How we make the content.

Lead author and editor: Eric Snyder, founder of Receipts Group. Marketing Director at Safeguard Impact ($800K → $2M/mo over 15 months). Former Operations Director at Cash Buyers Network ($0 → $6M/yr). Founder and exit of DeliveryLean. Receipts Group is operator-led content from someone who's been the buyer of marketing services for the last decade, not a content team optimizing for word count.

Every page on this site is reviewed and signed off by Eric before publish. No anonymous bylines. No "reviewed by editorial team" implying plural humans who don't exist. If the byline says Eric Snyder — which it does on every published guide — Eric read it, fact-checked it against the cited sources, and approved the final version.

Receipts Group's pillar and article content is AI-assisted, human-reviewed. First drafts of long-form research content are generated by a structured agent pipeline running on Anthropic's Claude. The output passes through 25 hard quality gates and a set of soft surgical-patch gates before a human ever sees it — and then it doesn't publish until Eric reads it, checks the sources, and approves it.

What AI drafts: pillar pages, cluster articles, blog research summaries, FAQ blocks, comparison tables, internal-link recipes. AI is good at the structural work — making sure headings cover the right subtopics, making sure the recipe (intro → comparison → pricing → FAQ) is followed, making sure schema and word counts and meta lengths are all within audit-clean spec.

What AI doesn't draft: the founder's bio, operator history, the case study (every screenshot is real), pricing claims (those come from current Receipts Group rate cards), the ROI deck math (derived from first-party operator data), client proposals, and the page you're reading right now. Anything where someone could read the byline and conclude "this is what Eric actually does" is human-written end to end.

Why we use AI for drafting: speed and structural consistency. A 2,000-word pillar with FAQ, comparison table, schema, internal-link coverage, and meta budget compliance takes a human writer 8–12 hours to get right. The agent pipeline does the scaffolding in minutes; the human time goes into the parts that actually matter — the contrarian wedge, the operator anecdote, the "no, that's not how this works in practice" rewrites — instead of into formatting and recipe compliance.

Why we disclose it: because the brand promise on the homepage is "receipts." Hiding the pipeline that produces the content would undermine the same standard we're selling. Plus, Google's official position is that AI-assisted content is fine if it's transparent and useful — transparency strengthens E-E-A-T; it doesn't weaken it.

The Receipts Group content pipeline enforces a fixed set of hard gates on every generated page before it's allowed to upsert into the database, plus soft gates that trigger a surgical-patch rewriter when they fail. A sample of what each page has to clear:

• · h1_keyword — the H1 contains the primary keyword in a natural form
• · meta_title — under 55 characters, no redundant brand suffix
• · meta_description — under 155 characters, descriptive (not boilerplate)
• · word_count — minimum thresholds by page type (pillars > clusters > blogs)
• · keyword_density — primary keyword present but not stuffed
• · block_variety — the page isn't a wall of one block type
• · required_links — minimum number of internal + external authority citations
• · og_image_alt — 60–125 char descriptive image alt, not "image1.jpg"
• · recipe_compliance — the composition recipe for that page type is followed
• · first_party_stat — at least one verifiable operator fact (not invented)
• · reading_level (soft) — grade 10–12, with trade vocabulary preserved
• · faq_question_length (soft) — FAQ questions stay under the length cap that AI extractors prefer
• · heading_length (soft) — sub-headings stay scannable

That's not the complete list — there are 25 hard gates plus the soft surgical-patch set — but it's a representative cross-section. The point is that "an LLM wrote it and we hit publish" is not what happens here.

Every external claim cites a primary source. We define a primary source as:

• · Tool dashboards we operate — Google Search Console, Microsoft Clarity, Ahrefs Site Explorer, PageSpeed Insights, SEOmator
• · Official documentation — Google Search Central, Schema.org, FTC, IRS, IndexNow protocol spec
• · First-party operator data — our own revenue numbers, our own ad-account performance, our own published case studies
• · Public regulatory filings — SEC, FCC, state license registries

What we don't count as a source: secondary aggregator blogs, AI search summaries, "industry studies" without methodology, vendor marketing pages quoting other vendor marketing pages, statistics with no traceable origin.

Live receipts on this site that are themselves the source:

• · Case study — the Safeguard Impact rebuild, every screenshot from tools we don't own (GSC, Ahrefs, SEOmator, PageSpeed, Microsoft Clarity)
• · SEO ROI deck — 5-year compounding math derived from first-party operator data
• · Charlotte Performance audit — Ahrefs Site Explorer pulls in real time, screenshots inline

Every page is timestamped with the original publish date and an "Updated for [year]" badge. Pillar guides get reviewed quarterly at minimum; cluster articles get refreshed when the underlying data moves (a Google algorithm update, a tool changes its dashboard, a regulation changes, etc.).

Found a factual error? Email eric@receiptsgroup.com with "editorial correction" in the subject line and a link to the page. Factual errors get fixed within 48 hours of being reported and verified. Significant rewrites get a new publish date; small fact corrections keep the original date with the "Updated" badge incremented.

We don't ghost-edit. If a page's claim is meaningfully revised, the page's "Updated" timestamp moves and the change is reflected in the sitemap lastmod. The version Google indexed last week is not silently replaced.

The relationships you should know about before reading anything on this site:

• · Safeguard Impact — Eric is the Marketing Director. Safeguard appears as a case study throughout this site. The work is real; the relationship is disclosed everywhere it's referenced.
• · Cash Buyers Network (housefastcashfl.com) — Eric was Operations Director during the $0 → $6M/yr build, pre-agency. When Cash Buyers appears as a portfolio case (e.g., the dead-domain rebuild on the homepage), the prior operator role is disclosed.
• · DeliveryLean — Eric founded and exited. No current commercial relationship.
• · Haven (havenhrg.com) — Receipts Group client; "Powered by Receipts Group" is in their site footer.

No other paid relationships drive editorial decisions. Receipts Group does not take affiliate commissions, referral fees, or sponsorship money in exchange for editorial mentions.