The Art of Business · Chapter IV

Chapter IV: Returning Fire

Guides · Art of Business · By The Glasswright

Founder handbookChapter IVThe Glasswright

← Designing the Machine · Series index

The broken bridge

A trading house learned a rival had talked an important carrier into ending the agreement. The founder wanted a public answer. His operations chief asked for one day.

That day she traced the affected shipments, called the customers whose orders were exposed, and arranged temporary carriage at a higher cost. She found the carrier had handled almost every route the house used. Nobody noticed the concentration because invoices arrived under several regional names.

The rival expected a shouting match. The house locked two new arrangements, changed purchasing terms, and built a delivery reserve. It cost money. It also killed a dependency that had been growing for years.

When the founder finally spoke publicly, he described the service customers could expect and the orders already recovered. The dispute got a paragraph. The improvement shaped the next decade.

The attack showed the bridge. The answer was a network of crossings.

The strike is also a report

Treat the event as intelligence as well as harm. Extract information before you decide how to answer. Contain the damage. Let the report show where the structure needs work.

What did the pressure reveal? A customer relationship hanging on one employee. An account with no recovery route. A public claim the company could not back with evidence. A vendor who gained power through gradual concentration.

Record the event without letting suspected motive replace observed facts. Service failure, commercial disagreement, accident, and deliberate interference can look similar at first. Misclassification sends resources the wrong way. Preserve evidence. Define the immediate consequence. Keep confidence levels attached to explanations.

Returning fire means restoring the ability to operate and improving the position. Enforce a contract. Correct a false claim. Replace a supplier. Repair a product. Decline further engagement. Choose the response for its effect on the enterprise.

Four pillars of the response

Intelligence. What happened, which assets or obligations are affected, how the condition arose. Assign someone to keep a chronology. Separate confirmed facts from reports awaiting verification. The person containing the fire should not also carry the whole investigation alone.

Realign power. Identify the dependency that allowed the damage. Reduce it, replace it, or keep it on purpose. A vendor dispute may resolve commercially and you should still check whether switching options are adequate. Repair the relationship and redesign the dependency in parallel when both are needed.

Reposition quietly. Prepare the operational change before you claim results. Transfer essential knowledge. Test the replacement. Establish a rollback where you can. Premature announcements can alert a counterparty while you still need cooperation you just made less likely.

Advance on purpose. Once service is stable, use the improved position: serve customers better, negotiate from stronger evidence, enter a market the old dependency blocked. That converts survival into advantage. Timing matters. Do not dress unfinished recovery as expansion.

Each pillar needs an output: evidence record, decision, tested capability, measurable benefit. Without outputs, the framework is language glued to improvisation.

Stillness that buys time

Stillness helps when it protects judgment and preparation. It hurts when customers stay uninformed or damage spreads.

Run two clocks. Operational: how fast must you contain harm, restore service, or meet an obligation? Communication: who needs to know what, and by when? You can move urgently inside and speak carefully outside. You can also issue an early factual update without guessing at cause.

Prepare a message with the known consequence, the action underway, the next update time, and how affected people get help. Do not attribute motives the evidence cannot support. When information changes, correct the record directly. Predictable updates cut the flood of "any news?" while the team works.

Stillness also stops you from amplifying the event. Before publishing, ask whether the audience needs it, whether it adds verified information, and whether it helps the company keep its promises. The urge to prove you noticed an insult is an expensive editorial standard.

Answer where you can change the outcome

Answer at the point where your work moves the result. A rival's cheaper price may call for a clearer product, stronger service, or a different segment. A hostile accusation may call for a concise correction backed by records. A departing partner may call for a cleaner operating model.

Asymmetry means changing the comparison. If another company can sustain a lower commodity price, matching it is not the only show of strength. Study the work customers still do around the commodity. Setup, reliability, education, recovery, integration may hold a valuable problem the rival leaves open.

Keep the response economically sound. Expensive service added to protect pride can make a losing position look "differentiated." Test what customers will pay for and what you can deliver repeatedly. Competitive pressure improves strategy only when it forces better decisions.

Sometimes the strongest response is a formal remedy. Let qualified professionals assess contractual or legal options when the matter requires it. Treat that route as a business decision: evidence, cost, probability, timing, defined objective. Unlimited appetite for vindication can consume the asset it claims to defend.

A reseller walks customers

A fictional MSP's former reseller contacts several customers and claims the provider cannot support them alone. The founder's first impulse is to publish the entire history of the relationship.

Ops verifies which accounts are affected, confirms access rights, and tests support. Real weakness: customers had been using the reseller's personal contact details. The company builds a direct support route, contacts affected customers with facts, and schedules a service review. It preserves disputed communications for appropriate review.

The message only works if delivery follows. Every affected customer must reach someone who can act. Every promised correction enters the work queue. Measure whether the new process resolves requests and whether customers stay by choice.

The reseller's behavior may still need a separate response. The company has recovered its direct relationship. Its future no longer depends on winning an argument in public.

End on a condition, not a feeling

Define the end before the campaign expands. Service restored. Access secured. False claim corrected. Contract closed. Dependency replaced. Once you hit the objective, stop spending unless new facts justify new action.

Review what you learned about yourselves. Which control existed only on paper? Which assumption never got tested? Who acted well? Which customer promise needs revision? Put the answers into the operating system.

Field order. For the next material incident, create one response sheet: confirmed facts, current harm, response owner, communication schedule, chosen remedy, cost limit, completion condition. After containment, authorize one structural improvement that reduces recurrence. Success means the company regains initiative and can point attention back at productive work.

Related

← Designing the Machine · Series index

How we run GlassHosting

This handbook is the doctrine behind how I build. GlassHosting operates with MSP discipline on what we run today. No invented uptime badges. Every location we show is one we run: Chicago, Toronto and Vint Hill, VA, with more as we grow. Read the MSP posture.

MSP standards Glass Gaming All guides