# The Competitive Threat Tier: Deep-Monitor, Watch, or Ignore a Rival

*You will score any single competitor across five weighted public dimensions and land on a defensible tier and review cadence that a second analyst can reproduce.*

- Canonical URL: https://www.refolk.ai/guides/competitive-threat-tier-rubric
- Pillar: Market and talent intelligence
- Format: Framework
- Published: 2026-09-30
- Last reviewed: 2026-09-30
- Reading time: 16 min
- Keywords: how to prioritize competitors to track, competitive threat scoring framework, competitor tiering model, how often to review competitors, rank competitors by threat level

## Key takeaways

- Job-posting acceleration leads net headcount by one to three months, so the earliest momentum read is posting velocity, not employee count; Momentum Group posted a 68.4% jump in active job postings against 0.6% headcount growth.
- The tier's real payload is its review cadence, not its depth: deep-monitor rivals get weekly refreshes, watch rivals monthly, and ignore rivals a quarterly scan.
- The weighting choice, not the individual scores, decides the tier, so lock weights to your strategic question before any grading and two analysts will converge.
- Reactivity is the least-adopted CI dimension - Porter's Four Corners did not make SCIP's top-ten tools - yet it is decisive: a fast rival that never reacts is a lower threat than a slow one that always does.
- Grade market overlap on a two-axis grid of product similarity by segment overlap rather than a binary direct-versus-indirect label, which hides the indirect rival drifting into your lane.
- Emerging entrants break a public-signal rubric because they lack a headcount curve or funding trail, so scan them quarterly rather than scoring them.

Deciding how closely to watch a competitor is a repeated judgement call, and most teams make it by gut. This guide is for competitive-intelligence analysts, strategy and research teams, and operators sizing a market who need to spend their watching hours on the rivals that can actually move the market. It gives you a weighted rubric, scored entirely from public signals, that outputs a monitoring tier - deep-monitor, watch, or ignore - and a matching review cadence, so two analysts grading the same company reach the same tier.

Existing public guidance stops at sorting rivals into direct versus indirect and telling you to focus on the biggest threats without saying how to measure "biggest." That is where the gut sort creeps in. What follows replaces the sort with a graded call: five publicly observable dimensions, explicit weights that sum to 1.0, a computed total, and a cadence that follows from the tier.

## Why a graded tier beats a direct-versus-indirect sort

A tier is a defensible, reproducible verdict; a direct/indirect sort is a label that hides the rival about to become a threat. The problem with the binary is that it is a snapshot of today's overlap, and the competitor you most need to catch is the one whose overlap is changing.

The cost of getting this wrong is measurable. Sellers go head-to-head with competitors in 68% of deals, yet the average team rates its own competitive effectiveness a 3.8 out of 10. That gap is not a data problem - the data is public. It is a prioritisation and cadence problem: teams watch the wrong rivals at the wrong frequency, and the January deck is still on screen in April.

**3.8/10 - Average team's self-rated competitive effectiveness, against rivals in 68% of deals**

The head-to-head gap is a prioritisation and cadence problem, not a data-access problem.

The fix is to make tiering a graded call. Teams that pair KPIs, a CI platform, and an executive sponsor achieve revenue impact at 3.6x the rate of teams with none of the three - and the through-line of all three is discipline: a defined method, a place to run it, and someone accountable for the loop. A weighted rubric is the KPI layer of that discipline.

> A label tells you what a rival is today; a graded tier tells you whether it can move your market.

## The five dimensions and what each one proves

Score every competitor on the same five publicly observable dimensions: market/segment overlap, momentum, resourcing, trajectory toward your segment, and reactivity. Each maps to an established model and to specific public evidence you can pull without a vendor relationship.

| Scorable dimension | Source model | Public evidence to score it |
|---|---|---|
| Market/segment overlap | Two-axis competitor map | Product-similarity by segment grid |
| Momentum | Growth-signal practice | Headcount curve, job-posting velocity |
| Resourcing | GE-McKinsey KSF weighting | Funding rounds, 10-K/10-Q filings |
| Trajectory toward segment | Segment-expansion signal | Job postings by target market |
| Reactivity | Porter's Four Corners | Observable Actions corners |

Here is what each dimension proves, and what it looks like when it lies.

