CommercialIntelligence
Adapted to your business strategy.
The traditional channel carries 40 to 70% of your revenue, and it is the one place you cannot see. That is not a reporting gap. It is where your growth has been hiding. Our models map the whole universe, including the stores nobody serves, and work out what each store could actually be buying. That is growth in two directions: more stores on the route, and more from every store already on it. So your team starts the week with an answer: where the next dollar is, and what to execute tomorrow.
Your strategy fails twice
Most consumer goods companies know where they want to grow. What they cannot see is the channel that carries most of it, so the strategy gets built on a picture that was already out of date when it arrived.
You cannot see or predict most of your market
Your reps work a partial map, with no read on which stores are worth more, which are about to go dark, or where demand is building unserved. A meaningful share of your highest-potential stores are on no route at all, and nothing in your reporting would tell you. And when a campaign finally goes out on that map, 30 to 50% of the stores on the list have closed or already belong to a competitor.
The costGrowth you never see, so you never plan for it, and coverage you lose without ever seeing it go.
The strategy fails in translation
Even the decisions you do make do not execute as intended at the shelf. Direction gets diluted between planning and the route, and what actually happened comes back as a photo in a chat thread, weeks after it could have been fixed.
The costNo signal on whether your decisions are working at all.
Both are decisions you cannot make with what you have today.
Commercial intelligence
Two engines. One closed loop.
Built for consumer goods companies whose growth lives in the traditional channel, and whose commercial decisions only count once they become a stop on a rep’s route.
Network Expander
Every store your product should be at, and what each one is worth.
Builds the most complete picture of the market from internal and alternative data: store sales, store census, mobility data, geographic profile and socioeconomic factors. The stores you are in and the stores you should be end up on one ranked list, so effort and investment follow what a store is worth rather than where the route already goes. Three commercial plays run on top of it.
SpotOn
Know exactly what to do next.
Knowing where the opportunity is does not move product. SpotOn turns the coverage picture into specific direction at every level of the commercial structure, all executing one strategy, each measured on what their own role controls.
What the field reports comes back to sharpen the intelligence.
Every visit your team makes feeds the models. What each store actually ordered, what was on the shelf, whether it is still open and still buying: all of it goes back into the potential estimate, the churn risk and the visit frequency for that store. So the ranking your team works from next week is built on what happened this week. Intelligence with no route into the field is a report. A route with no intelligence behind it is habit. Run together they close, and every cycle leaves the models sharper than the last.
What keeps it valuable
Analytics projects disappoint in two predictable ways. They go out of date, and they arrive configured for somebody else’s business. The loop closes the first. How we build closes the second.
Real time data, always current, never a picture that was true last quarter. Every visit your field team logs corrects the record it came from, so the ranking your reps work is sharper the week after than the week before, and the stores that changed hands are out of the list before anyone drives to them.
A study is worth most the day it lands. This is worth most a year in.
The store scoring, the routing logic and the task architecture are calibrated to the priorities you have already set, not a default configuration you have to bend to. If this year’s push is a category, a pack format or a region, the model prioritises for it, and the ranking changes when your plan does.
The model bends to your plan, not your plan to the model.
Want to see this run on your territory?
A 30 minute call. We will map a sample of your universe, including the stores you do not serve yet, and score what it is worth.
Talk with an expertFour things have to be true
Beverages, snacks and confectionery, dairy and bakery, personal care, home cleaning. The category changes what is in the box. It does not change the problem, and what decides whether this fits is not what you sell. It is how your business reaches its market.
Your market is thousands of independent stores, not a handful of chain accounts
Modern trade is a few negotiations a year with buyers who report their own numbers. This is the opposite: a long tail of small stores, each one a separate decision, and no two of them worth the same.
You have your own field force working them
Reps, promoters and supervisors covering more ground than anyone can hold in their head. The route was drawn once and has been worked from habit ever since, and nobody has had the time to ask whether it is still the right one.
