Get Your Red Pen Ready
By Lieutenant Colonel Matthew J. Fontaine
Article published on: July 1, 2024 in the July–December 2024 Issue of Military
Intelligence
Read Time: < 23 mins
Introduction
Everyone knows the S-2/G-2 must provide only two enemy courses of action (COAs) during the mission analysis
brief, and it is a job well done. These two COAs, the most likely and most dangerous, provide the commander and
staff with everything they need to know about how the threat will fight against friendly actions throughout the
execution of a complex operation, right? Sure, we know that doctrine asks us to “identify the
full set of courses of action available to the threat,”1 but who does that?
The intelligence cell should do that, of course. This article recommends that intelligence cells continuously
develop and refine three categories of enemy COAs, instead of the two standard most likely and most dangerous
enemy COAs, to better account for the tactical and operational options available to a thinking enemy in
large-scale combat operations.2 Better enemy COAs result in better friendly plans.
Better plans result in friendly forces more likely to seize opportunities or avoid disaster during the execution
of operations. So, get your red pen ready! The three enemy COA categories are:
- Operational enemy COAs.
- Critical event enemy COAs.
- Transition enemy COAs.
Now, don't get me wrong. The intelligence cell must still designate the most likely and most dangerous enemy
COAs, but they must do so for each category. Identifying the most likely and most dangerous enemy COAs
is essential because they enable the commander to develop optimized friendly plans in environments that are
often time constrained.3 The staff then develops contingency options (think
branches, sequels, or alternate COAs) should the enemy execute any other valid COA available to the
threat for each category.4 (I will discuss the value of the most likely and most
dangerous designations in the context of the three enemy COA categories again near the conclusion of this
article.)
In this article, I will explain why the typical enemy COAs drafted by many intelligence cells do not meet the
challenges of large-scale combat operations. I will then describe each enemy COA category in detail. I will then
conclude with a discussion on developing logical priority intelligence requirements (PIRs) to detect any valid
enemy COA selected by the threat.
The Problem with "Two and Done"
The standard two enemy COAs typically developed by intelligence cells often do not provide the complete
conceptual narrative and details the commander and staff need to create an effective plan. Effective plans
posture the unit to overcome the current enemy challenge and execute critical future transitions, like branches
or sequels, without unnecessary risk.5
Enemy COAs often come up short because they try to provide too much information from the start but do too little
to support the development of effective plans. This tendency is especially true for enemy COAs developed for
command post exercises, large-scale combat operations scenarios at the division and higher echelons, or the
initial COA a brigade creates before an Army combat training center rotation. Here is what I mean: the typical
command post exercise or initial combat training center enemy COA often focuses on how the threat will achieve
its operational or strategic end state from start to finish. I dub this enemy COA the operational enemy
course of action (OECOA, pronounced OH COA); it is an essential COA.6 However, an operational enemy
COA only tells part of the story to the commander and staff. The intelligence cell uses the operational enemy
COA to portray how the enemy could achieve its overall mission and end state. It has limited utility for
developing effective plans for two primary reasons.
First, an operational enemy COA cannot provide the necessary details for good staff work because it must cover so
much ground, literally and figuratively. It is not uncommon for an operational enemy COA in a typical
large-scale combat operations scenario to describe how an enemy division, corps, or even army will execute an
entire operation—from invasion to the destruction of the friendly forces over hundreds or even thousands
of square kilometers! Intelligence cells often describe an operational enemy COA using a single paragraph or
PowerPoint slide. How valuable can this analysis be in supporting detailed, friendly planning?
Second, the intelligence cell drafts operational enemy COAs during mission analysis before developing friendly
COAs. Given this arrangement, operational enemy COAs only consider opposing forces in the most general sense, as
the detailed, friendly plan does not yet exist. From its inception, an operational enemy COA is of limited value
because the intelligence cell did not construct it in relation to friendly actions.
But we know war, particularly large-scale combat operations, must be considered from both friendly and enemy
perspectives. Carl von Clausewitz imagined war as a match between two wrestlers: “Each [wrestler] tries
through physical force to compel the other to do his will; his immediate aim is to throw his
opponent in order to make him incapable of further resistance.” 7 Clausewitz's analogy evokes
an image of a violent fight. One sees two competitors locked in a fierce back-andforth struggle to gain
advantage before one side imposes its will in a final tremendous effort to emerge victorious.