**Market/segment overlap** proves whether the rival competes for your buyer at all. Grade it on two axes rather than a binary: a Y-axis of product similarity (low, medium, high) and an X-axis of customer-segment overlap (low, medium, high). Direct competitors cluster high-high; indirect ones sit high-medium or medium-high. It lies when you collapse it to a single label, because that hides the rival drifting from medium to high.

**Momentum** proves the rival is expanding rather than coasting. Headcount growth is the most-cited proxy - a company that grew from 50 to 150 employees in 24 months is signaling expansion even with no press release. But raw count lies. Score the composition of open roles instead: a 400-person company with 30 open roles is normal; a 40-person company that opened five roles in one department in three weeks is not.

**Resourcing** proves the rival can fund a move against you. Pull it from funding rounds and, for public rivals, the most recent 10-K or 10-Q for revenue, growth rate, gross margin, and operating loss. It lies when a large cash balance masks a company with no intent to spend it in your segment.

**Trajectory toward your segment** proves the rival is coming for your lane specifically. A jobs search shows which functions a company is staffing, which is the closest public view of its roadmap. The classic tell is an enterprise player launching a "light" version for SMEs, visible first as sales roles targeting mid-market buyers.

**Reactivity** proves the rival will actually respond to your moves. This is the Porter's Four Corners contribution: the model develops a profile of likely strategy changes and determines each competitor's probable reaction to industry shifts. Use the Actions corners - current strategy and capabilities - because they show steps you can observe first-hand. It lies when you score a rival high because it *could* react; grade only what it has visibly done.

> **Note:** Reactivity is underused, which makes it a differentiator
>
> In a 2005 SCIP survey of frequently used analytical tools, Porter's Four Corners did not figure in the top ten. A fast-moving rival that never reacts is a lower threat than a slow one that always does, so scoring reactivity from observable Actions is an edge most teams skip.

## How to weight the dimensions

Assign weights that sum to 1.0 before you score anything, because the weighting choice, not the individual scores, decides the tier. This is the mechanic borrowed from GE-McKinsey and key-success-factor practice: pick five to seven factors that matter, assign weights summing to 1.0, score each on a 1-to-10 scale, and calculate a weighted total. A Bain-experienced practitioner puts the common range at six to eight factors.

The reason weights come first is reproducibility. The same factor scores produce different totals under different weights, so two analysts agree only if the weights are locked. Fix them to your strategic question. If your near-term risk is a rival displacing you in your core segment, overlap and trajectory carry more weight than raw momentum. If your risk is a well-funded entrant, resourcing rises.

**Threat-tier scoring rubric (weights sum to 1.0)**

```
Dimension                     Weight   Score(1-10)   Weighted
Market/segment overlap        0.30     __            __
Momentum                      0.20     __            __
Resourcing                    0.15     __            __
Trajectory toward segment     0.20     __            __
Reactivity                    0.15     __            __
                              -----                  -----
Total weight                  1.00     Weighted total: __

Tier bands (on a 1-10 weighted total):
  7.0 and above  = Deep-monitor (weekly)
  4.0 to 6.9     = Watch (monthly)
  below 4.0      = Ignore (quarterly scan)
```

*Set the weights first, from your strategic question, then score each rival 1-10. Adjust weights, not scores, to reflect your objective.*

The bands above are a starting calibration, not a law of nature. Set them so the deep-monitor tier stays small enough to review weekly, and revisit the thresholds after your first calibration round.

> **Rule:** Lock weights before you score
>
> Derive and freeze the weights from your strategic question before grading any rival. Weights that do not reflect the objective let a rival post a high generic total while posing no threat to your specific segment.

## Reading momentum from public signals

Read momentum from job-posting velocity first and net headcount second, because postings lead hires by one to three months. Postings are funded commitments visible before anyone starts, so a cluster is the earliest honest read on where a rival is putting money.

The strongest evidence for this ordering is the contrast within a single company. Momentum Group's headcount grew 0.6% to 14,223 employees in one year - flat - while its active job postings rose 68.4% year over year. Reading the headcount alone, you would call it stalled. Reading the postings, you catch a re-acceleration a quarter early.

