Everything Reality Index needs in the world — accounts, infrastructure, distribution channels, content strategy, ongoing operations — before, during, and after the v1 public launch. Pragmatic, not preachy. Decisions to make, things to set up, plans to commit.
01Accounts & infrastructure
Things to register or set up so the project has a presence in the places it needs one.
Web infrastructure
Hosting decision: static site host (Vercel, Netlify, Cloudflare Pages) or WordPress?
Current build is static HTML — easiest deploy is Cloudflare Pages or Vercel. Free tier, custom domain support, fast. WordPress only if you plan to author blog posts inside the site rather than externally.
Domain DNS pointed to host (realityindex.co)
Already registered. Just needs DNS configuration once host is chosen.
SSL certificate (HTTPS) provisioned
All static hosts above provide free SSL via Let's Encrypt or Cloudflare. Should be automatic.
Email forwarding for info@realityindex.co
Currently methodology docs point to info@realityindex.co. Decide: forward realityindex.co email to that inbox, or create a dedicated inbox?
Analytics (Plausible, Fathom, or GA4)
Privacy-friendly analytics preferred. Plausible or Fathom are paid (~$10/mo) but ethical and lightweight. GA4 is free but Google.
CDN / caching
Cloudflare in front of host gives free DDoS protection and edge caching. Recommended for any project that might go viral on X.
Decision needed
Host: Cloudflare Pages, Vercel, or Netlify? All are roughly equivalent.
Cloudflare Pages: best for projects that might get traffic spikes, since CDN is integrated. Vercel: best dev experience but pricing can surprise you under heavy load. Netlify: also fine, slightly older interface.
Social accounts
X / Twitter: @realityindex (or similar) — register if available
Tom already has 220K+ followers at @tomselliott. Decision: launch from existing account, register dedicated handle, or both?
Substack: realityindex.substack.com (or similar)
Discussed below in "Distribution strategy." Whether to host on Substack vs Tom's existing newsletter is a real decision.
Bluesky: realityindex.bsky.social
Lower priority but worth grabbing the handle to prevent squatting. Bluesky audience skews younger and more economically curious.
Threads / Instagram: lower priority but reserve handle
Don't actively post; just secure the names.
LinkedIn page (Reality Index, as a publication)
Useful for credibility with journalists and analysts who use LinkedIn for sourcing.
Decision needed
Launch from @tomselliott or create dedicated @realityindex?
From @tomselliott: immediate 220K reach, established credibility, but the project is then tied to one person. Dedicated @realityindex: cleaner brand separation, signals institutional commitment, but starts at zero followers and requires aggressive cross-promo. Both: post charts on Tom's account, build @realityindex slowly in parallel with consistent posting.
02Distribution strategy
How the project gets in front of audiences. Distribution is the actual product; the website is just where people land.
The full-court-press distribution flow
Every data update or news-driven moment follows the same flow, in this order:
1. Post canonical update on realityindex.co/news
Short editorial post explaining how the new data fits the larger picture and what trend it confirms or contradicts. This is the source of truth — all subsequent posts link back here.
2. Cross-post to Substack with full essay treatment
Substack post expands the /news post into long-form, embeds the relevant charts, and ends with a link back to realityindex.co. Goes out to email subscribers same day.
3. Thread on X with the headline chart + 3-5 bullet takeaways
Posts from both @tomselliott and @realityindex (if dedicated handle is created). Pinned for the day.
4. Cross-post to LinkedIn / Bluesky / Threads
Shorter version of the X thread. Less editorial work needed — same chart, same takeaways, different format.
5. Mention in Grabien's existing newsletter
Brief callout linking back to the canonical /news post.
The /news page is therefore the load-bearing infrastructure of the entire distribution strategy. Without it, the canonical post lives nowhere; every channel becomes its own micro-site of record. With it, every channel amplifies one consistent canonical update.
Substack vs Grabien Newsletter — the big editorial question
Reality Index needs a recurring publication channel where it can drop new data, commentary, and updates. There are three options, each with real tradeoffs.
Decision needed
Where does Reality Index publish recurring updates?