Our enemy COAs should account for this dynamic nature of war–but often, they do not. Instead, many
intelligence cells develop operational enemy COAs without understanding the friendly plan. And how could this
not be the case? S-2s/G-2s present enemy COAs during mission analysis before friendly COA development. Look at
many operational enemy COAs (and friendly COAs, for that matter) to see how little we take our opponent's
actions into account. Most enemy COAs have a few blue opponent icons or tasks at the end of a sequence of enemy
steps as if the friendly forces were just along for the ride. Some include no friendly icons or tactical tasks
at all!
Developing a friendly COA with just an operational enemy COA is akin to a wrestler preparing for a live opponent
based solely on a session with a wrestling dummy.8 Like wrestling against a dummy, typical operational
enemy COAs provide no sense of the dynamic reactions and counteractions necessary to spur commanders' and
staffs' thinking on how best to design a friendly operation considering the complete set of options
available to a threat. The limited utility of operational enemy COAs becomes readily apparent after being
briefed during mission analysis. The S-2/G-2 receives a deluge of “how” questions from the commander
and staff: how will the enemy react to a particular aspect or critical event of a friendly COA? How long will an
enemy transition take? And so on.
I know what you think; the war game will address many of these questions. After all, the purpose of the S-2/G-2
during the war game is to project how the enemy will react to the friendly COA, including its constituting
critical events.9
But do they always? And how far do staff get during war gaming in time-constrained environments or under
demanding conditions (if the war game even happens)? If a war game does occur, is the S-2/G-2 prepared to
execute the basic war-gaming “action, reaction, and counteraction methods of friendly and enemy forces
interaction”10 for every critical event described in
doctrine, armed with only a most likely and most dangerous operational enemy COA? I am not convinced based on my
experience. Many of us have been guilty of using the same enemy COA paragraph or PowerPoint we initially showed
during mission analysis for the operations order at the end of the military decision-making process (MDMP). We
all agree that this should not occur if the command and staff truly wrestle with the problem presented by the
enemy.
Suppose you are an S-2 or G-2 that comprehensively updates the intelligence preparation of the operational
environment products at the end of war-gaming. (There should be significant changes, correct?) In that case, you
are ahead of the game!
Whether or not you update your products, don't you wish you had more depth in your enemy COAs before the war
game so as not to provide shallow responses or, to put it politely, baloney? Or that you had better enemy COAs
during mission analysis to give the commander and staff a better starting point for developing more
comprehensive friendly COAs, branches, and sequels, with the idea that better input—both friendly and
enemy—will result in a better war game output?
One thing is sure: no G-2 or S-2 wants to find themselves at a war game considering for the first time the
enemy's reaction to some critical event, like a wet gap crossing! The enemy fights to win in large-scale
combat operations and will use every technically and tactically ingenious method to prevail. We must think
deeply to beat these opponents, so broad enemy COAs alone will not cut it. The solution to providing better
enemy COAs—and better friendly COAs as a result—is to start with the big picture.
Operational Enemy Courses of Action
The first set of enemy COAs to develop are the operational enemy COAs. Even though I just seemingly maligned
them, creating quality operational enemy COAs is essential to understanding the threat from a complete narrative
perspective. Operational enemy COAs describe how the enemy might achieve its desired operational or strategic
end state from start to finish, arranged along a line of operation (LOO). It is a conceptual product that lets
the staff visualize how an enemy operation could evolve holistically. Operational enemy COAs are also essential
to anticipating how enemy forces can enter and exit the uni's deep area or flanks (a vital aspect for
targeting and intelligence handover line coordination). The intelligence cell derives the operational enemy COAs
from the enemy COAs developed by its higher headquarters. 11
For example, a division's analysis and control element derives its operational enemy COA from the corps enemy
COAs, with a slight emphasis on the forces templated in the division's area of operations (AO). I
say slight because the purpose of the operational enemy COA is to gain a holistic understanding of the big
picture, so focusing just on one's own AO misses the point, potentially obscuring how the enemy in one's
area of interest (AOI) could present a risk to the mission or forces. Operational enemy COAs are the first enemy
COAs presented during mission analysis and serve as the foundation for all future enemy COA development.
Importantly, if an intelligence cell has no higher enemy COAs on which to base its operational enemy COAs, that
cell must produce them. If a unit disagrees with the enemy COAs of the higher team, it cannot simply change or
ignore them. To do so would contravene the necessity of having a common understanding of the threat. Instead,
every intelligence cell must collaborate through both staff and command channels to arrive at a common
understanding of the threat with their higher headquarters before moving on with planning.
