You stop arguing about the numbers and start deciding from them.
Gully Sales builds the measurement layer under your revenue: one written definition for every figure, the data instrumented to produce it, a check that proves it reconciles, and a monthly analysis that says what moved and why.
- One agreed definition for every number your reviews depend on
- Data instrumented at source, so a figure can be traced back and trusted
- A monthly analysis that explains movement, not a screen nobody opens
Gully Sales Private Limited serves small and medium businesses across India, and builds this on the systems and records you already pay for.
In one paragraph
What is Revenue Analytics and Intelligence?
Revenue analytics and intelligence turns scattered marketing, sales and service records into one measurement model you can act on. Gully Sales agrees the definitions, instruments the events and data sources behind them, verifies that the figures reconcile, and hands your leadership a monthly analysis explaining what changed, where it changed and what to do next.
The problem
Your reports disagree, and none of them tells you what changed.
This is rarely a reporting failure. It is what happens when a business grows faster than its measurement. Somebody started counting enquiries in a sheet. Somebody else set up a CRM view. Finance kept the invoice ledger. Each was correct on the day it was built. Nobody agreed what an enquiry, a qualified lead or a closed deal actually means, so now three honest people produce three different totals for the same month, and the review spends its time reconciling instead of deciding.
You will recognise it as
- Two departments present the same month and the totals do not match, so the meeting turns into a discussion about the sheets.
- You can see that enquiries fell, but nobody can say which source, region or product fell, or when it started.
- The monthly numbers are assembled by hand over two days, and only the person who assembles them believes them.
- Reports show totals and trends, but never why a trend moved, so every review ends in opinion rather than action.
- Marketing spend is reported by channel, sales by seller, service by ticket, and none of the three can be joined to a customer.
- A question as simple as cost per closed customer takes a week to answer, and the answer changes if somebody else answers it.
What it costs the business
- Money keeps going to channels and activities nobody has ever proved are working, because the evidence to stop them does not exist.
- Problems are noticed a quarter late, when the revenue has already been missed, instead of in the month the leading number turned.
- Senior people spend their week building sheets rather than talking to customers, and the cost of that time never appears in any report.
- Decisions default to whoever argues most confidently, so the business becomes dependent on instinct that cannot be checked or handed over.
- New tools get bought to fix what is actually a definition problem, and the new tool reports the same confusion more attractively.
Why it persists. Measurement belongs to everybody and therefore to nobody. Marketing owns campaigns, sales owns the target, finance owns the ledger, and the definitions that would join them are not on anyone's list of goals. Fixing them is invisible work that never feels as urgent as this month's closing, and the manual workaround does keep producing a number. So the gap widens quietly. By the time the reports are openly contradicting each other, the habits are years old and buried inside several systems.
If it stays unresolved. Reporting stays a monthly argument, and it gets longer as the business adds people, regions and products. Spending decisions stay unexamined, so waste compounds. How the business really works stays in two or three heads, and leaves when they do. Eventually guessing becomes expensive, and the clean-up costs far more than defining the numbers properly would have while the business was smaller.
What changes
What changes once the numbers can be trusted.
In the first weeks
- One written definition for every metric, so a word means the same thing in every room.
- A single reconciled view of the current month that marketing, sales and finance all recognise.
- The manual assembly of the monthly pack largely replaced by a view that refreshes itself.
In how the work runs
- Leaks are located to a stage, source, region or product instead of being described as a general slowdown.
- Weekly reads on leading numbers, so a problem is caught while there is still a month to fix it.
- A named owner for every metric, and a routine that catches a wrong figure before it reaches a review.
In sales and marketing
- Spend can be judged against what it actually produced, so budget moves from the unproven to the proven.
- Forecasts built from recorded stage behaviour rather than from what the pipeline feels like this week.
- Cost to acquire a customer, and revenue per seller, on one definition everybody accepts.
