Internal Draft · Concept only, not sent to Social5 or anyone · Pre-engagement · Oct 16, 2026
Social5 logo social5
Company Review
The Social5 Build · Internal

Company Review: Social5

This is a real audit, not a sample. It reviews Social5's public presence, its website, its own social, its review profiles, its SEO, and the partner channels where its provable reach actually lives. Every finding was checked on October 16, 2026 against public sources only. No outreach was made to Social5 or anyone else. Where something could not be verified, the finding says so in plain words, and analyst conclusions are labeled as inference. The review leads with the assets, since 2012, loyal clients, channel trust, because they are what make the findings worth fixing.

Prepared for: FirePower / TLN executive leadership (owners-side)  ·  Subject: Social5 (Social5, LLC), social5.com, digital marketing agency for small businesses and franchises, founded 2012, Draper, Utah  ·  Checked: October 16, 2026  ·  Method: public web sources only; no contact  ·  Status: internal concept draft, part of The Social5 Build, not sent to anyone

13evidence-graded findings
406PMI franchised outlets in its channel (end of 2025)
4.9★Birdeye aggregate, 98 reviews, profile unclaimed
1–3likes per post on its own Facebook (~11K followers)
3brand domains splitting its equity
404on the public affiliate / partner pages

Hear it, see it, ask about it

Listen: the podcast summary

In production

Two hosts walk through this audit in plain language: what we found, what the evidence says, and what we would fix first. The audio summary of this review has not been produced yet, this slot fills when it exists, and not before.

Watch: the video explainer

In production

The same audit as a visual walkthrough, wrapped in Social5's branding: the scoreboard, every finding with its status, the head-to-head, and the fix list at the end. Not produced yet, no player is shown until there is a real video to play.

Ask: an explainer agent for this audit

In production

An explainer agent trained only on this audit's facts is part of the standard Build. It has not been provisioned for Social5 yet, so this slot states that plainly instead of embedding a stand-in.

Part 1, The Assets: why this is worth doing

This review leads with the assets, because the assets are the reason for everything after them. The case, in one line: a strong brand and loyal clients, missing the modern stack. Nothing in Part 2 changes that, the gaps are system gaps, and systems can be built. What Social5 already has cannot be bought:

Asset 1, Since 2012, and a story clients can trust

Founded in 2012 in Draper, Utah, Social5 has operated since 2012, through the social era, the mobile era, and two ownership eras, still serving the same promise it started with: done-for-you marketing for small businesses, month to month, no long-term contracts, content reviewed by real people. Longevity is the rarest proof in this industry, and Social5's headline claim to it is true.

Asset 2, Clients who stay when they could leave any month

The public testimonials cite relationships of three and five-plus years, on a service with no contract holding anyone. Reviewers praise communication and a named account manager they ask for personally, and one cites measurable response growth for their business. A loyal client base on month-to-month terms is not a list; it is a verdict, renewed monthly.

Asset 3, Channel trust money cannot buy

Social5 sits inside PMI's franchise marketing program (406 franchised outlets at end of 2025), holds an exclusive member-benefit position with InterNACHI that members were still asking about in 2025, and built a licensee network that sold its services for a decade. Franchisors and associations do not hand those seats to agencies they do not trust, and they do not hand them out twice.

Asset 4, One genuinely modern muscle: search for the AI era

The SEO/GEO program is current in a way most agencies are not: AI crawlers deliberately welcomed, a GEO product on the menu, AI-search topics on the blog, multiple posts a week. It proves the team can run a modern discipline at cadence when the system supports it, which is exactly the bet the rest of this Build makes.

Part 2, The Audit

The proof problem: channels, not clients

Gap found

Social5 publicly names almost no direct clients. Its current homepage shows no client logos and no named case studies. The proof it does publish is carried by aggregate self-reported numbers, first-name-only testimonials, one anonymized directory case study, and its channel programs. For a company whose product is making other businesses visible, its own clients are invisible.