Figures 1–5 provide simplified examples of higher echelons' enemy COAs and associated operational enemy
COAs developed by the fictional YOUR UNIT. Ideally, the intelligence cell would produce multiple operational
enemy COAs, each nested within the higher echelon's read of the situation.
Operational enemy COAs frame the possible range of valid enemy COAs to include the most likely and most dangerous
available to the threat based on the friendly's understanding of the enemy's mission, intent, key tasks,
and end state within the AO and AOI. As mentioned, this is usually as far as intelligence cells get at the start
of any large-scale combat operations scenario, but we know more is needed. Therefore, the next step is to
develop more detailed enemy COAs. Key to this is understanding the likely critical event of a given LOO.
Figure 1. Higher Echelon Enemy Course of Action One: Heavy North12
Figure 2. Higher Echelon Enemy Course of Action Two: Heavy South13
Figure 3. Operational Enemy Course of Action One: Heavy North, Southern Fix14
Figure 4. Operational Enemy Course of Action Two: Heavy South, Southern Fix15
Figure 5. Operational Enemy Course of Action Three: Heavy South, YOUR UNIT Defeated16
Critical Event Enemy Courses of Action
The second set of enemy COAs to develop is what I will dub critical event enemy COAs (CECOA, pronounced
“SEE COA”). Critical event enemy COAs are like the “snapshots in time” situation
template described in doctrine that represent a “potential threat COA as part of a particular threat
operation.”17 Like a situation template, a critical event enemy COA
describes how the enemy might achieve its desired tactical end state in pursuit of its
operational end state. The difference between a critical event enemy COA and a situation template is
that a critical event enemy COA emphasizes the anticipated enemy COA's relationship to the
anticipated friendly COA or action during a specific critical event. Critical event enemy COAs are detailed
products that enable the staff to visualize the separate ways (actions, reactions, and counterreactions) the
enemy will seek to gain the advantage (win) during a particular portion of a LOO given a friendly action.18 Critical event
enemy COAs ensure we approach enemy COA development from the back-and-forth perspective of Clausewitz's
wrestlers.
For example, imagine the LOO in figure 6 (on the next page) associated with a simple operational enemy COA and a
failure operational enemy COA (more on failure COAs later). The example LOO has three critical events
with these possible friendly actions and enemy counteractions:
- Critical Event 1.
- Friendly Action: Seize OBJECTIVE ONE (Capital City).
- Enemy Counteraction: Defend OBJECTIVE ONE.
The enemy can defend OBJECTIVE ONE broadly via a maneuver defense (CECOA 1 for CE 1) or an area defense to retain
the capital (CECOA 2 for CE 1).
- Critical Event 2.
- Friendly Action: Execute Wet Gap Crossing.
- Enemy Counteraction: Defeat Wet Gap Crossing.
The enemy can defend key crossing sites of the wet gap via an area defense (CECOA 1 for CE 2) or, broadly, via a
maneuver defense (CECOA 2 for CE 2).
- Critical Event 3.
- Friendly Action: Seize OBJECTIVE TWO.
- Enemy Counteraction: Defend to retain OBJECTIVE TWO.
(CECOA 1 for CE 3) or retrograde (CECOA 2 for CE 3).
This is simple stuff. The intelligence cell develops multiple initial critical event enemy COAs for each
critical event to present during the mission analysis brief that they refine throughout the MDMP. The S-2/G-2
designates each critical event enemy COA as the most likely, most dangerous, or some other valid enemy COA for
that critical event. The result is that the S-2/G-2 will develop the most likely and most dangerous critical
event enemy COAs (and other valid critical event enemy COAs) for the most likely operational enemy COA, and the
same goes for the most dangerous critical event enemy COA (and other valid operational enemy COAs). Given the
already high demands on an intelligence cell for the mission analysis brief, this is a tall order, but it will
pay dividends. If not possible, the intelligence cell should begin developing or refining critical event enemy
COAs immediately after the mission analysis brief as the friendly plan takes form. The enemy critical events
will likely be a mirror image of the friendly anticipated critical events.
