Joshua Love

Healthcare administration is Zite AI's version.

• The case for it: AI does not forget, tire, or make decisions based on office politics or gut feelings, so within three to five years, AI-driven tools will outperform seasoned experts on key tasks.
• Love urges readers to act: research AI startups in your field now, because early adopters will be harder to outmaneuver.

The post lists three uses to make the shift concrete: sales CRMs that listen to calls and generate timed outreach, emergency-room security cameras that flag high-risk patients on arrival, and legal tools that pull relevant case law while a lawyer interviews a client. These examples reach across sales, hospital security and law, beyond the healthcare-administration category Love's own company works in.

Read original

Because projections have rarely matched reality, systems fall back on rough rules of thumb or guesswork.

• Some set a viability threshold, such as 10,000 patients a year for a primary care site; others open a location assuming it will succeed, then close it about three years later.
• Love puts the failure rate at one in five new sites within the first five years, and says each closure diverts money from patient outcomes.

Two urgent care sites that look identical, in markets that look identical, drew 15,000 visits a year in one case and 5,000 in the other. Love uses the gap to explain why health systems stopped trusting their own projections.

Read original

The first was to use AI to predict commercial real estate development, buy land there and resell it; the idea stalled because Love and co-founder Jacob were graduate students without $100M.

• The second forecast how many customers a store should expect so retailers could choose a location.
• It reached the finals of the Cozad New Venture Challenge but did not crack the market after more than 100 customer discovery interviews.
• The turn came when a health system CEO described the same forecasting question in a harder form: systems need patient volumes to decide where to open clinics, set budgets and allocate staff.
• Unlike retail, not every patient generates revenue; a patient journey can involve emergency visits, referrals, research and word of mouth.
• Love says the priority is the outcome: building something that helps people get healthy instead of helping sell more hamburgers.

The CEO told them their system had paid $2M to a consulting firm for a site-of-care plan and another $1M to a second firm to validate it, then reacted to the Zite AI demo: "you mean AI can do this in seconds?

• Where were you when I hired those consultants!?"
• Love recounts.

Read original

Love describes the company's pivot in stages.

• The original plan was to use AI to predict where commercial real estate development would happen, buy land there, and later flip it.
• That stalled because, as they put it, they and co-founder Jacob were graduate students with no $100M.
• The next concept projected how many customers a store should expect so retailers could choose locations.
• It reached the finals of the Cozad New Venture Challenge, but 100-plus customer discovery interviews still did not crack the market.
• The turn came when a health system CEO explained that healthcare faces the same question in a harder form: systems must anticipate patient volumes to decide where to open clinics, set budgets, and allocate staff.
• Unlike retail, not every patient generates revenue, because the patient journey runs through emergency visits, referrals, research, and word of mouth.
• Love recounts the CEO paying $2M to a consulting firm for a site-of-care plan and another $1M to a second firm to validate the first, then reacting to a demo with "you mean AI can do this in seconds?"
• They close by saying they would rather build something that helps people get healthy than help sell more hamburgers.

For brands planning a campaign

This is the strongest founder-story asset: it shows a documented pivot validated by a paying healthcare buyer, not a market the founders chose on their own. It suits a narrative-led collaboration aimed at health system executives, with the $3M in stacked consulting spend making the problem tangible.

It is the sharpest and most quotable statement of scale in the material.

Love argues that health systems do not build pro-forma patient-volume projections for new sites of care, despite the size of the investments.

• Instead, they either set a viability threshold, such as needing 10,000 patients a year to make primary care work, or guess and check—opening a site, assuming it will succeed, and closing it three years later.
• The consequence they cite is that one in five new sites fails within the first five years.
• Their core argument is that past attempts rarely matched reality: two seemingly identical urgent cares in seemingly identical markets drew 15,000 visits a year in one case and 5,000 in the other.
• They frame each failed site as millions of dollars that could have gone to improving outcomes for communities.

For brands planning a campaign

Use this as the problem-statistics spine of a campaign.

• The one-in-five failure rate and the 15,000-versus-5,000 variance are concrete and memorable, and tie directly to budget consequences.
• They can anchor a short-form piece or a lead magnet aimed at planning and strategy roles.