In what management can see
- Leadership reviews start from evidence, and the first fifteen minutes are no longer spent agreeing whose sheet is right.
- Board, banker and investor questions can be answered the same day, from the same source.
Over the longer term
- A measurement model that survives new products, new regions and new people because it is written down.
- A management habit of deciding from evidence, which is the foundation any later automation depends on.
Gully Sales controls the definitions, the instrumentation, the accuracy checks, the analysis and the review routine. What you then earn depends on the decisions your team takes with that evidence, and your capacity to act. Measurement makes sound decisions likelier; it does not make them.
Who it is for
Who this work suits, and when to start it.
The businesses it suits
- Owners and directors who no longer have direct sight of every deal and need the numbers to tell them what they used to see.
- Businesses running marketing, sales and service across separate tools that have never been joined at the customer level.
- Sales and marketing heads asked to justify spend or headcount, with no evidence trail either side can agree on.
- Companies whose monthly management pack is assembled by hand and takes days that nobody has to spare.
- Multi-branch or multi-product businesses that need the same measures read the same way in every location.
- Founders preparing for a loan, an investor conversation or a professional board, who need numbers that hold up to questioning.
What usually prompts the call
- A month closed far below expectation and nobody can explain which part of the funnel actually moved.
- Two teams presented contradictory figures for the same period in the same meeting.
- A new CRM or marketing platform went live and the reporting still runs out of spreadsheets.
- Marketing spend was increased and there is no agreed way to tell whether the increase paid for itself.
- A funding, lending or due diligence conversation has asked for numbers you cannot yet produce with confidence.
- The one person who compiles the reports has resigned, or is about to go on leave.
What Gully Sales does
The work, component by component.
Measurement plan
We begin with the decisions your leadership actually takes each month, and work backwards to the smallest set of numbers those decisions need. Each metric gets a plain-English definition, a formula, an owner, a refresh frequency and a stated purpose. Metrics that support no decision are removed, which is usually half of what was being reported.
- Why it matters:
- Most reporting grows by accumulation. Nobody removes anything, so attention spreads thinly across figures that change nothing.
- You receive:
- A written measurement plan and a metric dictionary your team can read without us.
- Business value:
- Every number on the page earns its place, and disagreements about what a word means are settled once, in writing, instead of monthly.
Event and stage instrumentation
We map where each number is actually created in your systems: the form submission, the call log, the stage change, the quotation, the invoice. Where the event is not recorded at all, we specify what must be captured, by whom and at what moment, and set it up in your existing tools rather than adding new ones.
- Why it matters:
- A metric is only as honest as the moment it is recorded. Numbers reconstructed afterwards from memory cannot be trusted or repeated.
- You receive:
- An event and stage map, plus the field, form and stage changes required to capture each one.
- Business value:
- The figures start being produced by the work itself, so reporting stops depending on somebody remembering to update a sheet.
Data sources and joins
We inventory every system holding revenue-relevant data — website, ad accounts, CRM, telephony, WhatsApp or email tools, billing and accounting — and establish how each connects, how often it refreshes, and the common key that lets a customer be followed from first enquiry to invoice.
- Why it matters:
- Without a shared identifier, marketing, sales and finance can each be accurate and still be unable to describe one customer's journey.
- You receive:
- A data source inventory and a documented join, showing what each system contributes.
- Business value:
- You can follow revenue end to end instead of studying three separate summaries and guessing how they relate.
Analysis and intelligence build
We build the analysis on top of the model: stage-by-stage conversion, source and channel performance, cohort behaviour over time, win and loss patterns, product and segment contribution, sales cycle length, and a pipeline view that shows coverage against the target rather than a single total.
- Why it matters:
- A total tells you the score. Analysis tells you which part of the game produced it, which is the only thing you can act on.
- You receive:
- Built analytical views in the tools you already hold, with a sample extract you can circulate.
- Business value:
- Questions that used to take a week of sheet work are answered in the meeting where they are asked.