Figure 6. Simplified Line of Operations19
How can the S-2/G-2 recognize friendly actions during mission analysis to develop the initial critical event
enemy COAs when COA development has not started? My advice is not to overthink the initial critical event enemy
COAs. If a piece of key terrain is essential enough for the enemy to defend it, friendly forces will likely have
to seize it. If a large river flows through the AO, both sides may have to cross it. Therefore, it becomes
necessary for the intelligence cell to describe how enemy forces would react to friendly actions at that
objective if the staff is to build an effective plan. Hopefully, in their initial planning guidance, the higher
echelon's order or the unit commander will identify or at least indicate likely critical events for the
unit. If not, ask the commander and staff to get thinking!
If prepared correctly, the critical event enemy COAs will supercharge friendly COA development after mission
analysis. The staff will better understand the risk to the mission and force during critical aspects of the
overall operation from the start within the context of the general enemy's operation, thanks to the
operational enemy COA. Critical event enemy COAs also focus on the detailed planning of all warfighting
functions in more concrete situations within the larger enemy and friendly picture. For example, given a
particular critical event enemy COA, a member of the protection cell will have to think deeply about how to
shield forces during a wet gap crossing while staying nested within the general scheme of protection for the
overall friendly COA, which itself is designed to account for the most likely operational enemy COA. Detailed
planning like this is essential to understanding and mitigating risk.
Critical event enemy COAs are the intelligence cell's primary input to the war game; the war game refines
them. Providing draft critical event enemy COAs during mission analysis ensures that the staff has already had
an opportunity to think deeply about the valid, serious problems the unit will likely encounter during different
portions of the operation, including the most likely and most dangerous ones. Quality critical event enemy COAs
ensure that the COA analysis of the MDMP includes a genuine war game instead of what can be a series of ad hoc
responses to inch-deep tactical dilemmas.
The next step is to consider what happens if the threat fails (or succeeds) in achieving its objectives.
Transition Enemy Courses of Action
The third set of enemy COAs to develop is what I will dub the transition enemy COA (TECOA, pronounced
“TEE COA”). A transition enemy COA anticipates what actions the enemy might take initially if unable
to achieve a critical event on its LOO. Or, a transition enemy COA might envision how an enemy could seize an
opportunity because the enemy completed its assigned objectives at costs below what was anticipated for factors
such as time or battle damage. First, I'll discuss a failure transition enemy COA and second, a
success transition enemy COA.
Failure transition enemy course of action. The failure transition enemy COA describes how the enemy will
attempt to regain the conditions necessary to achieve the current end state described in the operational enemy
COA or a modified end state based on the new battlefield realities.20 Stated another way, the failure
transition enemy COA describes how the threat will transition from a state of relative disadvantage to
a situation of relative advantage to the friendly force.21 The Save Face COA in figure 6 is an example of a
transition enemy COA at the operational level. The enemy sought to seize OBJECTIVE ONE but transitioned to
retaining OBJECTIVE TWO when it could not. Remember: the enemy constantly fights to win, and our enemy COAs must
always reflect this.
Failure transition enemy COAs are issued with their respective operational enemy COA or critical event enemy COA.
They enable the commander and staff to develop success branches and sequels to exploit the threat's
momentary failure before they shift to a failure transition enemy COA.
Figures 7 and 8 (on the next page) provide simplified examples of transition enemy COAs developed by the
fictional YOUR UNIT. Ideally, the intelligence cell would produce multiple transition enemy COAs for each
operational enemy COA and critical event enemy COA while also designating the most likely and dangerous
instances.
Consider a wet gap crossing for another example of the power of sequel planning thanks to a quality transition
enemy COA. A high-performing intelligence cell presents an initial wet gap crossing critical event enemy COA
during mission analysis to kick off detailed, friendly planning for this event. At this point, staff typically
do one of three things.
One staff only designs a plan for crossing the wet gap against the threat described in the critical event enemy
COA. This isn't bad; it's certainly better than only planning against an operational enemy COA or only
planning for the critical event enemy COA after publishing the base operations order. But, as we will see with
the following staff scenario, a critical event enemy COA only improves a friendly plan by so much.
The second staff war-games the critical event enemy COA and identifies the possibility of a sequel, which leads
to the creation of a new decision point. Something like, Decision Point 1: conduct sequel after wet gap
crossing. Unfortunately, little detailed planning goes into the sequel to increase the odds of success
because the enemy situation becomes too murky at this point. As a result, the unit culminates after crossing the
river during execution and watches as the enemy retrogrades, unable to exploit their initial success. In other
words, even with a decision point, this staff mainly reacts to the enemy situation as it emerges. It cannot
effectively sequence its actions to maintain pressure on the enemy.