Evidence: social5.com homepage, accessed Oct 16, 2026, no client logos, no named case studies; testimonials name individuals (Cade Walker, David Smart, Janelle Jenkins, C Lund) but never their businesses. Audit file 02, headline finding.

In this Build: the fix-first list leads with named, permissioned proof, real clients, real numbers, on the record.

Its own numbers contradict each other

Gap found

The homepage claims "20,000+ Happy Clients · 150,000+ Projects Delivered · 13+ Years Experience · 5★ Avg. Client Rating." The About page claims "12,000+ Businesses Served · 98% Client Satisfaction." The two flagship pages disagree on the company's most basic fact, and neither figure is independently verified. A third-party directory separately lists "10+ active clients" with major clients marked "Confidential", aggregator data, treated as unverified, but it shows how far the public record drifts from the headline claims.

Evidence: social5.com and social5.com/about, accessed Oct 16, 2026; toppragencies.com/profile/social5 (aggregator, unverified). Audit files 00 and 01 §1.

In this Build: fix item 2, one set of numbers, sourced from the system that runs the work, on every page.

A reputation company with an unclaimed reputation

Gap found

Birdeye aggregates a 4.9-star profile from 98 reviews for Social5, and explicitly flags it as "not claimed by the business owner." The Manta listing is likewise "Unclaimed." Social5 sells reputation management and directory-listing services; in at least two public places, its own listings are unmanaged. The reviews themselves are good, the failure is that nobody is minding them.

Evidence: reviews.birdeye.com/social5-163405496192592 ("not claimed by the business owner"), accessed Oct 16, 2026; manta.com listing status "Unclaimed." Audit file 01 §3–§4.

In this Build: fix item 1, Social5's own profiles get claimed and completed in week one, as the demonstration of the review playbook.

Directory depth: perfect scores, one review deep

Gap found

Clutch shows 5.0 from a single verified review. The Manifest shows 5.0 from one review, the same engagement. iTRate shows 5.0 from one verified review. No Social5 profile surfaced on Yelp, the BBB, Trustpilot, G2, or Capterra. The newest directory review samples shown date from June 2024 and September 2020. That is a thin third-party footprint for a company claiming five figures of clients: the reputational issue is thinness, not negativity.

Evidence: clutch.co/profile/social5; themanifest.com/company/social5; itrate.co/company/social5, all accessed Oct 16, 2026. Searches for Yelp/BBB/Trustpilot/G2/Capterra profiles returned none. Audit file 01 §3.

Its own social does not perform

Gap found

Social5 posts near-daily templated promotional reels to a Facebook page of about 11K followers, and sampled posts draw 1–3 likes each, with almost no comments. Instagram (@social5_media, 1.7K followers) cross-posts the same reels to the same result: 1–3 likes per post. A social-media agency's own feed is the first portfolio a prospect checks, and this one demonstrates production volume, not engagement.

Evidence: Meta content-search index samples, Jul–Oct 2026 (posts dated Jul 13 – Oct 9, 2026), accessed Oct 16, 2026; instagram.com/social5_media. Audit file 01 §2.

In this Build: the content step treats engagement as the metric, not post counts, planned, produced, and measured inside the system.

No verifiable X or YouTube presence

Gap found

Social5 posts to X for clients, it is one of the channels in its social plans, but no official Social5 X account, handle, or follower count could be verified from public search. No official YouTube channel surfaced either, which is notable for an agency that sells short-form video. A Vimeo account under "social5" exists per the search index, but its details were blocked by a bot check, so its content is unverified. A LinkedIn company presence exists, but its follower count and posting cadence could not be verified publicly.

Evidence: public searches, Oct 16, 2026, no official X account or YouTube channel found; vimeo.com/social5 per search index (details unverified); LinkedIn via tomba.io listing (metrics unverified). Audit file 01 §2.