Quality assurance and reconciliation
Every reported figure is tested against an independent record — reported revenue against the invoice ledger, reported leads against raw form and call logs, stage counts against the underlying records. Gaps are investigated and either corrected or explained in writing. We then set the routine checks that keep the numbers honest afterwards.
- Why it matters:
- A single wrong figure spotted in a review destroys confidence in the whole pack, and people quietly return to their own sheets.
- You receive:
- A reconciliation record with variances explained, and a written data quality checklist with owners.
- Business value:
- The numbers survive being challenged, which is what makes people willing to act on them.
Decision views
The analysis is surfaced as a small number of role-appropriate views: an owner view of the whole funnel and cash conversion, a sales view of pipeline and stage movement, a marketing view of source performance. Each view is limited to what that person can act on, and states when it last refreshed.
- Why it matters:
- A screen with forty tiles is read once, admired, and abandoned. A screen with six numbers somebody owns gets opened every week.
- You receive:
- Working decision views, plus a short handover recording so your team can extend them.
- Business value:
- Each role sees its own evidence without needing a request, an export or a reminder.
Decision cadence
We set the rhythm that turns analysis into action: which numbers are read weekly, what the monthly review covers, what is examined quarterly, who attends, what each meeting is expected to decide, and how last month's decisions are checked against what followed.
- Why it matters:
- Analysis with no meeting attached becomes an archive. The cadence is what converts evidence into a changed decision.
- You receive:
- A written cadence, meeting agendas and a monthly analysis pack template.
- Business value:
- The reporting keeps working after we leave, because it is attached to meetings that already happen.
What you will have at the end.
- A written measurement plan naming every metric, its definition, its formula, its owner and the decision it supports.
- A metric dictionary in plain English, so one word means one thing across marketing, sales, service and finance.
- An event and stage map showing exactly where each number is created, updated and closed in your existing systems.
- A data source inventory listing every system we read from, what it contributes, and how often it refreshes.
- The instrumentation itself: the fields, forms, stage rules and tracking changes needed to capture what was missing.
- Built analytical views covering funnel conversion, source performance, cohorts, cycle length and pipeline coverage.
- A sample anonymised extract of the monthly pack, so you can see the format before the first live month is produced.
- A reconciliation record comparing reported revenue and leads against invoices and raw logs, with variances explained.
- A data quality checklist: what is verified, by whom, how often, and what happens when a figure looks wrong.
- A monthly analysis pack: what moved, where it moved, the likely reason, and the two or three actions it suggests.
- A decision cadence document setting the weekly, monthly and quarterly reviews, their agendas and their attendees.
- A handover note and screen recording so your team can maintain and extend every view without depending on us.
How it runs
The engagement, step by step.
- 1
Decision inventory
We sit with the owner and the heads of marketing and sales and list the decisions actually taken each week, month and quarter — where to spend, whom to hire, which product to push, which region to expand, which deals to escalate. Each decision is written down with the evidence it currently rests on. This list, not the data, is what the measurement model is built to serve.
- You provide:
- Ninety minutes each from the owner and the functional heads, and honesty about how decisions are currently made.
- We produce:
- A written decision inventory, with the evidence gap named against each decision.
- Done when:
- You agree the list of decisions the reporting must improve.
- 2
Measurement plan and definitions
Working from that list, we define the smallest set of metrics that supports it. Each gets a definition, a formula, a source, an owner and a frequency. Where two teams currently count the same thing differently, we put both versions side by side and get one agreed. Metrics that support no decision are deliberately dropped.
- You provide:
- Existing reports, sheets and CRM views, plus one working session to settle contested definitions.
- We produce:
- The measurement plan and metric dictionary, circulated for written sign-off.
- Done when:
- Every metric has one definition and one named owner.