**68.4% - Momentum Group's year-over-year rise in active job postings, against 0.6% headcount growth**

Posting velocity is a leading indicator; net headcount confirms it one to three months later.

There is no publicly agreed annualised headcount band that separates "accelerating" from "stalling," so treat the numbers as directional. A stable, mature company does not need double-digit growth - single-digit is normal for it - which implies double-digit annualised headcount growth reads as accelerating. One trading-signal implementation flags hiring momentum with a 15% growth threshold over an 8-week lookback and treats job data as a one-to-three-month leading indicator on revenue. Borrow the shape, not the exact constant.

#### Momentum signals, earliest to most confirmed

| Stage | Figure | Note |
| --- | --- | --- |
| Job-posting spike | leads by 1-3 months | role clusters, not raw count |
| Net headcount delta | confirming | filters out churn-driven postings |
| Revenue / filings | lagging | 10-K/10-Q for public rivals |

*The earliest read is posting velocity; net headcount and revenue confirm it later.*

Because there is no fixed threshold, write down the band you use and re-check it against your own market's base rate each quarter. Where senior-hire inflow matters - a new CMO, CRO, VP Sales, or RevOps - treat it as a distinct momentum signal, not folded into the headcount count.

Ask me this: `Companies posting sales roles targeting mid-market buyers in the segment we sell to, that previously only served enterprise` - [run the search](https://www.refolk.ai/start?q=Companies%20posting%20sales%20roles%20targeting%20mid-market%20buyers%20in%20the%20segment%20we%20sell%20to%2C%20that%20previously%20only%20served%20enterprise).

*Returns rivals whose job postings reveal a downmarket trajectory into your lane before any product launch is announced.*

Pulling that trajectory signal by hand means scraping careers pages across a set of rivals and reading each posting for its target buyer. [Refolk](/) runs the same read as a plain-English query across the public professional graph, so you can grade the trajectory dimension in minutes instead of an afternoon of tab-switching.

## The scoring procedure

Run the tier as an eight-step procedure. Steps one through four produce the score; steps five through eight make it reproducible and keep it current.

#### Score a competitor into a tier

1. **Define the set and weighted dimensions** - Fix five to seven publicly observable dimensions - overlap, momentum, resourcing, trajectory, reactivity - and assign weights summing to 1.0 from your strategic question.
2. **Grade each dimension from public evidence** - Score overlap on the similarity-by-segment grid, momentum from headcount and posting velocity, resourcing from funding and filings; attach a source link to every score.
3. **Compute the weighted total and set a provisional tier** - Multiply each score by its weight and sum to one number that maps to deep-monitor, watch, or ignore.
4. **Apply the Four Corners reactivity check** - Test whether the rival is likely to actually react using observable Actions corners plus low-confidence inferred motivation, then confirm or adjust.
5. **Calibrate with a second analyst** - Have a second analyst grade the same rival independently against the locked weights and reconcile to one tier or a documented reason for divergence.
6. **Assign the matching cadence and owner** - Deep-monitor weekly, watch monthly, ignore quarterly, potential entrant semi-annual; give each rival a named owner and next-review date.
7. **Wire off-cycle triggers** - A funding round, leadership change, or major launch forces a re-score within 48 hours; wire alerts to those three event types.
8. **Re-score on cadence** - Run a quarterly deep dive across the full set for strategic re-assessment, threat re-ranking, and win/loss review, then log the update.

Step four carries a genuine disagreement in the sources: some treat reactivity as one of the weighted dimensions, others apply it as a gate after scoring. Pick one and document it. Because Four Corners is underused and inferred motivation is low-confidence, I score reactivity from observable Actions only and apply it as an adjustment that can demote a rival that scores high on overlap and momentum but shows no pattern of reacting.

## Mapping tier to review cadence

The tier's real payload is its cadence, not its depth. Cadence, not analysis depth, is what drives CI value: soliciting sales feedback weekly correlates with strong CI adoption at 69% versus 22% for teams that do not. So the point of tiering is to set how often you look.