Nobody reports sell-out, so you plan on shipments and habit
Independent stores do not tell you what left the shelf. What you have is what you dispatched, a census from last year and a panel that reports the category. None of it comes from the stores where the revenue actually moves.
Growth depends on coverage and how well each store is worked
Not on one big account renewing. On how many stores you reach, which of them are worth reaching, and what actually happens once a rep is standing in the doorway.
That is the operation Commercial Intelligence was built for.
Route, launch, price and margin
Four planning decisions your team already makes every cycle. What changes is the input to each one.
Which stores get served, how often, and by whom. Visit frequency follows what a store is worth rather than how convenient it is to reach.
Where to launch and where to push. Zones chosen on live demand signal instead of last year’s territory map.
Price, pack and promotion decided against the potential of the individual store, not the average of the region.
Where the margin actually sits once cost to serve is counted, so growth is not bought at a loss one store at a time.
Forward deployed
Plug-and-play works for software. It does not work for a commercial field operation, where a model only counts once it has survived contact with a route, a supervisor and a rep with a phone in one hand. Most software implementations end at go-live. Kin starts there.
We join the team.
We onboard like a new hire. System access, a seat in the commercial planning conversations, and our hands on the real process. That is how we learn the day to day of each role: what a director needs to see to set direction, what a supervisor needs to catch the week before it is lost, and what a rep needs in a shop doorway with a phone in one hand. We build for each of them, not for an analyst with time to interpret a dashboard.
Deep business understanding.
We do not build anything until we understand how your commercial operation runs and where it is trying to go. That knowledge goes directly into the store scoring, the routing logic and the task architecture, so the intelligence supports the strategy you have already set instead of pulling against it. It is the difference between intelligence that changes what happens on a route and software that just gets installed.
We show up.
Remote calls and video are convenient, and we use them. But nothing replaces working alongside the people a decision actually passes through: the directors who set it, the supervisors who run the week against it, and the reps who carry it into a store. That proximity is what builds trust, surfaces the reasons a plan bends in the field, and makes the work stick.
We’re accountable for the outcome.
A project does not end when the solution is delivered. We define the success metrics with you up front, coverage gained, effective visit rate, volume against plan, the numbers your business lives by, then track measurable impact against them over time. Efficiency and profitability read against your business objectives, not against ours.
Why the tools you already have stop short
None of these are wrong. They just answer a different question than the one your commercial team is actually asking.
Your last census is already wrong.
A foot census is built street by street over months, and it goes out of date with every store that closes, opens or changes hands while it is being collected. It is a snapshot, and it is stale on delivery. Kin is continuous, so the ranking your team works from is current that morning.
We go where panel data cannot.
Panels and macro reports stop at the distributor. They tell you what happened to a category across a market, which confirms the problem without locating it. They cannot tell you which individual store is underperforming its potential. Panels see the category. We see the shelf.
Built for your scale, not theirs.
Enterprise revenue growth platforms are built to be configured and then run by your own analysts, which assumes a BI team dedicated to operating them and a rollout measured in quarters. Commercial Intelligence is not a lighter version of that. It is a different arrangement: our team operates it alongside your commercial team instead of handing it over, so it starts producing decisions in weeks.
They track. We decide.
Your SFA confirms a rep was at a store. That is a record of activity, not a decision about where the activity should have gone. Kin sits above the app you already run as the intelligence layer, choosing which stores matter and what to do at each one, and it can feed those decisions straight into the SFA or CRM you already run.
The status quo has a price. You just cannot see the invoice.
Manual reports and coordination over chat feel free because the cost never appears as a line item. It shows up instead as coverage you never gained, execution drift nobody caught in time, and decisions made on numbers that were already weeks old.
A decade turning data
into commercial decisions
These models were not adapted for Latin America. They were built in it, inside the channel structures the region actually runs on.
These models read the channel, not the company. A traditional channel has the same structure whether you ship a million cases a month or a fraction of that, so the same models run for the largest bottlers in the region and for teams a tenth their size. Nothing here is an enterprise product scaled down.