Figure 7. Failure Transition Enemy Course of Action One: Key Terrain One and Wet Gap
Defense22
Figure 8. Success Transition Enemy Course of Action Two: Friendly River Defeat23
Here is where things get interesting. A third staff receives a wet gap critical event enemy COA and a
failure transition enemy COA. The commander and staff listen with great interest as the S-2/G-2
explains how, if provided the opportunity and with sufficient remaining combat power, the threat will withdraw
(transition) to new defensive positions if defeated at the wet gap and immediately dig in to establish a
deliberate defense in as little as 24 to 36 hours. The commander tells the operations officer that the unit
cannot allow the enemy to establish a deliberate defense. As a result, the staff designs a detailed sequel that
prevents the unit from culminating after the wet gap crossing and disrupts enemy defensive preparations in the
unit's deep areas. The proactive friendly plan prevents the enemy from regaining the advantage during
execution.
Success transition enemy course of action. The success transition enemy COA (figure 8) describes how the
enemy will exploit unanticipated success to achieve its current end state, as described in the operational enemy
COA, or a modified end state based on the improved battlefield realities for the threat. Stated another way, the
success transition enemy COA describes how the threat will transition from a state of relative
advantage to a situation of greater relative advantage to the friendly force.
As with failure transition enemy COAs, success transition enemy COAs are issued with their respective operational
enemy COA or critical event enemy COA. They enable the commander and staff to develop failure branches and
sequels to mitigate or completely head off the friendly unit's momentary disadvantage (failure) before the
threat fully executes its success transition enemy COA.
Consider again the third staff in our failure transition enemy COA wet gap critical event enemy COA. After
getting a pat on the back from the commander for their insights on the enemy failure transition enemy COA, the
S-2/G-2 asks to brief one more slide. Again, the command and staff listen with great interest to how the threat
may commit additional ground, fires, and aviation assets to defeat the wet gap crossing in the unit's AO.
The S-2/G-2 then demonstrates how the threat could exploit this opportunity to transition to the offense with
the enemy's operational reserve in a follow-and-support role to defeat the unit. To mitigate this
worst-case scenario, the staff designs a more robust COA with enhanced levels of protection and combat power at
the crossing sites, coordinates for higher echelon fires to disrupt the movement of the operational reserve in
the AO, and designs a whole new sequel COA that rapidly transitions the unit to a deliberate defense.
Quality transition enemy COAs ensure that a friendly unit can seize opportunities and weather setbacks. The next
step is to ensure the team has a collection plan to detect these operational enemy COAs, critical event enemy
COAs, and transition enemy COAs.
Priority Intelligence Requirements Development
We know that uncertainty and ambiguity are unavoidable qualities of war.24 Returning to the wrestler
analogy, no wrestler goes into a match believing they know precisely how a contest will play out. Sure, a
wrestler has a plan, given what they know about their own and their opponent's strengths and weaknesses,
but if their plan is off track, they change their approach if they want to win. The wrestler expects their
opponent to do the same. Notably, some wrestlers may only commit to an initial approach once the match
begins, when both opponents first receive cues about what the other might do. However, as a rule, we expect
those wrestlers who have planned and prepared for alternate approaches to win more matches.
We can draw three simple points from the wrestling analogy to inform the development of PIRs. First, planning and
preparation are essential even in uncertain environments.25 Second, the uncertain nature of war is partially due to
the sometimes unpredictable outcomes resulting from the clash or posturing of forces. We cannot predict with 100
percent accuracy how we or our opponent will react or counteract in a situation. Third, because both sides
approach war knowing that they select COAs based on many friendly, enemy, and environmental factors, it would be
foolish to assume that our opponent has already determined what they will do from the start. Instead, the enemy
may keep their options open for as long as possible. In other words, we cannot tell what COA our opponent will
pick with complete certainty from the get-go because the enemy may still need to commit to a decision or may
transition to an alternate COA partway through execution. Because forces react unpredictably and can make
decisions based on a wait-and-see attitude, units must develop collection plans that constantly scan the
environment for multiple enemy COA possibilities. Commanders and staff cannot simply pick one enemy COA and
ignore the rest–or they do so at their peril.
