A VP of Sales I worked with last year split his territories on a Sunday night. Six reps, one spreadsheet, a column for US states, and a rough sort by the company logos he recognized. By Tuesday the team had their new patches. By April two reps were sitting at 140 percent of quota and three were stuck under 60, and he was sure the three were the problem. They were not. The map was.
That is the most common way territory planning happens at companies between Series A and Series B. Someone opens a spreadsheet on a weekend, divides accounts by region or by an even count, and ships it. Then the team spends the next four quarters arguing about rep performance when the real variable got set in that one quiet session nobody reviewed.
I have rebuilt territory models for SaaS teams with 4 reps and for teams with 40. The pattern is the same every time. The plan is treated as a one-off admin chore instead of a data problem, and the cost shows up later as missed quota, rep churn, and a pipeline gap nobody can explain. This post is how I run it now.
Why account count is the wrong unit
Most first territory plans get balanced on the one number that is easiest to count: how many accounts each rep owns. Fifty each, fair and done. It feels fair because the columns line up. It is not fair, because those fifty accounts have wildly different value.
Give one rep fifty accounts in a dense metro with three Series C buyers and a cluster of mid-market targets actively hiring. Give another rep fifty accounts spread across four states, half of them already churned or too small to matter. Same count. One rep has five times the real opportunity. Then we put them on the same quota and act surprised when one coasts and the other burns out.
Account count is a proxy for effort, not for revenue. And it is a bad proxy even for effort, because a dense patch of named accounts you can work in a tight motion is far less draining than fifty scattered logos that each need their own research and their own angle. The unit that matters is potential, measured against the capacity to actually work it.
Revenue that unbalanced territories quietly cost a sales org, according to territory research summarized across vendor studies. It does not show up as a line item. It shows up as reps who never had a fair shot at their number.
What a real territory model is built from
When I design territories, I am balancing four things, not one. None of them is account count.
The first is total addressable potential. For each account, what is it actually worth to you? Pull average contract value by segment and vertical from your closed-won history, not a guess. A patch full of 200-seat companies in your best-fit vertical is worth more than a patch of 20-seat companies you rarely win.
The second is signal density. How many accounts in the patch are showing buying intent right now? New funding, a relevant exec hire, a tech stack change, headcount growth in the team you sell to. A territory can look strong on paper and be dead this quarter if none of it is in motion. This is where your ideal customer profile does real work. If the ICP is sharp, signal density is something you can measure instead of feel.
The third is your existing footprint. Where do you already have customers, expansion paths, and warm logos a rep can multithread into? A patch with three happy customers ready for upsell is not the same as a cold patch of the same size.
The fourth is rep capacity, and this is the one teams get most wrong.
Capacity is not headcount
Here is the trap. A team plans territories for the headcount on the org chart, not the selling capacity that headcount can actually deliver this quarter. Those are different numbers, and the gap between them is where unworked territory sits, the kind a competitor walks right into.
If you hired two AEs last month, they are not at full capacity. A mid-market AE typically takes 3 to 6 months to ramp. An enterprise rep takes 6 to 12. If you carve a high-potential patch and hand it to a rep who is eight weeks in, you have parked your best opportunity behind someone who cannot work it at full speed yet, and you will read the slow start as a performance issue instead of a planning one.
So before I draw a single boundary I write down, per rep: current ramp state, realistic accounts they can actively work in a quarter, open roles and when they will actually be filled, and known attrition risk. Planned capacity is a fantasy number. Realistic capacity is the one that produces a territory map that holds.
This is also where I push back on the instinct to give every rep a full, dense patch from day one. If your realistic capacity is lower than your account inventory, the honest move is to leave some accounts in a holding pool and assign them as reps ramp, not to spread everyone thin so the spreadsheet looks complete.
How to design a territory that balances
The process is not complicated. It is just rarely done in order. Here is the sequence I run, and it works whether you are in HubSpot, Salesforce, or a Clay table feeding both.
The order of the last step matters. If you ask reps for input before you have a data baseline, the loudest voice gets the best patch and the quietest rep gets robbed. Set the baseline from data, then let reps correct it with on-the-ground knowledge the data could not see, like a relationship with a buyer who just moved companies. Data first, voices second.
One more thing about scoring. Most of the inputs you need are already in your CRM, you just have not connected them. Closed-won ACV by segment is there. Win rate by vertical is there. The signals are the part you usually have to enrich for, and that is a smaller lift than people expect once the CRM data quality underneath it is clean. If the underlying data is a mess, no territory model will save you. Garbage in, unfair map out.
The numbers that tell you a map is broken
You do not need a consultant to know your territories are unbalanced. The signals are sitting in your own reporting, if you look at the right cuts.
Watch the spread in attainment first. If your top reps are at 140 and your bottom reps are at 55, and they are similar in skill, the map is doing that, not the people. Healthy territories produce a tighter band. Watch pipeline coverage by rep next. If one rep has 4x pipeline coverage and another has 1.2x against the same quota, you handed them different games and asked for the same score.