The growth engine was switched off: the dead affiliate program

Gap found

Social5 was built by partners, an affiliate program and a licensee network that put it inside franchise systems and national associations. Today the public affiliate and partner pages return 404. The plumbing still exists in the walls: the sitemap carries an affiliate subdomain with signup and academy paths, and partner subdomains per franchise brand. The heritage is intact; the program on top of it is publicly dead.

Evidence: social5.com affiliate/partner pages return 404 (Oct 16, 2026); sitemap.xml lists affiliate.social5.com with /signup and /academy, plus pmi.social5.com and other partner subdomains. Audit files 00 and 01 §1.

In this Build: Phase 2 of the plan rebuilds the partner engine, the concept partner page is already drafted in this Build.

Fulfillment without a platform, and the complaint that proves the risk

Gap found

Social5 sells roughly 27 à la carte services, fulfilled the way they were a decade ago: no client CRM the client can see, no AI that answers, no measurement on the wall. The failure mode that model invites is already on the public record: a one-star review dated September 21, 2024 states that Social5 had not posted to the reviewer's business Facebook account since May 17, yet kept billing monthly. That is exactly the failure a workflow-driven CRM prevents, scheduled work either ships and proves it, or an alert fires.

Evidence: Chamber of Commerce business directory, review by Dawn Markus, Sep 21, 2024 (paraphrased in the audit); social5.com/pricing menu (~27 SKUs), accessed Oct 16, 2026. Audit files 00 and 02 §1.3.

In this Build: fix item 4, every client on their own subaccount, where a silent stop is impossible to hide.

One brand, three domains

Gap found

Brand and search equity are split across at least three live domains: social5.com (the current site), social5.net (the legacy WordPress site, still live, still carrying 2023 content), and s5devgroup.com (the Development Group's site). Old content competes with and duplicates the main brand, and third-party data for the legacy domain shows it still carries measurable authority that belongs to the flagship.

Evidence: social5.net and s5devgroup.com live, Oct 16, 2026; legacy .net metrics via sitelike.org crawl (~Oct 2026). Audit file 01 §1 and §4.

In this Build: fix item 6, consolidate onto one domain, with redirects that keep the equity and end the duplication.

The one documented case study anonymizes the client

Gap found

The single documented engagement in the public record is a higher-education institution in Salt Lake City (1,001–5,000 employees), engaged since June 2019 and listed as ongoing when reviewed in September 2020: content and social strategy, SEO, a university news hub, and two newsletters, with named senior Social5 staff attached. It reads as one of Social5's largest direct engagements, and the institution's name is anonymized by the directory, so even Social5's best story cannot be told with a name on it. Whether the relationship continues in 2026 is not public.

Evidence: clutch.co/profile/social5 (verified review, reviewer Dan Felsing, Senior Director of Content Strategy, Sep 9, 2020); themanifest.com/company/social5. Audit file 02 §1.1.

Strength, credited: a genuinely current SEO / GEO program

Strength

Credit where due, this is the one modern muscle. The robots.txt deliberately welcomes AI crawlers (GPTBot, ClaudeBot, PerplexityBot and peers) with an in-file rationale: the point of its GEO/AEO blog programme is to be cited by AI assistants. The pricing menu sells a GEO blog product, the blog runs AI-search topics ("How to Get Your Business Recommended by ChatGPT," "Get Cited in AI Overviews & Perplexity"), and the sitemap shows multiple posts per week, about 95 blog posts, with the sitemap itself updated the day before this audit. The model is dated; the people running this program are not incapable.

Evidence: social5.com/robots.txt; social5.com/sitemap.xml (lastmod 2026-10-15, ~140 URLs incl. ~95 posts); social5.com/pricing (GEO Blog SKU). Audit file 01 §4.

