Crisis Communications and Reputation Management Case Studies
Marketing5 Case StudiesUpdated Sep 2026
Reputation work rarely fails in the moment of a public incident. It fails earlier, in the quiet weeks when no one has confirmed who signs off a holding statement, what response time a client was promised, or which report the account director actually reads. The five cases below come from fictional agencies and in-house marketing teams that do crisis communications and reputation management. Each begins with a team under strain, then moves through diagnosis, redesign, and measurement. Read them as examples of how roles, service levels, workflow data, stakeholder expectations, and delivery targets interact under pressure.
Case Study 1
Missed response targets at a regional agency
Calder Pike Communications, a 64-person marketing agency whose reputation practice serves four retained clients
The situation
Calder Pike Communications runs a reputation practice for four retained clients, two of them regional hospitals and one a transit authority. Over the previous quarter, the rapid response desk missed its internal standard of a first holding statement within ninety minutes in 19 of 61 escalations. Duty officers, account leads, and the social monitoring analysts each believed a different person owned the final call on overnight messages. A hospital client had been told by its account lead that approvals took two hours; the duty officer had been staffing the desk alone since the last rotation change. Meanwhile new business pitches were pulling the same senior people away.
The challenge
Leadership wanted the desk back under its service level within the quarter, but the team could not agree on whether the problem was staffing, approval design, or priority. Account leads resented being called for overnight sign-off they had never agreed to cover. The analysts blamed slow handoffs. Two senior people had quietly stopped attending the desk review meetings, which meant no one was looking at the same numbers.
Analysis
Teams grounded in Crisis Communications and Reputation Management usually begin with the workflow record before they argue about people. At Calder Pike the desk logged timestamps for intake, draft, approval, and release, and those logs showed the delay sat almost entirely between draft and approval, not in intake. Median approval time for the 19 missed escalations was 137 minutes, while the approval step had no named owner in the overnight window. A RACI chart existed but nobody had signed it. Once the gap was visible, the disagreement about staffing became a question with a measurable answer.
Expectations and targets were the other half. Clients had been promised a first holding statement within ninety minutes, but nothing in the retainer defined who the client should call overnight, so the promise was being made by whoever answered the phone. Targets had been set at the agency level rather than the account level, which hid the fact that the hospital accounts carried most of the overnight volume. A workable plan needs a single overnight approver per account, a written service level in each retainer, and a weekly review that uses the same four measures every time: first-statement time, approval wait, reassignments, and client-reported satisfaction.
Actions taken
- Name one overnight approver for each client account in writing, with a backup, and publish the rota to the desk and the client contact sheet before the next rotation begins.
- Rewrite each retainer's response clause so it names the first-statement window, the approver, and the exception path for after-hours incidents, then have account leads sign it.
- Move the desk targets from agency level to account level, and report approval wait separately from drafting time so the bottleneck stays visible in the weekly review.
- Separate the analyst role from the approver role on the desk roster, so monitoring staff flag and route incidents while a named duty officer decides, and log each reassignment with its reason.
Results
Within the third week after the changes, the overnight approver was named on every account, and approval wait fell from a median of 137 minutes to 58. The share of escalations meeting the ninety-minute first-statement window rose from roughly two thirds to 91 percent over the following month. Two of the four retainers were amended at renewal to carry the new clause. The hospital client's account lead said the team had answered the question she kept asking, which was who would pick up at two in the morning. Senior people returned to the review.
Questions for discussion
- Which of the four measures would you add to the weekly review for a client whose volume is mostly overnight, and why?
- How would you handle an account lead who refuses to sign a named approver clause in a renewal?
- When analysts and duty officers disagree about whether an incident needs a statement, who should break the tie and under what timeline?
Case Study 2
Board chair escalates a delayed service update
Ashgrove Pell Communications, a 40-person agency whose reputation team handles issues work for Corvane Valley Transit Authority and two other public bodies
The situation
Corvane Valley Transit Authority suspended a bus route after a bridge inspection, and Ashgrove Pell Communications was asked to run the public and stakeholder messaging. The first two updates went out on schedule, but the third was four hours late because the operations director had not approved the revised timetable. The board chair, Marcus, sent an email to the agency's managing director that used the word unacceptable twice. He wanted a named contact, a daily written update, and a timeline for when the route would return. The authority's own service level had been set at six hours for each update, but nothing in the statement of work said who had to approve a timetable change.