Option A — Hosted on Substack as a standalone publication. Pros: independent brand, easy paid-tier conversion later, native email + RSS, audience portability. Cons: starts at zero subscribers, requires consistent posting to build, separate auth/edit flow from Grabien's existing infrastructure.
Option B — Hosted on the Reality Index site itself (using a static-site CMS like Astro, or a simple blog directory). Pros: full design control, no platform dependency, all traffic comes to realityindex.co. Cons: more engineering work, fewer native distribution features, email delivery needs separate service (Buttondown, Beehiiv, etc.).
Option C — Published as a section/tag on Grabien's existing newsletter. Pros: instant existing audience, no duplicate auth, leverages Grabien brand. Cons: Reality Index brand subordinated to Grabien, harder to spin out later, requires Grabien editorial calendar alignment.
Recommendation: A. Substack gets you the cleanest brand separation, the easiest path to a paid tier if/when relevant, and the lowest engineering overhead. Cross-promote heavily from Tom's X and Grabien's newsletter to seed initial audience.
X / Twitter strategy
Posting cadence: 2-3 chart-driven posts per week
Each post = one chart from realityindex.co + 1-3 sentences of editorial framing + link. High signal, low filler.
Reactive posting: respond to viral inflation charts within 24h
When ZeroHedge / Kobeissi / etc. drops a chart, Reality Index quotes it with the corresponding RI page. Positions the project as the credible reference.
Monthly "Reality Index update" post tied to BLS CPI release day
Each BLS CPI release (typically mid-month) is a reliable hook. Build a calendar; pre-write skeleton tweets.
Charts-as-image strategy: every chart page generates a static PNG for sharing
Currently charts are SVG-on-page. Add a "download as image" feature or pre-generate share images. Twitter algorithm favors native images over links.
Cross-promotion targets at launch
Reach out to friendly financial-Twitter accounts pre-launch
Kobeissi Letter, ZeroHedge (already aligned data-wise), WSJ Daily Shot, Charlie Bilello, Larry McDonald. Goal: 2-3 retweets at launch from accounts with 100K+ followers.
Brief journalists at WSJ, Bloomberg, FT, NYT economics desks
Use Grabien's existing media relationships. Pitch as "we built the tool you'll want to cite next time inflation comes up." Offer exclusive embargo previews.
Prepare pitch for FT Alphaville / Marginal Revolution / Money Stuff
Economics-curious blogs that link generously. Reality Index methodology rigor is the angle.
Reach out to economics podcasts
Odd Lots, Macro Voices, Forward Guidance, EconTalk. ADI methodology is podcast-shaped — Tom is media-trained, this should work.
03Content & maintenance strategy
Reality Index is a living dataset, not a one-time publication. The whole project gets stronger with each monthly BLS update and breaks if it goes stale.
Data update cadence
Monthly: refresh BLS CPI subindexes and APU data on CPI release day
BLS releases mid-month, typically around the 10th-13th. Set calendar reminder. The data pull is scripted; refreshing 13 food items + utilities + health care anchors takes ~30 minutes.
Quarterly: refresh FHFA home prices, KFF health care data when available
FHFA publishes quarterly; KFF annually in September. Lower frequency, automate where possible.
All annual releases. AAA typically September, NCES typically August, Census typically September, IRS typically October.
Build refresh script that updates all data files in one command
Currently each data source is pulled manually. Consolidate into one Python script (refresh_all.py) that runs in <5 minutes.
Document the refresh process for handoff
So someone other than you can run the monthly refresh if needed. Notes on each data source location, expected file formats, common gotchas.
Editorial calendar
CPI release day post (monthly)
"BLS just reported CPI of X. The Reality Index shows what families actually paid for [headline category]." Template-able once we have the workflow.
Quarterly housing update tied to FHFA release
Housing is the biggest single inflation driver in the project. Tying updates to FHFA gives a regular cadence on the most editorially powerful topic.
Annual September: "American Dream Index updated for 2025" major post
When Census MHI for the prior year publishes (typically September). Big annual moment to refresh the ADI headline number.
Reactive: viral chart response posts
When a chart like Kobeissi's ground beef goes viral, Reality Index has a relevant chart ready to share within 24h. Maintain a "viral chart inventory" of pre-rendered RI charts mapped to common claims.