Strength, credited: channel assets most agencies would kill for

Strength

Social5's real distribution is channels. It produces the Facebook content inside PMI's franchise digital marketing program, white-labeled, through the program operator, and PMI counted 406 franchised outlets at the end of 2025. It holds an exclusive member-benefit partnership with InterNACHI, the home-inspector association, launched in 2020 through Social5-led webinars and still surfacing in member discussion in April 2025. And its own licensee network, Social5 Development Group, sold territories from 2015, 46 units in franchise directories, with licensees selling and Social5 fulfilling everything. This network is the asset. It is why Social5 is worth rebuilding rather than replacing.

Evidence: PMI program features page ("Facebook content production by Social5"), live Oct 2026; PMI outlet counts per Franchise Chatter's FDD summary (406, end of 2025); InterNACHI webinar records 2020–2021 and forum thread Apr 28, 2025; S5DG franchise directories (46 units). Whether the InterNACHI partnership is contractually active in Oct 2026 is not public. Audit file 02 §2–§3.

Strength, credited: tenure, transparency, and a promise clients keep

Strength

The homepage testimonials cite relationships of three and five-plus years, on a month-to-month, no-long-term-contract service, where every client can leave any month. Praise concentrates on communication and the named account-manager relationship; one reviewer cites customer response up 8%, then 12%. The offer itself is transparent: an à la carte menu a prospect can read end to end before ever talking to sales. Longevity plus tenure plus transparency is a foundation a rebuild can stand on.

Evidence: social5.com homepage testimonials (Cade Walker 5+ years; David Smart 3+ years), accessed Oct 16, 2026; social5.com/pricing (full public menu); contact-page terms (month-to-month). Audit files 01 §1/§3 and 02 §1.2/§3.7.

Part 3, The Market It Plays In

Social5 does not compete the way a normal agency competes. Its reach runs through three channel arenas, and a fourth force, the platform shift, is changing what every player in all three arenas is expected to deliver. Facts only, from the audit record.

The franchise program arena, PMI, run by PMW

Where a franchisor's marketing program decides which vendors hundreds of locations ever meet

Strength of the position: systematized distribution no direct-sales agency can buy its way into.
Weakness: invisibility, the end client never learns Social5's name, so the channel builds PMI's brand, not Social5's proof.

The association arena, InterNACHI's member benefit

Where an association's endorsement is the marketing

Strength of the position: exclusivity plus trust transfer, the association's credibility arrives before Social5 says a word.
Weakness: single-channel dependence, and a benefit whose current contractual status cannot be confirmed from outside.

The reseller arena, Social5 Development Group

Social5's own licensee network: independent sellers, centralized fulfillment

Strength of the position: a decade of licensee heritage and a fulfillment engine built to serve sellers it never meets.
Weakness today: the public front door of this arena, the partner pages, returns 404 (finding 7).

The fourth force: the platform shift

What clients in every arena above now expect an agency to run on

Strength of the position: the demand side is already educated, Social5's own clients already buy AI lines from its menu.
Weakness: every month on the old operating model, platform-native competitors get to define what "modern" means to Social5's channels.

Head to Head

Social5 today vs Social5 on the Business Hub OS stack. The right column claims only what the platform's capability inventory records as of October 16, 2026, live means live, and anything in build or planned says so.