The challenge
The managing director, Priya, needed to answer Marcus within a day without promising anything the operations team could not deliver. The account team wanted to send a long apology with a full timeline, but the timeline itself was unsettled. Operations, communications, and the legal office each had a version of what had been agreed at the briefing, and none of them had written it down.
Analysis
The first step was to separate what the chair was angry about from what he was asking for. His email named the late update, but the rest of the message was about having no contact and no return date. Those are different failures with different owners. The late update came from a missed approval step, which belonged to operations. The contact and the return timeline were communications responsibilities that the statement of work had never assigned. Once the two sets of expectations were mapped, it became clear that the agency had been answering a timing complaint when the chair was really asking for a single accountable voice.
Measurable commitments came next. A trust problem with a public body tends to stall when the response is only a promise to do better, because the chair has no way to check it. The better approach was to fix three things that could be verified: a named contact with a phone line that answered within thirty minutes during service hours, a written update every weekday at a fixed time, and a commitment to confirm the return timeline only after operations had signed the timetable. Each of these has a delivery target and a report that the chair could read without asking anyone.
- Clarify expectations Separate what the chair is angry about from what he is asking for
- Resolve immediate concerns Name a single contact and publish a fixed weekday update
- Commit measurable targets Write response time, update frequency, and approval rules into the contract
- Review with stakeholder Walk the chair through the three measures each month and adjust
Actions taken
- Send the managing director's reply within one working day, naming Priya as the single contact and stating the phone line and hours in the first paragraph.
- Ask operations to approve the revised timetable in writing by a fixed hour, and treat any approval that misses that hour as a published delay in the next update.
- Publish a written update each weekday at ten in the morning, with the status, the next checkpoint, and a clear line on what is still unknown, and keep the format identical.
- Add a service level to the statement of work covering response time, update frequency, and who approves timetable changes, then review those three measures with the chair each month.
Results
Before the end of the first week, the managing director had met the chair, and the chair confirmed in writing that the single contact was the right arrangement. Operations approved the revised timetable on the fourth day, and the route returned about a week later. Updates went out on time for every weekday after the change, and the service level was added to the statement of work at the next renewal. The chair's emails dropped from daily to once a week, and his latest message thanked the team for the updates rather than asking for a call.
Questions for discussion
- What would you have needed to see in the first forty-eight hours to be confident that the single contact arrangement would hold?
- How should an agency handle a client whose approval step is the source of the delay, without making the client feel publicly blamed?
- Which of the three verifiable commitments would you be most reluctant to put in writing, and what would you do about that reluctance?
Case Study 3
Rework loops in holding statement approvals
Brackett Ridge Foods, a packaged food company with 1,240 employees and an in-house communications team of nine that handles recall and quality messaging
The situation
Brackett Ridge Foods' in-house communications team issues holding statements when a product quality question reaches the press or a retailer. A holding statement needs input from the quality lead, legal review, a regulatory check for labeling claims, and a final sign-off from the communications director. Over the last half year, drafts were returning to the start in roughly a third of cases, usually because the quality lead's facts had changed after the draft left the team. Median time from first request to approved statement was 6.5 hours, against a target of three. Nobody had mapped the handoffs since the process was written, and each function described the delay as someone else's.
The challenge
The communications director, Nadia, had been asked to cut approval time to three hours without adding staff or weakening the legal check. Her team had already tried a shared tracker, which added another place to look. Legal said the review was not the bottleneck. The quality lead said the problem was that requests arrived with no defined facts. Both claims could be partly true, and neither had data behind it.
Analysis
Start with the handoff log. The team pulled timestamps for 84 requests and found seven handoffs, not four as the written process suggested. Two of the extra steps were informal: the quality lead forwarded drafts to a regional manager for an unrequested read, and legal sometimes returned a draft to communications instead of to the quality lead, adding an unplanned loop. Of the 84 requests, 31 returned to the start, and in 24 of those the draft had waited more than two hours for the next reviewer. Root causes sat in three places: unclear intake facts, untracked side reviews, and a sign-off rule that allowed anyone to reopen the draft.