- 3
Source and event mapping
We trace each defined metric back to the moment it is created in a real system and record how it gets there. Missing events are listed, along with what it would take to capture them. We check what your website, ad accounts, CRM, telephony and billing systems already hold, and what is being lost because nobody asked them to record it.
- You provide:
- Read access to the relevant systems, and time with the people who use them daily.
- We produce:
- A data source inventory, an event map and a gap list ranked by effort against value.
- Done when:
- Each metric has a traceable path from the work to the number.
- 4
Instrumentation
We make the changes that let the numbers be produced by the work rather than reconstructed later: form fields, tracking, call and message logging, stage entry and exit rules, required fields at the points that matter, and the joins that let one customer be followed across systems. We change your existing tools; we do not sell you new ones.
- You provide:
- Administrator access, and a decision on any field or stage change that affects daily routine.
- We produce:
- Configured tracking, fields, stage rules and connections, with every change documented.
- Done when:
- A test enquiry can be followed from first click to invoice without manual repair.
- 5
Analysis build
We build the analytical views on the instrumented model: conversion at each stage, source and channel performance, cohort behaviour, win and loss patterns, segment and product contribution, cycle length, and pipeline coverage against target. Each view is built in the tools you already own and is limited to what its reader can act on.
- You provide:
- A review of the first drafts, and a clear statement of which views each role will use.
- We produce:
- Working analytical and decision views, plus a sample extract of the monthly pack.
- Done when:
- Each role can open its own view and read it without explanation.
- 6
Quality assurance
Before anything is trusted, every headline figure is tested against an independent record: revenue against the invoice ledger, leads against raw form and call logs, stage counts against the underlying records. Variances are chased to their cause and either corrected or documented. We then write the routine checks that hold the model honest month after month.
- You provide:
- Access to invoice records and one finance contact who can confirm what was actually billed.
- We produce:
- A reconciliation record with variances explained, and a data quality checklist with named owners.
- Done when:
- The reported numbers reconcile to their independent source, or the difference is written down.
- 7
Cadence and handover
We run the first review cycles with your team — the weekly read, the first full monthly analysis — so the routine is learned in practice rather than described in a document. We agree agendas, attendees and what each meeting must decide, then hand over the model with a recorded walkthrough so your people can extend it themselves.
- You provide:
- Attendance at the first review cycles, and a named internal owner for the model afterwards.
- We produce:
- The cadence document, the first live analysis packs, and a handover note with a recording.
- Done when:
- Your team runs a review from the model without us in the room.
Ways to work with us
Start with a diagnostic, or keep us through the first quarter.
Measurement diagnostic
A short review of what you currently measure, where the definitions conflict, which numbers cannot be traced to a source, and what your existing systems could already produce. Delivered as a written findings note and a working discussion.
Analytics and intelligence build
The full engagement: decision inventory, measurement plan, source and event mapping, instrumentation, analytical views, reconciliation, quality checks, the decision cadence and a recorded handover to your team.
Build with a supported first quarter
The build, followed by Gully Sales preparing and presenting the monthly analysis for one quarter while your team learns the routine, so the cadence is established before responsibility passes across.
Ongoing analysis retainer
For businesses that want the monthly analysis prepared and interpreted for them: refreshed views, verified numbers, a written pack of what moved and why, and attendance at the monthly leadership review.
Why Gully Sales
What you are actually choosing when you choose us.
We start from decisions, not from data
The model is built backwards from the choices your leadership actually makes. That is why it stays small, why it gets used, and why it does not become another report that is admired once and then ignored.
We work inside the tools you already pay for
Most Indian SMBs already own more reporting capability than they use. We would rather make your CRM, ad accounts and billing system tell the truth together than add a licence you have to justify next year.
We reconcile before we present
Every headline figure is tested against an independent record before anyone is asked to believe it. Numbers that cannot be reconciled are marked as such, because a quietly wrong figure costs more than a missing one.