A complete, prioritised store universe in four weeks, and a customer base growing 22% a year instead of 10%.
Phigital Prospecting, the horizontal play inside Network Expander, built that picture from data rather than street by street, ranked by potential and ready for the field, and the growth attributable to it took the customer base from roughly 10% organic annual growth to 22%, more than double the rate the business was growing on its own. Store Growth Navigator, the vertical play, works the other direction on the stores a team already serves, and across Latin America it has moved brand share, price realisation and coverage. Churn Radar is the third, and it flags the stores about to stop buying. Neither turns into growth until it reaches the field as routes and daily direction, which is the half SpotOn does. And every role involved in the strategy sees execution from their own angle, with the metrics that role is measured on, so the rep, the supervisor and the director read the same week instead of three different reports.
Trusted & certified
now, they know
Working with Kin has enabled us to carry out high-impact AI-based projects in over 11 Latin American countries, achieving completely disruptive results.”
The idea of processing technical tests using AI became a reality thanks to the expertise of Kin Analytics, paving the way for us to innovate the evaluation industry. For years, they have been an important ally in driving disruption.”
Questions commercial teams ask
What is the traditional channel, and why is it so hard to measure?
The traditional channel, also called traditional trade, is the network of independent stores that sit outside modern retail chains. It carries 40 to 70% of consumer goods revenue in most emerging markets, and almost none of it reports sell-out data. Companies end up planning against distributor shipments and panel estimates, which describe the category but never the individual store. Kin builds the complete store universe and scores each store, which panel data cannot do.
How do I find the stores my reps are not visiting?
By building the universe rather than sampling it. Network Expander crosses your existing portfolio against external sources and resolves them into one list of every active store in your territory, then flags the gap between where your team goes and where the volume actually is. What comes back is not a longer list, it is a ranked one, with the stores you do not serve yet scored alongside the ones you already do.
How do you grow the stores we already serve, not just find new ones?
That is the second play inside Network Expander. Store Growth Navigator profiles each store you already serve, its geographic and socioeconomic profile, foot traffic and shopper preferences, and works out what it could be buying against what it buys today. The gap between those two numbers is revenue available without adding a single store, and it sets the visit frequency, the offer and the investment for that store. Phigital Prospecting grows the base horizontally. This grows it vertically. Churn Radar keeps it.
How is this different from a store census or panel data?
A census is a snapshot and it is stale on delivery. Panel data stops at the distributor and reports categories, not stores. Kin is continuous and store-level, and it updates from what your own field team confirms on every visit. Panels see the category. We see the shelf.
We already use a field sales app. Where does Kin fit?
A field sales app records that a rep was there. It does not decide where they should have been. Kin sits above it as the intelligence layer, deciding which stores matter and what to do at each one, and it can feed those decisions into the SFA or CRM you already run. They track. We decide.
What data do you need from us to start?
Less than most teams expect. Your store-level sales history and your current customer master are enough to begin. We bring the external and alternative layers ourselves, census records, mobility, geographic and socioeconomic data, along with the entity resolution that reconciles them against your records. If your data is messy or incomplete, that is normal, and sorting it out is part of the work rather than something you have to finish before we start.
How long before we see results?
The complete store universe is typically ready in about four weeks. Commercial results follow the first execution cycle, because the model only pays once the field acts on it. We agree the success metrics with you before we start and we report against them.
Do we need an analytics team to run this?
No. Kin is forward deployed, so our analysts and engineers work inside your commercial operation through onboarding, pilot and the full results cycle rather than handing you a login. Teams that do have their own analysts keep them on the questions only they can answer, instead of on operating the model week to week.
See your territory
as it actually is
A 30 minute call, then an audit of one territory, so you can put a number on the opportunity you are not working yet.
Talk with an expert
Tell us where your commercial decisions are running blind, and which territory hurts most.
We will build a sample of your full store universe, including the stores no rep visits, and rank it by potential.