Options were scored against the three-hour target, legal risk, and effort. Option one was a fixed intake form that required the quality lead to state the facts and a version number before a request could open, which addressed the largest root cause. Option two was a pre-approved template library for recurring recall questions, which would clear common requests without full review but needed legal to accept narrower sign-off. Option three was a new workflow platform, which promised routing and tracking but would take months to configure and would leave unclear facts unresolved. The team recommended the intake form and template library together, with the platform held back.
Actions taken
- Make the quality lead's intake form mandatory, with a stated facts summary and version number, and refuse to open a request that arrives without them.
- Write the side reviews into the process map as named steps with an owner and a time limit, and remove any review that cannot justify its slot in the pilot.
- Route every returned draft directly to the reviewer who requested the change, with the old and new facts side by side, so no draft waits in a general queue.
- Set the three-hour target as a shared measure for the quality lead, legal, and communications, reported weekly from the timestamps rather than from memory.
Results
During the pilot on one product line, median approval time fell from 6.5 hours to 2 hours and 40 minutes, and the share of drafts returning to the start dropped from roughly a third to under one in ten. Legal kept its full review on anything that touched a labeling claim, and the template library cleared about 40 percent of routine requests without a full review. Nadia said the larger gain came from the quality lead, who now arrived at each request with the facts already written down. The platform decision was deferred to the next budget cycle.
Questions for discussion
- Which root cause would you test first in a pilot, and what result would convince you to move to the next one?
- How would you keep legal's full review intact while still clearing routine requests faster, and who should approve that boundary?
- If the quality lead resists writing facts before a request opens, what evidence would you show them, and whose authority would you use?
Case Study 4
Spokesperson approval policy meets local resistance
Dunmere Freight Group, a regional freight operator with 3,850 staff across eleven depots and a six-person communications team
The situation
Dunmere Freight Group introduced a new policy requiring all media statements about driver safety, depot incidents, or customer claims to come through the central communications desk. Before the change, depot managers often answered local reporters directly, and some had given confident figures that later proved wrong. The policy added a required review step with a target of two hours for routine questions and four hours for incidents. Within the first month, eleven of the fourteen depot managers had sent at least one complaint, and the regional operations director had asked for the policy to be paused until the review step was tested.
The challenge
Gareth, the head of communications, did not want to pause the policy, because the last two incidents had shown how quickly a wrong number could spread. But he also could not ignore that depot managers felt cut out of conversations about their own sites. The policy had been written by a small group that did not include a single depot manager, and training had been a single hour-long briefing. Adoption had to be fixed without reopening the principle.
Analysis
The concerns sorted into three groups, and each group needed a different answer. Depot managers were worried about speed and about losing their standing with local councils and customers. Central communications was worried about accuracy and about being blamed for a message it had not approved. Regional operations doubted the two-hour target, which the desk had set without counting questions. The workflow data showed the problem. Depots had logged 212 media or customer questions, and 158 were routine, so the four-hour incident window was rarely the pressure point. The two-hour window applied to questions that often came in the late afternoon, when the desk had one person on shift.
Adoption was sequenced, not announced again. The first change was the target: routine questions would get a two-hour window only during staffed hours, and an out-of-hours route would go to a named duty editor with a ninety-minute window. Depot managers were then given a short list of pre-approved facts for common questions, which let them answer the reporter with confirmed information while the desk checked the rest. The policy was reissued with depot managers named as co-authors of the template library, giving them a stake in upkeep. Uptake was measured by the share of questions routed through the desk, the median review time, and corrections issued after publication.
Actions taken
- Hold a half-day session with depot managers to rewrite the two-hour and four-hour windows, recording each objection and the answer given, so the terms are theirs as well as the desk's.
- Issue a pre-approved fact sheet for the twenty most common questions, with a named owner for each sheet and a review point set inside the same month.
- Put an out-of-hours duty editor on a published rota, with the ninety-minute window in writing, and test it with two mock reporter calls before the policy is reinstated.
- Report uptake weekly as three figures, routed share, median review time, and post-publication corrections, and share the same sheet with depot managers and the operations director.
Results
Early in the second week of the revised windows, the share of depot questions routed through the desk rose from 61 percent to 89 percent. Median review time for routine questions fell from two hours and twenty minutes to one hour and five minutes, and the desk logged no corrections in the final two weeks of the review. Four depot managers asked to join the template working group, and the operations director withdrew his request to pause the policy. Gareth noted that two complaints had continued, both from a single depot, and that concerns were now logged by name.