We know the sales side, not only the data side
Gully Sales works on pipelines, sellers and campaigns every week. So the analysis is read the way a sales head reads it, and the recommendations are ones an operating team can actually carry out.
We hand the model over
Definitions, sources, checks and views are documented and recorded so your own people can maintain and extend them. The work is finished when your team can run a review without us in the room.
Where it applies
The same service, in different businesses.
Industrial manufacturing
- The situation:
- Enquiries arrive from a website, two marketplaces, dealer references and trade exhibitions, and the sales team records them in one CRM field called source, filled in differently by each person.
- How it applies:
- Source definitions are fixed, capture is standardised at the point of entry, and enquiries are followed through quotation to dispatch so each origin can be judged on orders rather than on volume.
- Likely benefit:
- Exhibition and marketplace spend can be compared against dealer-led business on the same basis, and the budget follows what actually converts.
Multi-branch healthcare
- The situation:
- Three clinic locations report appointment numbers weekly, but no one can say how many new patients each location keeps beyond a first visit, or which service line brings them back.
- How it applies:
- Patient records, call logs and billing are joined on a common identifier, and cohort views show retention and repeat value by branch, service and referral source over time.
- Likely benefit:
- Investment goes to the branches and service lines that hold patients rather than the ones that merely record the most first appointments.
B2B technology services
- The situation:
- A long sales cycle means the quarter's revenue was decided months earlier, but the pipeline is reviewed as a single total, so a shortfall is discovered only when it arrives.
- How it applies:
- Stage-by-stage conversion, cycle length by segment and pipeline coverage against target are measured, and leading indicators are read weekly instead of monthly.
- Likely benefit:
- A coverage shortfall is visible with a quarter's warning, while there is still time to add pipeline rather than explain a miss.
Real estate and property
- The situation:
- Portal spend, site visits and bookings are tracked by three different teams in three different sheets, and cost per booking is estimated at the end of a campaign, if at all.
- How it applies:
- Portal, call centre and site-visit records are joined into one funnel, with cost carried through to booking by project, portal and campaign period.
- Likely benefit:
- Portal and campaign spend is judged on bookings by project, so money moves off listings that generate enquiries but not visits.
Consumer brands and retail
- The situation:
- Online sales, marketplace sales and offline distribution are reported separately, so the same customer is counted twice and the true cost of acquiring one is never established.
- How it applies:
- Channel data is unified, repeat purchase behaviour is measured in cohorts, and acquisition cost is calculated on one agreed definition across every channel.
- Likely benefit:
- Discounting and acquisition decisions are made on repeat value rather than on first-order revenue alone.
Education and training
- The situation:
- Admissions counsellors work from a shared enquiry list, and the season's conversion is only understood after it has ended, when the intake is already fixed.
- How it applies:
- Enquiry-to-admission conversion is measured weekly by counsellor, course and source, with follow-up activity tracked against outcomes through the admission window.
- Likely benefit:
- Counselling effort is redirected mid-season towards the courses and sources still converting, instead of being reviewed after intake closes.
Questions buyers ask
Before you enquire, the answers you will want.
Which business decisions should this reporting improve?
The ones you already take and currently take blind: where to spend the next marketing rupee, which channel or region to expand, whether to add a seller, which deals need escalation this week, which products carry the business and which quietly drain it, and whether a shortfall is a demand problem or a follow-up problem. We list your decisions first, then build only the numbers those decisions need. Anything that supports no decision is deliberately left out.
How is this different from the dashboards already in our CRM?
A CRM dashboard reports what the CRM was told. If stages mean different things to different sellers, if half the enquiries never reach the system, and if billing lives elsewhere, the dashboard will be neat and still wrong. This work fixes the layer underneath: agreed definitions, events captured where the work happens, sources joined on a common key, and figures reconciled against invoices. The screens matter less than whether what feeds them is true.
How long before the first reliable month?
