Social media monitoring, escalation and a plan for the first hour
Most social flare ups are not crises. The damage usually comes from treating a small one as an emergency, or from nobody noticing a real one until Monday morning.
What is social media reputation management?
Social Media Reputation Management is the monitoring, triage and response system a brand uses when criticism appears on social platforms. It covers listening across channels, a severity scale, an escalation path with named decision makers and rehearsed crisis procedures. It suits Australian brands whose customers complain in public and where a slow response compounds the damage.
Get a fixed written quote- Typical timeline
- 4 to 6 weeks to stand up, then ongoing
- What drives cost
- Depends on monitoring volume, how many brands and locations are covered, whether you need out of hours cover and how many people need training.
- Best for
- Consumer brands, franchises and organisations with public accountability
- You own
- The playbook, the monitoring configuration and every log we keep
- Built with
- Listening tools, severity matrices, escalation paths, holding statements
Your handover
What to listen for, and where the mentions actually are
Monitoring your own page notifications catches perhaps half of what is said about you. The rest happens where you are not tagged: a screenshot in a suburb community group, a comment thread under a news outlet's post, a video on TikTok that names you in the audio but not the caption, a Reddit thread, a review platform, or a message in a private group that reaches you only when a customer forwards it. Brands are frequently surprised by how much of the conversation sits outside their notifications.
- 01Listening configuration across platforms, keywords and variants
- 02Tuned alerting with noise reduced to a readable level
- 03Severity matrix with response times and named owners
- 04Escalation contact tree including deputies
- 05Approved holding statements for likely scenarios
- 06Published comment moderation policy
- Response playbook for recurring themes
- Tabletop crisis rehearsal and written debrief
- Exportable incident log and monthly sentiment reporting
The rest of the answer
So we configure listening across brand names, common misspellings, product names, executive names, store or suburb combinations and the phrases people use when complaining about your category. Then we tune it, because an untuned listening setup generates so much noise that people stop reading the alerts, which is worse than having none. The first month is mostly tuning: adding exclusions, adjusting thresholds and learning which sources produce signal for your brand rather than for brands in general.
Who is allowed to speak, and who decides
The worst moment to work out who has authority to make a public statement is while a thread is doubling every twenty minutes. Escalation paths need names, mobile numbers and a deputy for each name, because the one person with authority is invariably on a plane when it happens. We document that before anything goes wrong, and we test it with a call that is not a drill announcement.
Getting this line wrong in either direction is costly
We also set the boundaries of what an agency may say. We answer routine matters from the playbook. We do not comment on legal proceedings, employment matters, incidents involving injury, regulatory contact, or anything about an individual's personal information. Those get a holding acknowledgement and go straight to your named contact. Getting this line wrong in either direction is costly. An agency that answers everything will eventually say something that becomes evidence, and one that escalates everything is not providing a service.
How the engagement runs
The first-hour when something goes wrong
Speed matters, but the first move is not a statement. It is establishing what actually happened, because a fast public response built on an incomplete picture is the thing that turns a bad day into a bad quarter. The sequence below is deliberately structured so that acknowledgement happens quickly while the substantive response waits for facts.
- 01Stage 1Pause scheduled content across every channel before doing anything else
- 02Capture the recordScreenshots, timestamps, URLs and reach, saved somewhere outside the platform
- 03Stage 3Assess severity against the scale and notify the named owner for that level
- 04Stage 4Establish the facts internally with whoever actually knows, not with whoever is available
- 05Stage 5Acknowledge publicly if warranted, briefly, without speculation or admission of fault
- 06Stage 6Respond substantively once facts are confirmed, in one place, and link to it rather than replying separately everywhere
- 07Stage 7Monitor and log for at least seventy two hours, then debrief and update the playbook
Two decisions on your side that keep the project moving
We rehearse this. A tabletop exercise with an invented scenario takes ninety minutes and reveals the gaps every time: nobody knows the password to the account that needs to post, the approver is unreachable, scheduled content keeps publishing cheerful promotions in the middle of it, and no one has thought about what to tell frontline staff who are being asked about it by customers.
Choose the right level
A severity scale so that not everything is a crisis
The most common failure we see is a single angry comment triggering an executive call, while a genuinely serious safety allegation sits unanswered because it arrived at 6pm on a Friday. Both are triage failures. A written severity scale fixes both, because it tells whoever is watching the channel exactly what to do without needing to make a judgement call about seniority.
Level
01
1 Routine
Looks like
A single complaint about service, delivery or price
Response time and owner
Same business day, community manager, no approval needed
02
2 Elevated
Looks like
A repeated theme, a post gaining traction, or a visibly upset customer
Response time and owner
Within two hours in business time, community manager plus marketing lead notified
03
3 Serious
Looks like
Safety, discrimination, staff conduct, or media contact
Response time and owner
Within one hour, any hour, escalate immediately to the named executive
04
4 Critical
Looks like
Coordinated pile on, legal exposure, or an incident with real-world harm
Response time and owner
Immediate, crisis group convened, all scheduled content paused
How we work this out during scoping
The scale below is a starting point that we adapt per client. What makes it work is not the wording. It is that each level names a response time, an owner and an approval requirement, and that the people named have agreed to it before anything happens. A plan nobody has read is not a plan.