Questions for discussion
- Which stakeholder group's concern would you address first, and what evidence would you need before changing a target that the desk set?
- How would you decide whether a pre-approved fact sheet is ready for depot managers to use without a review step?
- What would make you pause a policy rollout, and who should have the authority to call that pause?
Case Study 5
Choosing one reputation initiative with partial data
Kestrel Hollow Mutual, a mutual insurer with 2,300 employees and a marketing function that owns brand protection and issue monitoring
The situation
Kestrel Hollow's marketing leadership had budget for one new reputation initiative, with two more requests waiting. The options were a retained crisis agency on call, a social listening upgrade that would widen coverage to regional forums, and an employee advocacy program that would prepare staff to respond to questions in public channels. Each had a sponsor. The advocacy program was championed by the chief people officer, and the listening upgrade by the digital lead. The retained agency was recommended after a near miss, which had been logged as a single event. Leaders had some data for each option, but none of it was complete, and the proposals used different measures.
The challenge
Kestrel Hollow's chief marketing officer, Teresa, needed a recommendation that could be defended if the initiative failed, and she needed it before the next budget conversation. Each sponsor argued that their option would have prevented the last issue. None of them could show the counterfactual. The finance partner wanted a rule that could be applied to all three, not a case built on the strongest anecdote.
Analysis
Fix the criteria before scoring anything. Teresa's team settled on four: expected reduction in response time for issues that had reached the press, annual running cost, dependence on one person or vendor, and how easily it could be stopped. The chief people officer and the claims director each weighted the first two. Service levels entered the test as well, since an option that could not meet a four-hour response for claims issues scored poorly whatever it cost. The workflow data was thinner than anyone wanted. Only 9 issues in the previous period had needed an outside voice, so the team wrote down how small that sample was.
The recommendation was a staged decision rather than a single award. The retained agency was the strongest on response time but carried vendor dependence and a fixed annual fee, so it was funded for a six-month retainer with a review at the midpoint. The listening upgrade was cheap to reverse and scalable later, so it received a pilot budget. The advocacy program was deferred, since its benefit depended on staff who had not yet agreed to the role. Each stage had a delivery target: four-hour response on claims issues, coverage of at least 12 regional forums, and 40 trained staff willing to answer questions.
- 1Set criteria firstAgree four criteria and weights before any option is scored
- 2Score with uncertaintyRecord the sample size behind each score and flag thin evidence
- 3Check risk exposureTest vendor dependence, reversal cost, and service levels for each
- 4Fund stages, not totalsRelease money in tranches tied to named delivery targets
- 5Review at checkpointsStop, scale, or continue each stage on its stated evidence
Actions taken
- Fund the retained agency for six months with a written four-hour response clause on claims issues, and require a midpoint report that measures the clause against actual incidents.
- Launch the listening pilot in the regional forums with a fixed coverage count of twelve, and record which posts were found by the tool and which by staff, to test its value.
- Hold the advocacy program until a pool of 40 staff has agreed in writing to the role, and set a checkpoint for that review so the deferral does not become a quiet cancellation.
- Write the decision rule into the minutes: each option is reviewed at its checkpoint against its stated target, and any option missing its target loses funding before the next cycle.
Results
At the midpoint review, the retained agency had answered every claims issue within the four-hour clause, and the listening pilot had found 3 posts that staff had missed, though it also produced noise that needed filtering. The advocacy program stayed deferred, because only 22 of the 40 staff had agreed to the role. Teresa presented the decision to finance with the criteria, the targets, and the stage gates on one page. Finance accepted the staged approach, and the next budget conversation focused on whether to extend the agency retainer rather than on whether to keep it.
Questions for discussion
- Which of the four criteria would you weight most heavily if the claims director and the chief people officer disagreed on it?
- What evidence would convince you that a pilot on the listening tool had earned a larger budget, and how soon would you expect to see it?
- How would you explain a deferred initiative to its sponsor without suggesting the decision was a judgment on their judgment?
Learning Resources
Marketing CASE STUDIES
China, India, United States, Indonesia, Pakistan, Nigeria, Brazil, Bangladesh, Russia, Japan, Mexico, Philippines, Vietnam, Egypt, Turkey, Germany, United Kingdom, France, Italy, Canada, South Korea, Australia, Spain.