Staff must design PIRs to determine what the enemy is doing now (critical event enemy COA), next (transition
enemy COA), and within the big picture (operational enemy COA) to reduce the unavoidable uncertainty of war.
PIRs ensure units use their scarce collection assets to answer the commander's most important questions.26 What else would
be worth prioritizing our limited collection assets against than determining what COA the enemy is undertaking
or will undertake (besides support to targeting to enable our selected COA)? Nothing in my mind.
The description of a PIR offered in FM 3-0, Operations, supports this reasoning. FM 3-0 states that PIRs
“identify information about the threat and operational environment that a commander considers most
important to making decisions in a specific context.”27 Certainly, the set of enemy COAs described in this
paper qualifies as requiring friendly decisions!
However, the straightforward process of drafting PIRs to identify which enemy COAs the enemy selects breaks down
too often. Many PIRs (even well into execution) often say nothing of enemy COAs at all and instead use generic
or unhelpful statements like:
- PIR 1: Where will the enemy employ its reconnaissance?
- PIR 2: Where will the enemy employ its fires?
Knowing where the enemy reconnaissance or fire assets are located is beneficial, but why? Read on.
Let's return to our wet gap crossing example. Recall that our third staff worked through the whole gamut of
enemy COAs. The intelligence cell prepared critical event enemy COAs for the threat's anticipated defense: a
failure transition enemy COA if the threat could not defeat the friendly crossing operation and had to withdraw,
and a success transition enemy COA if the enemy defeated the crossing operation and transitioned to the offense.
To provide support for the commander's decision making, a better set of simplified PIRs for this phase of the
friendly operation might look like this:
- PIR 1: Will the enemy conduct an area defense (critical event enemy COA 1) or maneuver defense (critical
event enemy COA 2) to oppose a friendly wet gap crossing? Friendly Decision: Execute a COA to defeat the
most likely critical event enemy COA 1 option with a contingency option should the threat adopt the second,
less likely COA. (This PIR may be broken into two separate requirements).
- PIR 2: Is the enemy transitioning to defensive operations east of the wet gap (failure transition enemy
COA)? Friendly Decision: Pursue withdrawing forces and disrupt defensive preparations in depth (success
sequel COA).
- PIR 3: Is the enemy transitioning to offensive operations west of the wet gap (success transition enemy
COA)? Friendly Decision: Execute a defense west of the wet gap (failure sequel COA).
Instead of looking for reconnaissance or fires assets as the sole purpose of collection as we did in the first
sample set of PIRs, these examples focus collection efforts to broadly identify what the enemy is doing (enemy
COA). Collection still looks for fires and reconnaissance assets, in addition to other critical systems and
activities, but now they serve as indicators to support the assessment of which COA the enemy is executing. Next
(or concurrently, if possible), the unit focuses collection via additional PIRs to target assets on the high
payoff target list, which enables the execution of the optimized friendly COA. Too often, the tendency is to
jump right into targeting without understanding what the enemy is trying to do as a combined arms team, both
operationally and tactically.28 The collection approach represented in the second set
of PIRs fixes that.
For any operation, a generic PIR framework that considers the uncertainty inherent to large-scale combat
operations would look like figure 9 (on the next page). While seemingly complicated, it has clear advantages
over the standard two-and-done enemy COAs often generated at the start of many large-scale combat operations
scenarios. Figure 10 (on the next page) suggests how to keep the number of PIRs more manageable throughout an
operation's execution.
Figure 9. Priority Intelligence Requirement Framework29
Figure 10. Managing Priority Intelligence Requirements by Phase30
Return to the Most Likely and Most Dangerous Enemy Course of Action
How do you position friendly planning efforts against all these enemy COAs? Here is where the most likely and
most dangerous labels return to the picture. S-2s and G-2s evaluate and prioritize all valid enemy COAs within
the three categories.31 Prioritization is essential for two reasons. First, as
discussed at the beginning of this article, prioritization ensures that most planning time is devoted to
developing the most likely and most dangerous enemy COAs when time is limited.32 Second, prioritization enables
the staff to develop a single friendly COA “optimized to counter the most likely threat COA,
while allowing for contingency options should the threat choose another COA.”33 So, if we do our enemy COA
development correctly, we wind up with one very resilient friendly COA with the necessary number of contingency
options to account for every valid enemy COA in our three categories over an entire operation. This optimized
COA is far preferable to a friendly critical event COA that does not take the big picture into account or an
operational COA that lacks the details of the tactical situation, with just a single contingency option to
account for the most dangerous enemy COA.