Run three cadences in parallel rather than one uniform rhythm: a full refresh quarterly, a pricing-and-feature delta monthly, and a sales-ready snippet weekly, with major moves forcing an out-of-cycle update within 48 hours. Map the frequencies to tiers as below.

| Tier | Review cadence | What is tracked |
|---|---|---|
| Deep-monitor / direct | Weekly | Pricing, features, reviews, hiring, messaging |
| Watch / indirect | Monthly | Product expansion, positioning, funding |
| Ignore / replacement | Quarterly | Tech trends, adoption curves, category shifts |
| Potential entrant | Semi-annual | Job postings, M&A, roadmap signals |

Two adjustments. First, match the market's speed: e-commerce needs daily attention while enterprise software works on weekly or biweekly cycles, so scale the whole table to your category's clock. Second, emerging entrants require a separate scanning cadence, quarterly at minimum, because they lack the public data footprint to score on the rubric at all.

#### From score to review loop

1. **Weighted total** - maps to a provisional tier
2. **Tier** - selects weekly, monthly, or quarterly cadence
3. **Owner + review date** - assigned so the loop cannot go stale
4. **Off-cycle trigger** - funding, leadership, or launch forces a 48h re-score

*The weighted total selects a cadence, and events can jump a rival forward at any time.*

> **Watch out:** Uniform cadence turns intelligence into noise
>
> Strategic reports produced monthly become noise, and tactical flash reports produced only annually miss every event they exist to address. The cadence must differ by tier, and off-cycle triggers must sit on top of it.

## How this goes wrong

The rubric fails in predictable ways, and knowing them is more valuable than the rubric itself. Each failure below is a false positive or a stale verdict, with a check that catches it.

**Headcount growth without net growth.** Companies with high hiring activity but flat or negative employee growth are churning, not expanding - a retention problem read as momentum. Momentum Group's 0.6% net growth against a 68.4% posting jump can cut either way, so check the net headcount delta, not the posting count alone.

**Landing-page momentum.** A landing page can claim momentum; a payroll cannot. A rebrand or a funding press release scored as trajectory is a false positive. Require a payroll, headcount, or shipped-release corroborant before you move the trajectory score.

**Binary overlap labels.** Treating direct versus indirect as yes or no hides the dangerous drifter - an indirect rival scored "ignore" while it repositions into your lane. Re-grade the segment-overlap axis every quarter so the drift shows up as a trajectory signal.

**Uniform cadence.** Covered above: a single rhythm applied to every tier either buries you or misses events. Cadence must differ by tier.

**Over-indexing one source.** A common failure is over-indexing on one source, usually search rankings, and missing signals that live elsewhere. Notably, the most-cited CI source is employee knowledge, internal docs, and call recordings at 54%, ahead of competitor websites at 48%. Confirm pricing pages, job posts, ads, and reviews are all covered.

**Weights that do not reflect the objective.** A weighted scorecard turns dimensions into one number, but the weights must reflect the objective. A rival can score high on generic weights while posing no threat to your specific segment. Re-derive weights from your strategic question, not a template.

**Reactivity assumed, not observed.** Scoring a rival high because it *could* react inflates the top tier. Use only observable Actions corners; because Four Corners is not a top-ten-adopted tool, treat inferred motivation as low-confidence.

**Stale tiers.** The quarterly deck looked sharp in January; by April three rivals had changed packaging and sales was still quoting the January slide because nobody owned the loop. Every tier needs a named owner and a next-review date.

#### Where a rival lands, and what to do

Horizontal axis runs from Low overlap to High overlap. Vertical axis runs from Low momentum to High momentum.

| Quadrant | What it means |
| --- | --- |
| Emerging watcher | Scan quarterly; may lack a public footprint to score |
| Rising threat | Deep-monitor weekly; check trajectory into your segment |
| Dormant peer | Ignore, but re-grade overlap axis each quarter for drift |
| Coasting incumbent | Watch monthly; reactivity decides if it can still move |

*Overlap and momentum place the rival; reactivity and trajectory adjust the verdict.*

## Before you commit a tier

Verify the rubric held before you publish a verdict and set a cadence. The checklist below is the gate between a graded call and a gut sort dressed up as one.