Then watch the quiet one: unworked accounts. Pull the count of high-fit accounts with zero activity in 90 days. That number is your unworked territory, and it usually clusters in the patches you gave to ramping or overloaded reps. It is revenue sitting in your CRM that nobody is touching. The sales velocity math gets distorted by this too, because a rep drowning in a patch they cannot cover will show a slow cycle that has nothing to do with their selling.
A bad territory map looks exactly like a bad rep on your dashboard.
Before you put someone on a plan, check whether the patch you gave them was ever winnable. Most of the time the answer rewrites the conversation.
Territory planning is a quarterly habit, not an annual ritual
The second biggest mistake, after balancing on account count, is treating territory design as a once-a-year event. Markets move faster than your planning cycle. Accounts get funded, get acquired, churn, or change ICP fit between January and June. A static map made in Q1 is already wrong by Q2, and the misalignment compounds quietly until your annual review.
I am not arguing for chaos. Reshuffling patches every month destroys rep relationships and kills momentum, and reps stop trusting the system if their accounts keep moving. What I run is a light quarterly pass. Keep the boundaries stable, but rescore potential, reassign accounts from the holding pool to reps who have ramped, and pull back patches that are clearly dead. Companies that adjust territories on live data, instead of freezing them for a year, see meaningfully more revenue per rep. The number floating around the research is up to 30 percent more, and even if your reality is half that, it pays for the afternoon it takes.
The discipline is what makes this work, not the tooling. A quarterly two-hour review with fresh scores beats a perfect annual model that goes stale by spring.
If you want this connected to the rest of your revenue engine rather than living in a side spreadsheet, that is the kind of plumbing we build in our CRM and RevOps work, and the scoring and enrichment side sits in our AI and automation work.
Reps missing quota in patches that were never fair?
Book a free 30-minute audit. We will pull your attainment spread and pipeline coverage by rep and show you whether the map is the problem before you put anyone on a plan.
Book an audit →Where this breaks in practice
A few honest failure modes, because the clean version above hides some mess.
The first is data you cannot trust. If half your accounts have no segment, no ACV, and a "0" employee count, scoring is fiction. Fix the data before the map. Otherwise you are balancing on noise and calling it rigor.
The second is overfitting to signals. Buying intent is useful, but a patch built only on this-quarter signals goes cold next quarter. Score on a blend of durable fit and current signal, not signal alone, or you will be redrawing the map every eight weeks.
The third is political carve-outs. The named-account list that a senior rep "has always owned" is often where the real imbalance hides. If those accounts are not being worked, they are not relationships, they are hostages. Look at activity before you grandfather anything.
The fourth is going fully automated. Territory software can balance patches in minutes, and the math is genuinely better than a spreadsheet. But the rep who knows a buyer just moved to a target account holds context no model has. The good version is data does the heavy carve, humans do the final correction. Tie it back to how you run pipeline management and the metrics you track, so the map and the scoreboard tell the same story.
Get this right and the payoff is not subtle. Fair maps mean reps trust the system, quota attainment tightens into a band you can forecast, and the pipeline gap you could never explain turns out to have been a planning artifact all along. The work is mostly unglamorous. Clean the data, score honestly, balance on potential, review every quarter. That is the whole game.
FAQ
What is sales territory planning?
Sales territory planning is how you divide your market and accounts among reps so each one has a fair, workable patch. Done well, it balances total opportunity, not account count, against what each rep can realistically cover in a quarter. Done badly, it is a spreadsheet split by geography that quietly decides who hits quota before the year starts.
How do you balance sales territories fairly?
Balance on scored potential, not headcount or account count. Weight each account by fit, average contract value history, and live buying signals, then carve patches so total scored potential is roughly even across reps. Adjust for ramp state so newer reps are not handed patches they cannot work at full speed. Then show reps the data and take their local input as a correction, not as the starting point.
How often should you redo territory planning?
Run a full design once a year and a light review every quarter. The annual pass sets the structure. The quarterly pass rescores potential, reassigns held accounts to reps who have ramped, and retires dead patches. Static annual maps go stale by spring because accounts get funded, acquired, or change fit faster than a yearly cycle can track.
What data do you need for territory planning?
Most of it is already in your CRM: closed-won ACV by segment, win rate by vertical, account segment, and existing customer footprint. The part you usually enrich for is buying signals, such as funding, exec hires, headcount growth, and tech changes. None of it works if the underlying CRM data is dirty, so clean and dedupe first.
Why are my reps missing quota in some territories but not others?
Often the patch, not the rep. If attainment ranges from 140 percent down to 55 across reps of similar skill, and pipeline coverage per rep is uneven, the territory map is creating that spread. Check the opportunity gap between your strongest and weakest patches and the count of high-fit accounts sitting untouched before you put anyone on a performance plan.
Stop planning territories on a Sunday night
If your last territory map was drawn in a weekend spreadsheet and has not moved since, it is almost certainly costing you quota and reps without showing up as a line you can point at. We rebuild territory models on real data, tie them to your CRM and your scoreboard, and set up the quarterly rhythm that keeps them fair as the market moves.
Book a free 30-minute audit and we will show you the attainment spread and the unworked accounts hiding in your current map.