Moderation, hiding versus deleting, and the record you keep
Comment moderation needs a written policy, published where people can see it. Without one, moderation decisions look arbitrary and any removal becomes its own story. The policy should state plainly what gets removed, which is abuse, discriminatory content, spam, doxxing and anything unlawful, and state equally plainly that criticism of the business is not removed. That second part is what makes the first part defensible.
Deleting is more honest and more visible
Hiding a comment on some platforms leaves it visible to the author and their friends while hiding it from everyone else, which people discover and treat as deception. Deleting is more honest and more visible. Our default is to leave critical comments up and respond, remove only what breaches the published policy, and log every removal with the reason. If you are a government body or a publicly funded organisation, there may be record keeping obligations attached to that log, so we design it to be exportable rather than trapped in a tool. For organisations in that position we cover the surrounding requirements alongside public sector delivery.
When this is more machinery than you need
If you are a business to business firm with a quiet LinkedIn page and a few dozen clients who ring you when they have a problem, a full listening and escalation programme is over engineering. What you need is someone checking the channels regularly, a short written note on who to call, and good ordinary social media management. We will tell you that rather than sell a crisis framework for a channel with eleven comments a year.
The rest of the answer
This service is worth it when public criticism is frequent enough to be a pattern, when a single incident could reach media, when you operate across many locations, or when you are accountable to a regulator, a funder or a board that will ask what your process was. If your reputation problem is mostly about reviews and what a branded search returns, that is a steadier and cheaper programme, covered under reputation management.
How we scope it
Four ways to scope your Social Media Reputation Management project
We do not publish package prices, because the same brief can be a short build or a long one. These are the shapes the work usually takes. Tell us which one sounds like you and you will get a fixed written quote that spells out exactly what it covers.
Essentials
The core of it, scoped and quoted
Fixed written quote, agreed before work starts
- Listening configuration across platforms, keywords and variants
- Tuned alerting with noise reduced to a readable level
- Severity matrix with response times and named owners
Social Media Growth
The version most businesses need
Fixed written quote, agreed before work starts
- Everything in Essentials
- Escalation contact tree including deputies
- Approved holding statements for likely scenarios
- Published comment moderation policy
Social Media Platform
The largest version, built around your operation
Fixed written quote, agreed before work starts
- Everything in Social Media Growth
- Response playbook for recurring themes
- Tabletop crisis rehearsal and written debrief
- Exportable incident log and monthly sentiment reporting
Social Media Care
Ongoing support once it is live
Rolling monthly, quoted in writing
- A named engineer rather than a ticket queue
- Patching, monitoring and a tested backup
- Changes and improvements worked through monthly
- Rolling, cancel with 30 days notice
These are shapes, not menus. Most quotes end up somewhere between two of them, and we will say so when the honest answer is the smallest one. Describe the problem and we will tell you which it is.
Questions buyers usually ask
Frequently asked questions
How long does it take to set this up?
Usually 4 to 6 weeks. Listening configuration and tuning take the first two to three weeks because untuned alerts are useless. The severity matrix and escalation tree need workshop time with your leadership, since the value is in agreement rather than in the document. The tabletop rehearsal happens at the end, and it always changes something in the plan.
What does social media reputation management cost?
It depends on monitoring volume, how many brands and locations are covered, whether you need out of hours cover and how many people need training. Business hours monitoring for one brand is a modest scope. Twenty four hour cover across a franchise network is a different arrangement entirely. We define the response windows precisely and quote in writing.
Do you monitor outside business hours?
Only if that is what we agree and quote for, and we are direct about it because the gap between what a client assumed and what was scoped is where these relationships fail. Many brands find that alerting to a rostered internal phone after hours, with agency support from the morning, is the sensible balance. Genuine round the clock cover is available where the risk profile justifies it.
Should we respond to every negative comment?
No. Respond to genuine customers with a real problem, always. Respond once to a factual inaccuracy that others will read, then stop. Do not engage with abuse, deliberate provocation or an argument someone wants to prolong, because continued replies push the thread back into more feeds. Silence is a legitimate tactical choice and the playbook should say when it applies.
What do we own at the end of the engagement?
The playbook, the severity matrix, the escalation tree, the holding statements, the moderation policy and every incident log, all in editable formats. Monitoring tool accounts are set up in your name where the licensing allows. If you take this in-house or move to another agency, the system moves with you intact rather than needing to be rebuilt.
Can you help during an incident that is already running?
Yes, and we take those calls. The first tasks are the same regardless of when we arrive: pause scheduled content, capture the record, establish severity and find out what actually happened. We will not write a public statement about facts we cannot verify. Once it settles we run a debrief and build the playbook, because the second incident is much cheaper to handle than the first.
Related services
Stand up a monitoring and escalation plan before you need it
Tell us where your customers talk about you and who currently watches those channels. Inside one business day you will have a scope and a fixed price, both in writing.