#### Tier sign-off

- [ ] Weights were fixed from the strategic question before any dimension was scored, and they sum to 1.0.
- [ ] Every dimension score has a source link, and momentum uses posting velocity as well as net headcount.
- [ ] Overlap was graded on the two-axis grid, not as a direct/indirect label.
- [ ] Reactivity was scored from observable Actions, with inferred motivation flagged as low-confidence.
- [ ] A second analyst reached the same tier, or the divergence is documented.
- [ ] The rival has a named owner, a cadence matching its tier, and a next-review date.
- [ ] Off-cycle triggers for funding, leadership change, and launch are wired to this rival.

## Keeping the tiers current

Tiers decay, so treat re-scoring as the job, not the deliverable. Quarterly deep dives handle the full strategic re-assessment, threat re-ranking, and win/loss review across the entire set; between them, each tier runs its own cadence and off-cycle triggers catch the moves that will not wait.

The scarce resource in all of this is analyst time, and it is scarcer than the market assumes. In Refolk's index of professional profiles, the United States holds 245 people with Competitive Intelligence Analyst or Manager titles - concentrated at HP, ADP, Apple, PwC, and IBM - while the United Kingdom returns just 23, a US pool roughly 10.7 times the UK's. US CI leadership titles number only 18, concentrated in New York.

| Segment | People | Top employer signal |
|---|---|---|
| US, Analyst + Manager | 245 | HP, ADP, Apple, PwC, IBM |
| UK, Analyst + Manager | 23 | Frazer-Nash, Logitech, Dataiku |
| US, Head/Director/VP | 18 | Salesforce, Dell, BMS, Glean |

The thinness of that bench, especially outside the US, is the strongest argument for a written rubric with locked weights: it lets a non-specialist apply the standard and reach the same tier a senior analyst would, and it survives the analyst who built it moving on. To benchmark how rivals resource their own watching function, Refolk can pull competitive-intelligence leads and analysts at comparable firms so you can see who is staffing up to watch you back.

Re-check three things on every cycle. Re-derive weights if your strategic question has shifted. Re-grade the overlap axis for every "ignore" and "watch" rival to catch drift. And re-confirm your momentum band against your own market's base rate, since no public standard fixes it for you. A tier that is never re-scored is a January slide, and by April it is quietly wrong.

## Frequently asked questions

### How many competitors should be in the deep-monitor tier?

The rubric decides, not a quota, but keep the tier small enough to sustain weekly review. Deep-monitor demands weekly tracking of pricing, features, reviews, hiring, and messaging, which is expensive analyst time. If your weighted total pushes more than a handful of rivals into deep-monitor, re-check whether your weights actually reflect your strategic question, since generic weights inflate the top tier.

### Should reactivity be a weighted dimension or a separate gate?

Sources disagree, so decide explicitly and document it. Some treat reactivity as one of the five to seven weighted dimensions; others apply it as a gate after scoring, demoting a rival that scores high on overlap and momentum but shows no observable pattern of reacting. Because Porter's Four Corners is underused and inferred motivation is low-confidence, I favour scoring reactivity from observable Actions only and applying it as an adjustment.

### How often should I re-score the full competitive set?

Run a full quarterly deep dive across every rival for strategic re-assessment, threat re-ranking, and win/loss review, regardless of tier. Between quarters, each tier runs its own cadence - weekly, monthly, or a quarterly scan - and any funding round, leadership change, or major launch triggers an out-of-cycle re-score within 48 hours. Stale tiers are a common failure, so give every rival a named owner and a next-review date.

### Why not just sort competitors into direct and indirect?

A binary label hides the dangerous drifter: an indirect rival repositioning into your segment scores 'ignore' under a yes/no scheme right up until it lands in your lane. Grade overlap on two axes instead - product similarity and customer-segment overlap, each low, medium, or high - and re-grade the segment axis every quarter so a rival launching an SMB version of an enterprise product shows up as trajectory before it becomes a direct threat.

### What single signal is the earliest read on a competitor gaining momentum?

Job-posting velocity, because it leads net headcount by one to three months. Postings are funded commitments visible before hires land, so a cluster of five roles in one department in three weeks signals compressed urgency that employee-count curves will only confirm later. Momentum Group illustrates the gap: 0.6% headcount growth against a 68.4% year-over-year rise in active postings.

---

*From the Refolk guide library. I revise these guides rather than replacing them, so the current version is always at https://www.refolk.ai/guides/competitive-threat-tier-rubric*