Conclusion
Intelligence cells must commit to determining the complete set of valid enemy COAs to support effective decision
making in the uncertain conditions of large-scale combat operations. Drafting operational enemy COAs guarantees
we never lose sight of the big picture. Operational enemy COAs serve as the basis for all future COA
development. Critical event enemy COAs ensure we execute detailed planning on the areas that matter most.
Transition enemy COAs force us to consider what happens next and account for dangerous what-if scenarios. We
leverage this understanding, gained during planning, to recover or gain an advantage in every valid situation
during execution. The three enemy COAs acknowledge that the enemy and friendly have a vote and incorporate this
dynamic into their narratives.
The inescapable result of the recommendations in this article is that the staff will make many enemy COAs and
friendly branches and sequels. That's okay. Staff need to adopt the view that COA development is never
finished. Once the team wrestles with one COA, they move to understand the operational, critical event, and
transition enemy COAs tied to the next most likely or most dangerous situation–situations that large-scale
combat operations are guaranteed to produce in abundance. As enemy COAs are updated, so are the PIRs
prioritizing the unit's limited collection assets to determine which COA the enemy will select.
If all these COAs sound too intimidating, start small. Develop a failure transition enemy COA and a success
transition enemy COA to go with the standard most likely and most dangerous enemy COA during the mission
analysis brief. Move to critical event enemy COAs and additional permutations of the three enemy COA categories
as your commander, staff, and you see the benefits that the three categories bring to planning and execution.
So, grab some red pens. You're going to need them!
Endnotes
1. Department of the Army, Army Techniques Publication
(ATP) 2-01.3, Intelligence Preparation of the Operational Environment (Washington, DC: U.S.
Government Publishing Office [GPO], 1 March 2019), 6-5. Change 1 was issued on 6 January 2021. Change 2 was
issued on 23 January 2024. (emphasis added).
2. Department of the Army, Field Manual (FM) 5-0,
Planning and Orders Production (Washington, DC: U.S. GPO, 16 May 2022), 2-1. Please review the
discussion cited here for the distinction between warfare's tactical and operational levels.
3. Department of the Army, ATP 2-01.3, Intelligence
Preparation of the Operational Environment, 6-7.
4. As with friendly courses of action (COAs), valid enemy
COAs must be “feasible, acceptable, suitable, distinguishable, and complete.” Ibid.
5. Many people from my daily duties and professional
military education influenced my thinking for this article. I first gained an appreciation for the value of
an enemy COA beyond the standard most likely and most dangerousduring Intermediate Level Education at the
Command and General Staff Officer Course. The curriculum exposed me to the value of the failure enemy COA. I
developed and implemented those ideas as a brigade S-2. In my Military Intelligence Professional
Bulletin (MIPB) article, “Enemy Course of Action Development,” I captured how failure
and operational enemy COAs helped my team keep pace in a time-constrained environment during a Joint
Readiness Training Center rotation. This article builds upon those ideas along with conversations with my
co-workers (notably Majors Kyle Ferguson, Zachary Lawson, Michael Lapadot, Michael Hoffman, and LTC Patrick
Vogt) and lessons gleaned from my leadership, principally MG John Meyer's thoughts on transition points,
the need for careful development of enemy COA indicators, and the need for detailed (effective) planning to
understand the risk to the force and the mission. Of course, all errors are my own. You can access
“Enemy Course of Action Development” at https://mipb.army.mil/articles/2020-1qtr/fontaine-15oct2019.
6. Matthew Fontaine, “Enemy Course of Action
Development,” MIPB (October-December 2019): 23-24, https://mipb.army.mil/articles/2020-1qtr/fontaine-15oct2019.
I first use the term operational enemy COA in figure 1 of this article. The operational enemy COA's
value, paraphrased in the current paper, is initially described in this article's text as the COA
belonging to the next higher headquarters.
7. Carl von Clausewitz, On War, ed. and trans.
Michael Howard and Peter Paret (New Jersey: Princeton University Press, 1976), 75. (emphasis original).
8. Full disclosure–I am not a wrestler. But
wrestling dummies exist and are helpful in some situations (much as an operational enemy COA can be!). See
Kase Nipe, “Wrestling Dummies in Wrestling: Top Options,” Wrestle Love, https://wrestlelove.com/guides/wrestling-dummies-in-wrestling-top-options/.