DimensionSocial5 today (from this audit)Social5 on the BHOS stack (per the capability inventory)
Client system of recordFulfillment without a visible platform; work can stop silently (finding 8)A CRM subaccount for every client, pipeline, contacts, conversations, follow-up workflows in one place. The subaccount system is live across the network (63 live subaccounts counted Oct 15, 2026).
AI answeringSells AI chatbot and voice lines from its menu; the platform underneath them is not publicNamed, per-client trained agents, the pattern is live today: Sarah (Pegasus Surfaces), Sterling (Argent Reach), Sage (Homeward Wise), each capturing into the client's own subaccount.
Reviews & reputationOwn Birdeye profile unclaimed at 4.9★/98; directories one review deep (findings 3–4)A guided review engine, live at reviews.businesshubos.com, branded per-client flows with a private path for unhappy customers; the claiming/monitoring/response playbook is in build, stated as such.
Content engineLeads: a genuinely current SEO/GEO program, ~95 posts, weekly cadence (finding 11)Keeps Social5's SEO/GEO strength and adds the BHOS daily blog + podcast pipeline, episodes are generated daily today; public distribution is not yet switched on, stated as such.
Own social proofNear-daily posts drawing 1–3 likes on ~11K followers; no verifiable X or YouTube (findings 5–6)Planned, produced, and published from one compose into one calendar the client can see, with engagement as the metric. Honest cap: only channels verified for the business get named, for Social5 that is Facebook and Instagram today.
Partner / affiliate enginePublicly dead: partner pages return 404 (finding 7)Rebuilt on the BHOS affiliate discipline, per-partner subaccounts and tracked referrals across offers, at [Level 1 rate - set at launch]; the affiliate earnings dashboard is a planned surface, stated as such.
Client learningSocial5 University existed historically, a single secondary source; treated as unverifiedA live shared learning hub (learn.businesshubos.com) with a dedicated Social5 track, one academy, no duplicated content.
Leadership visibilityAggregate claims that contradict each other across pages (finding 2)Mission Control, one command center where leadership sees every client, campaign, review, and open item; the numbers on the wall are the numbers in the system.
The front doorA digital business card line on the menu; no always-on entry point of its ownA virtual card built to the house standard, on-page QR, take-a-selfie, and a capture form wired into the right CRM subaccount. Stated honestly: the network-wide wiring audit (Oct 15) found that standard not yet met on existing cards, so this is a build commitment, not a finished claim.
Channel assetsLeads, and it is not close: PMI program seat, InterNACHI exclusivity, licensee heritage (finding 12)Unchanged, the channels stay Social5's. The Build's job is to give them a platform worth distributing.

Swipe to see all three columns →

The bottom line: the asset worth rebuilding is the channel network plus the fulfillment model, not a brand with public proof, because the proof was never built. Keep the channels, the name, the tenure, and the SEO/GEO muscle. Replace the invisible, unmeasured client experience with per-client subaccounts, AI agents that answer, a review engine that gets claimed and fed, an affiliate engine that exists again, and named proof where today there are only aggregate claims. None of that requires Social5 to become a different company. It requires Social5 to run on a platform, which is the one thing it never built.

What we would fix first

  1. Claim Social5's own house: claim and complete the Birdeye and Manta profiles and the rest of the review footprint in week one, the firm that sells reputation management demonstrates it on itself first. Phase 0 of the Build Plan
  2. One set of numbers: reconcile the client counts across every page and source them from the system that runs the work, so the headline proof can survive a prospect checking two pages. Phase 0
  3. Publish named proof: convert the tenure the testimonials already prove (3–5+ year relationships) into named, permissioned case studies, starting with the channels, where the evidence is already public. Phase 0–1
  4. Put every client on a subaccount: migrate in waves, channel clients first (PMI program first), so scheduled work ships with proof or fires an alert, the September 2024 failure mode becomes structurally impossible to hide. Phase 1
  5. Turn the review engine on per client: claimed profiles, guided review flows, monitoring, and human-approved response drafts, reputation as a product, not a hope. Phase 1
  6. Consolidate the domains: one flagship domain, redirects from social5.net and s5devgroup.com, equity kept and duplication ended. Phase 0–1
  7. Relaunch the partner engine: partner pages live again on the affiliate platform, aimed first at the channels that already trust the name, PMI franchisees, InterNACHI members, former licensees and resellers. Phase 2
  8. Make its own social the portfolio: engagement as the metric, the SEO/GEO program's topics carried into social, and the missing channels (a verifiable X and YouTube presence) decided deliberately rather than by default. Phase 3

Re-audit: per the Build standard, this review re-runs quarterly once a Build is live, findings are re-checked, each pass dated on this page, and a finding that changes gets a new dated entry, never a quiet rewrite. This is pass one: October 16, 2026.