The first three groups in our main guide, 18 buying signals that show a company is about to change IT providers, come from what firms do. This page covers the fourth.
What owners say in public is a buying-signal group built from four sources. First, owners on the record: a managing partner names a technology problem on a podcast, giving a firm, a quote, and a date. Second, forums and social posts, where owners and IT managers describe pain in plain language; most are anonymous, but a named firm posting publicly moves to the top. Third, AI mistakes in local courts: public rulings on fabricated citations give a dated reason to offer every firm in that county a governed AI setup, without naming the lawyer. Fourth, silence on AI among regulated advisers: a firm with no meaningful AI use while examiners ask about it is a candidate for an exam-readiness talk. These signals are low volume, sitting at the top of a list rather than replacing it.
Everything in this group is gathered within each platform's rules, by people reading, not by scraping.
1. Owners on the record
Owners on the record is a signal in which a managing partner names a specific technology problem during a podcast or webinar, and it predicts a buying decision because a spoken problem is a decided problem.
The signal has three parts: the firm, the quote, and the date. The record usually lives on a regional business podcast, an industry vertical podcast, or a trade association webinar, and the published episode page or recording carries the date the partner spoke. A managing partner does not name a problem in front of an audience without having already raised the same problem inside the firm; the public version of the problem follows the internal one, not the other way around. Once the words are on record, the firm needs a documented plan for the exact problem the partner named, not an internal workaround nobody outside the firm can see. That plan has to come from an outside party the partner can point to, whether the next question comes from a client asking how the firm handles its own technology or from a regulator asking the same thing during an exam.
By hand this is slow, because nobody on an MSP sales team can listen to every regional business podcast. An outbound system can watch for these mentions and hand the sales lead a quote and a link.
2. Forums and social posts
Forums and social posts are a signal in which owners and IT managers describe a technology problem in plain language, and it predicts a buying decision because public complaints show patience with the current provider has run out.
The record itself usually sits in an owner group on LinkedIn, a trade association member forum, a local business Facebook group, or a Reddit community built around that profession, each a place owners already gather to compare notes rather than a place built to be searched. A written complaint predicts a decision because the owner has already tried the obvious fix inside the firm and is now asking peers who they use instead, which is a vendor search happening in the open. The firm does not need a general pitch in response; it needs an answer to the exact question the owner asked, from a provider who can show the fix rather than describe it.
By hand this is slow because nobody can read every forum and group where owners in a territory talk shop.
Most posts are anonymous. The exception is a real name at a real company describing a real problem, which moves that account to the top of the list.
3. AI mistakes in local courts
AI mistakes in local courts is a signal in which a judge issues a public ruling on a fabricated citation from generative AI, and it predicts a decision because every other firm in that county needs a governed AI policy.
Courts are catching filings that cite cases which do not exist, from generative AI and never checked, a pattern the UNC Kathrine R. Everett Law Library tracks under judges' standing orders (case rules) nationwide. The State Bar of California's guidance on generative AI names the risk: tools that produce "plausible but inaccurate information, including purported citations to legal authority that does not exist."
Each ruling is a reason to talk to every other firm in that county about a governed AI setup: approved tools, a use policy, and training. ABA Formal Opinion 512 says firm managers must set clear policies and train staff. By hand this is slow because nobody reads every ruling in every county courthouse.
One rule, no exceptions: never name the lawyer involved. The ruling is the reason; the person is not the pitch.
4. Silence on AI among regulated advisers
Silence on AI among regulated advisers is a signal in which a registered investment adviser discloses no meaningful AI use while examiners inquire, and it predicts a decision because unaddressed AI oversight becomes an exam finding.
Registered investment advisers (firms managing client money, registered with the SEC or a state regulator) file public disclosures. The SEC Division of Examinations' fiscal year 2026 examination priorities say the Division will "assess whether firms have implemented adequate policies and procedures to monitor and/or supervise their use of AI technologies." The same document flags oversight of vendors under Regulation S-P (the SEC's customer-data privacy rule); 2024 amendments require a written incident response program. By hand this is slow because nobody reads every adviser's public disclosure filing and notices what it leaves out.
5. What an MSP sells into these signals
A named podcast problem is a lead for managed IT work built around it. The first project is an assessment of the exact system the partner described, ending in a written plan the partner can put in front of the firm. The managing partner decides, afraid it stays unfixed.
A named firm complaint is a lead for provider replacement, starting with a current-state assessment. The first weeks of that assessment produce a network and security review the firm did not have before, plus a migration plan built around it. An owner or IT manager decides, afraid of downtime.
A local ruling on fabricated citations is a lead for a governed AI setup: approved tools, a use policy, and training. The first project delivers the approved tool list and the written policy, with staff training rolled out once both exist. A managing partner decides, afraid of sanction.
A silent adviser is a lead for exam-readiness work: a supervision policy, a vendor inventory, and an incident response program. The first project produces a draft of each of those three, in a form the compliance officer can show an examiner. A compliance officer decides, afraid of a deficiency letter.
| Signal | What changed | What it needs | Service model | Who decides |
|---|---|---|---|---|
| Owners on the record | Partner names a problem | A fix for it | Managed IT | Managing partner |
| Forums and social posts | Firm airs IT complaints | A new provider | Assessment, then migration | Owner or IT manager |
| AI mistakes in local courts | Ruling names fake citations | Controlled AI use | Governed AI setup | Managing partner |
| Silence on AI | Examiners ask, no answer | Exam-ready records | Exam-readiness work | Compliance officer |
SiteSmith builds target lists around signals like these, on top of an MSP's ideal-client base list.
6. Two caveats
First, nobody buys the day the event happens; follow-up runs over weeks.
Second, the sharpest signals are low volume: few managing partners say a technology problem out loud. These accounts sit at the top of a list built to the MSP's ideal-client definition, not the whole list.
Questions owners ask
Does a podcast mention actually mean a firm is ready to switch IT providers?
Not by itself. It shows an owner decided a problem is real, but nobody buys the same week.
Are forum posts about IT problems useful if the poster is anonymous?
Mostly for language. An anonymous post shows how an owner describes a problem, but reaching that person by name is usually impossible.
Why do AI citation mistakes in court matter to an unrelated law firm?
A ruling in one firm's case gives every other firm in that county a reason to check its own AI policy.
What does it mean if a registered investment adviser never mentions AI in its filings?
It can mean no policy exists yet, while examiners say they will ask about one. Silence is not proof, but it removes a reason to assume one exists.
Do these four signals replace a normal target list?
No. They sit on top of a list built to an MSP's own ideal client definition, since few owners say a problem out loud in public.
Where to go from here
For the full list, see 18 buying signals that show a company is about to change IT providers. The public-records group is the natural companion to this one: see filings and public records that predict an IT decision. To build a sales handoff, read how to build an MSP outbound system.
Sources and editorial note
SiteSmith publishes practical operating guidance and cites external sources for factual industry and security claims. This article is not legal, regulatory, or cybersecurity advice.
- SEC Division of Examinations, Fiscal Year 2026 Examination Priorities (PDF)
- SEC, press release on the 2024 amendments to Regulation S-P
- ABA Formal Opinion 512, Generative Artificial Intelligence Tools (PDF copy hosted by LawNext)
- State Bar of California, Practical Guidance for the Use of Generative Artificial Intelligence in the Practice of Law (PDF)
- UNC Kathrine R. Everett Law Library, Judicial Guidance on the Use of GenAI in Court