9. Department of the Army, FM 5-0, Planning and Orders
Production, 5-51.
10. Ibid., 5-36. The manual states, “each critical
event within a proposed COA should be war-gamed.”
11. Fontaine, “Enemy Course of Action
Development,” 24.
12. Adapted from author's original graphic. The figure
is a simplified example of figure 6-8. Threat course of action statement example in Department of the Army,
ATP 2-01.3, Intelligence Preparation of the Operational Environment, 6-16.
13. Ibid.
14. Ibid.
15. Ibid.
16. Ibid.
17. Department of the Army, ATP 2-01.3, Intelligence
Preparation of the Operational Environment, 6-7.
18.Fontaine, “Enemy Course of Action
Development,” 24. Critical event enemy COAs are akin to the key terrain and population center
microanalysis COAs described in my earlier MIPB article. I now prefer the term critical event enemy
COA better because it encompasses a broader set of situations and includes a friendly plan to consider enemy
counteractions at that event. Key terrain is still valuable, as it can be the starting point for the staff
to identify critical events.
19. Adapted from author's original graphic.
20. Fontaine, “Enemy Course of Action
Development,” 24.
21. Department of the Army, FM 3-0, Operations
(Washington, DC: U.S. GPO, 1 October 2022), 1-3. See this reference for a discussion on the term
relative advantage.
22. Adapted from author's original graphic. The figure
is a simplified example of figure 6-8. Threat course of action statement example in Department of the Army,
ATP 2-01.3, Intelligence Preparation of the Operational Environment, 6-16.
23. Ibid.
24. Clausewitz, On War, 80, 85.
25. Public Papers of the Presidents of the United
States: Dwight D. Eisenhower, 1957: Containing the Public Messages, Speeches, and Statement of the
President, January 1 to December 31, 1957 (Washington, DC: Office of the Federal Register, National
Archives and Records Service, General Services Administration, 1958), 818. President Eisenhower recalled
this fact during a speech to the National Defense Executive Reserve Conference on November 14, 1957.
26. Department of the Army, FM 2-0, Intelligence
(Washington, DC: U.S. GPO, October 2023), 6-7.
27. Department of the Army, FM 3-0, Operations,
3-9.
28. I benefitted from presentations by COL Blue Huber on
the hazards of focusing solely on targeting at the expense of the bigger intelligence picture for this
insight.
29. Adapted from author's original graphic, using
information from Department of the Army, ATP 2-01.3, Intelligence Preparation of the Operational
Environment, 6-7; and Clausewitz, On War, 80, 85. See also Department of the Army, Army
Doctrine Publication 6-0, Mission Command: Command and Control of Army Forces (Washington, DC: U.S.
GPO, 31 July 2019), 3-7, for a discussion on “exceptional information” as the doctrinal broad
category which “contradictory information” would belong to.
30. Ibid.
31. Department of the Army, ATP 2-01.3, Intelligence
Preparation of the Operational Environment, 6-7.
32. Ibid. Prioritizing planning time for the various enemy
COA permutations is critical to success and requires judgment. Even at the division level with a fully
staffed analysis and control element, it is difficult to truly identify, let alone
analyze, each enemy COA available to the threat. In my experience, the teams I have participated in
have successfully developed the most likely and most dangerous operational enemy COAs. It has also been
possible for us to develop the most essential critical enemy COAs post-mission analysis. During mission
analysis, we also successfully promoted the most likely transition enemy COAs for the operational enemy
COAs. I share this to inform the reader that executing the ideal framework proposed in this article is no
small feat! I am still striving for it. But, keep in mind, every added COA we developed brought value to the
unit, and the other extreme alternative, only developing two broad COAs and being done, is far too hazardous
for large-scale combat operations.
33. Ibid.
Author
LTC Matthew J. Fontaine is the G-2 for the 1st Infantry Division, Ft. Riley, KS. He previously served as the
G-2 for the U.S. Army Joint Modernization Command, Fort Bliss, TX. His deployments include two tours to Iraq
and two to Afghanistan, serving as an executive officer, platoon leader, battalion S-2, military
intelligence company commander, and analysis and control element chief. He holds two masters of military art
and science degrees, one in general studies and the other in operational art and science, from the U.S. Army
Command and General Staff College.