Content series ideas (Substack/long-form)
"Why CPI gets [category] wrong" — one essay per major divergence (housing, health care, education)
Each essay drills into the specific methodological choice CPI makes that produces the gap. Highly linkable, citable, evergreen.
"The American Dream by the numbers" — narrative breakdown of the ADI
For non-econ readers. Tells the story through one specific family in 1980 vs today.
Quarterly: "What's distorting CPI this quarter"
Each quarter, identify the category where the Reality Index gap is widening fastest. Make it a recurring beat.
04Launch sequencing
The actual order of operations for going from "ready to launch" to "live with momentum."
T minus 2 weeks
Soft launch & pre-brief
Site goes live at realityindex.co but is not yet promoted. Reach out to 5-10 trusted journalists and friendly financial-Twitter accounts with a private preview. Solicit feedback on methodology, copy, missing items. Fix anything they catch. Build social proof for launch ("X said Y about Reality Index" pre-quotes).
T minus 1 week
Final polish + scheduled content
Final pass on all pages. Pre-write the launch-day X thread and Substack post. Schedule the launch tweet. Set up monitoring for traffic spikes. Make sure the data refresh script works end-to-end. Confirm CDN is in place.
T minus 1 day
Stress test & staging
Load test the site (Cloudflare's load test simulator). Verify mobile rendering one more time. Have backup hosting ready. Make sure email forwarding works. Final read-through of methodology docs for typos.
Launch day
Multi-channel coordinated drop
Morning of a CPI release day or major economic news day is optimal. Tom posts a long-form thread on X with the ADI as the centerpiece chart. Substack publishes the launch essay (~1,500 words, the manifesto behind the project). Email to Grabien newsletter audience. Personal DMs to top-tier financial accounts inviting them to RT. Be on call to respond to questions and methodology critiques within 1-2 hours.
T plus 1 week
Second wave: long-form pickup
Pitch long-form pieces to FT Alphaville, Bloomberg Opinion, WSJ, NYT economics. Goal: 2-3 mainstream pickups in the first week. Pitch podcast appearances. Continue daily X posting on schedule.
T plus 1 month
First monthly data update
First post-launch CPI release. Reality Index updates its data, publishes the monthly update post, and demonstrates the recurring cadence. This is the moment that establishes the project as a living index rather than a one-time launch.
05Ongoing operations
Once launched, the project needs to be maintained or it dies. These are the recurring commitments.
Monthly data refresh (1-2 hours)
Refresh BLS data, regenerate charts, push updates. Most steps automatable.
Weekly X posts (~30 min/week prep)
2-3 chart-driven posts. Skeleton content prepared in advance.
Monthly Substack essay (1-2 days work)
One substantive post per month. Tied to that month's CPI release or quarterly data updates.
Reactive posts as viral charts appear (varies)
Time-sensitive responses to financial Twitter moments. ~1-3 per month average.
Quarterly methodology review
Sit with the data. Check whether the editorial framing is still right. Pick which v1.5 features to ship next quarter.
Annual major release: ADI updated with new Census MHI
September each year. Major moment to refresh the headline ADI number and revisit the project's editorial argument.
06Open questions
Things that don't have obvious answers yet. Worth thinking about before launch, not necessarily resolving.
Should Reality Index ever monetize?
Free forever is the simplest answer and probably right for v1. Paid tier (data API access, custom basket builder, white-label consulting) could come later if the project gets enough traction.
Editorial voice: principled-not-partisan or more sharp-edged?
Current methodology copy is principled-not-partisan. X account could be sharper-edged for distribution. Worth deciding in advance which voice goes where.
Who is the public face of the project?
Tom + Grabien attribution is current default. Worth deciding if there should be a dedicated "Reality Index" institutional voice separate from Tom.
How to handle methodology critiques and pushback?
There will be econometricians who attack the basket spec. Plan for engagement: ignore? respond once? maintain a public FAQ of past critiques and responses?
When does v2 happen, and what's in it?
v1.5 closes the immediate gaps (alt weights, comms rebuild, state-by-state). v2 could be geographic, demographic disaggregation (single parents, retirees, etc.), or international comparison. Worth roadmapping at 6-month mark.