · stage · region · team · segment (enterprise / mid-market / small) · channel (direct / distributor) · source (inbound / outbound / partner) · product family · deal-size band · forecast category · loss reason · competitor-present flag · approval-line flag. Marked personal and never broken down by: sales representative name (the nearest allowed cut is team), contact name.
Data edge: measured by the engine via max(stage_change_date) on the CRM extract; the model declares its extract lag.

03Assumed KPIs — what sales leaders track

KPI	Definition	Unit	Typical band	Usual breakdown
Bookings vs quota (attainment)	closed-won value ÷ quota	%	≈ 43 % of reps hit quota (2025 survey) [S202][S203]	team · territory
Win rate	won ÷ (won + lost)	%	20 % all-industry; software 22 %, finance 19 % [S201]; 19–21 % B2B 2025 [S202]	segment · deal-size band
Pipeline coverage	open qualified pipeline ÷ remaining quota	×	3–4× rule of thumb [S204]	team · quarter
Pipeline created	value of new opportunities in period	currency	no public benchmark found	source · segment
Sales cycle length	opportunity create → close	days	~84 days B2B SaaS; enterprise 6–9+ months [S202][S203]	segment · deal-size band
Average deal size	won value ÷ deals	currency	no public benchmark found	product line · segment
Stage conversion	advanced ÷ entered per stage	%	lead→opportunity ~15–25 %, proposal→close ~25–40 % (unverified) [S204b]	stage · team
Forecast accuracy	\|commit − actual\| ÷ actual	%	no public benchmark found	team · forecast category
Activity volume	calls / e-mails / meetings per rep	count	no public benchmark found	team · activity type
Price realisation / discount	realised ÷ list price	%	no public benchmark found	product · channel
Sales-loaded CAC	sales & marketing spend ÷ new customers	currency	~$239 blended B2B SaaS [S205]	channel · segment
Comparable-store sales growth (retail)	same-store sales vs prior year	%	no public benchmark found	store · region
Sales per square foot (retail)	net sales ÷ selling area	currency/ft²	~$325 US average (secondary) [S206]	store · category
Conversion and units per transaction (retail)	transactions ÷ footfall; units ÷ transactions	%, units	no public benchmark found	store · daypart

Sales: bookings, pipeline, cycle and discount, read before the forecast call

The questions the head of sales asks every month

Sample questions people type into the conversation

Three example agents

Chief Sales Officer Agent

Pipeline Performance Agent

Pricing & Discount Agent

Agents for the topics you choose

Three example sets, one run each

Q3 2026 to 20 September: Bookings +9.87 % on 12.62 % Fewer Deals — Mid-Market Win Rate 24.09 % → 19.61 % With an 88-Day Cycle, Q4 Coverage 2.54× With 8.20 % of Pipeline Past-Dated

Break Q3's decided Mid-market opportunities down by loss reason and by competitor-present flag, and record the 176 losses (39.02 %) that carry no reason as a data gap to close

Mid-Market Win Rate 24.09 % → 19.61 % (−4.48 Points) on 494 → 561 Decided Opportunities While Cycle Length 71 → 88 Days (+23.94 %) — 12.30 % of Decided Opportunities Lack Stage Dates and Are Excluded From the Cycle Measure

Put stale-dated open opportunities on a weekly watch by team, and ask each team to re-date or close them before the forecast call

Q4 Coverage 2.54× Against 3.21× a Year Earlier (96.4 M € Open Qualified Pipeline Against 38.0 M € Remaining Quota) — 8.20 % of the Pipeline Carries a Close Date in the Past

Bookings–Deal-Count Scissors: Bookings +9.87 % (41.2 vs 37.5 M €) While Won Deals 214 → 187 (−12.62 %) — Average Deal 175.2 → 220.3 k€ (+25.73 %); South +30.77 %, North −8.41 %

August 2026: Pipeline Created −18.55 % (−10.79 % per Working Day), the Drop Is Outbound and Partner — 38.44 % of August-Dated Opportunities Pushed, Two Teams Hold 52.31 % of the Pushed Value

Break August's pushed opportunities down by stage and by push count (first push versus repeat) for the two teams that hold 52.31 % of the pushed value

Close-Date Slippage: 38.44 % of Opportunities Due in August Were Pushed (143 of 372) Against 27.12 % in July (96 of 354), +11.32 Points — Pushed Value 21.6 M €, Two of Six Teams Hold 52.31 %

Pipeline Created 12.4 → 10.1 M € (−18.55 %; −10.79 % per Working Day) on 312 → 249 Opportunities — the Drop Is Outbound (−32.76 %) and Partner (−22.58 %) While Inbound +8.57 %

Q3 2026 to 20 September: Price Realisation 91.61 % → 89.62 % of List, Distributor Channel at 84.19 % — Above-Line Discounts on 21.39 % of Won Deals, Up From 13.08 %, Won at 34.78 % Against 21.18 % Inside the Line

Break above-line discounts down by product family and by deal-size band, and put the approval timestamp beside the close date

Above-Line Discounts on 21.39 % of Won Deals (40 of 187) Against 13.08 % in Q2 (28 of 214), +8.31 Points — the Win Rate of Above-Line Deals Is 34.78 % (40 of 115 Decided) Against 21.18 % Inside the Line (147 of 694)

Load the distributor agreement's floor as a metric, and watch distributor realisation against it monthly by product family

Price Realisation 91.61 % → 89.62 % of List (−1.99 Points) on 41.2 M € Booked Against 45.97 M € List — Distributor Channel at 84.19 % Against Direct 92.71 %, on 33.98 % of Bookings

Every number carries an address

Derived figures show their operation

AI governance is the publication layer

Four questions this page is usually asked

See which of these signals your own warehouse can carry.

The data, the model and the measures behind the three sets

Where these pages lead

Industries whose sets use these measures

On this site