It depends on how many systems hold your data and how much instrumentation is missing, so we scope it after the free audit call rather than quoting a standard duration. The pattern is consistent though: definitions and mapping come first, instrumentation next, then the analysis, then reconciliation, then two or three live review cycles run with your team so the routine is learned in practice. We agree the sequence and the dates in writing before starting.
What do you need from us to define the metrics?
Read access to the systems holding revenue data, administrator access where instrumentation must change, and time from the people who use those systems daily. From leadership we need roughly ninety minutes each for the decision inventory, a session to settle contested definitions, and one finance contact who can confirm what was actually invoiced. After go-live we need a named internal owner and attendance at the first review cycles. Without that owner, any reporting model decays.
How do we know the analysis itself is correct?
Against the baseline we record before anything is built. Operationally: data completeness, the share of figures that reconcile to an independent record, and hours spent assembling the monthly pack. Commercially: lifecycle conversion by stage, forecast reliability, lead-response speed and revenue per seller. We also track decision follow-through, because analysis that changes no decision has failed regardless of how accurate it is. Every measure is reported against its starting point, not in isolation.
What is outside an analytics engagement?
We do not run your campaigns, sell your pipeline or take over your finance function. We do not build a data warehouse or write custom software. Deep dashboard suites for every role, large-scale record cleaning, CRM rebuilds and AI-generated commentary are separate services, and we will say when one of them is the real requirement. Anything outside the agreed scope is quoted separately rather than absorbed quietly into this work.
Our data is messy. Should we clean it before starting?
No. Waiting for clean data is the most common reason this work never begins. We start by measuring how complete and traceable your records are, which tells you exactly which gaps matter for the decisions you take. Some are fixed by instrumentation so the mess stops growing; some need a separate cleansing exercise. Either way you learn what the data can honestly support now, instead of postponing measurement until a tidy state that never quite arrives.
Do we need to buy a new analytics tool for this?
Usually not. Most businesses we work with already own more reporting capability than they use inside their CRM, ad accounts, billing system and spreadsheet tools. Our preference is to make those agree with each other. If a genuine gap remains after mapping, we say what class of tool would close it and why, with the trade-offs written down. The recommendation is yours to accept, and we hold no reseller arrangement with any vendor.
4 more questions
Who maintains this after the engagement ends?
Your named internal owner, which is why we insist on one being appointed early. Definitions, sources, checks and views are documented, and the handover includes a recorded walkthrough so views can be extended without us. Where a business has no one able to hold it, we offer an ongoing retainer that prepares and interprets the monthly analysis. What we avoid is leaving a model that only works while we are still being paid.
Can you measure marketing and sales together, or only sales?
Together, which is the point. Measuring them separately is what produces the argument about lead quality. We join campaign, enquiry, sales activity and billing data on a common key so one customer can be followed from first click to invoice and, where relevant, to repeat purchase. That lets you see cost per closed customer by source rather than cost per lead by channel, which is a very different and far more useful number.
We are a single-owner business. Is this too much for us?
Not necessarily, but the scope should be smaller. A business where the owner still sees every deal needs perhaps six numbers, captured properly and read weekly, rather than a full analytical model. The diagnostic will tell you honestly whether that is your situation. If it is, we scope a short engagement to fix definitions and capture, and leave the deeper analysis until the business is large enough to need it.
Will this tell us why the numbers moved, or only that they moved?
That is the difference between reporting and analysis, and the reason this work exists. The views break movement down by stage, source, segment, region, product and cohort, so a fall can be located rather than described. The monthly pack states what moved, where it moved and the most likely explanation, with the evidence shown. Where the data cannot settle the question, we say so plainly instead of offering a confident guess.
Talk to us
Start with an honest look at the numbers you already have.
Request a Revenue Operations Assessment. Bring the monthly report you dread assembling and the figure two teams keep disputing, and we will tell you what your existing systems could already prove.
- No obligation and no sales script
- A reply from someone who does the work
- Your details are never sold or shared