Love argues that domain-specific AI is already reshaping industries—citing weather prediction, fraud detection, and supply chain optimization—while attracting far less attention than general-purpose AI.

• They position Zite AI as an example in healthcare administration.
• They predict a three-to-five-year window in which organizations either adopt AI-driven tools or fall behind, and list illustrative uses: sales software that listens to calls and generates timed outreach, security cameras that flag high-risk patients on arrival at an emergency room, and legal tools that pull case law while a lawyer interviews a client.
• AI, they argue, wins because it does not forget, tire, or decide based on office politics or gut feelings.
• They close by telling readers to research AI startups in their own field.

For brands planning a campaign

This cross-industry piece is useful when a collaboration needs to reach beyond healthcare operations or take a conversational rather than technical approach. The security-camera and legal examples are speculative illustrations of AI generally, not described Zite AI products; ask which capabilities, if any, are shipped before presenting them as products.

The post claims AI can process more information than people, does not forget or tire, and will outperform experienced staff on key tasks within three to five years.

• It gives three concrete illustrations of that shift: sales CRMs that listen to calls and time outreach, security cameras that flag high-risk patients arriving at an emergency room, and legal tools that pull case law while a lawyer interviews a client.
• The closing instruction is to research AI startups in one's own field before competitors do.

A forward-looking prediction, backed by a short bulleted list of examples, followed by a direct challenge to the reader.

The three examples span sales, hospital security and law rather than healthcare administration alone, so the post reaches beyond Love's own product category.

For brands planning a campaign

A brand selling vertical AI can use this framing directly: the argument is not that AI is generally useful, but that industry-specific AI creates a deadline. It suits campaigns that pair a category claim with named, tangible use cases rather than abstract capability language.

Examples: Source 1

Some set a rough threshold, such as 10,000 patients a year for a primary care site to be viable; others open a location, assume it will work and close it roughly three years later.

• They report that one in five sites fails within five years, projections rarely match reality, and two apparently identical urgent care sites in apparently identical markets can draw 15,000 and 5,000 visits a year.
• The stakes are multi-million-dollar investments and money that could have gone to patient outcomes.

Problem-first structure: state the counterintuitive practice, supply a failure rate and a variance example, then close on the cost to patients.

For brands planning a campaign

This is the most useful item for a healthcare technology brand. It names a specific, costly buyer problem with numbers attached, making it a natural anchor for a content collaboration on forecasting, site selection or capital allocation rather than a general AI theme.

Examples: Source 1

The founders abandoned it because they were graduate students without capital.

• Their next concept forecast customer traffic so retailers could pick store locations.
• It reached the finals of the Cozad New Venture Challenge but did not crack the market after more than 100 customer discovery interviews.
• The pivot to healthcare came through a conversation with a health system CEO, who described paying $2M to a consulting firm for a site-of-care plan and another $1M to a second firm to validate it, then reacting with surprise when Zite AI's technology worked in seconds.

First-person narrative in chronological order, anchored to specific figures: $100M, 100+ interviews, $2M and $1M in consulting fees.

For brands planning a campaign

For a brand that wants founder-story content, this supplies a complete narrative arc: an idea that failed for a concrete reason, validated discovery work and a customer conversation that redirected the business. It works for campaigns about building in healthcare, customer-discovery discipline or why domain knowledge beats a general model.

Examples: Source 1

The subject matter narrows from broad AI adoption to the operational and economic detail of healthcare site planning. Two posts use founder biography, and all three keep the language plain and reader-directed rather than technical.

The posts vary in structure, evidence type and voice.

For brands planning a campaign

A brand can plan content around this pattern rather than a single topic. Love is likely to produce usable content on subjects with specific cost or volume figures, and less useful content on abstract themes without numbers to anchor them.

Examples: Source 1 · Source 2 · Source 3

The posts say Zite AI makes domain-specific AI for healthcare administration, including forecasting patient volumes so health systems can plan sites of care, budgets, and staffing.

View example

In a post on physician staffing, Belkin says coverage gaps are addressed only once they become urgent and that proactive workforce planning means forecasting demand by service line before gaps emerge.

• A post on burnout supplies specialty-level numbers and treats burnout as a leading indicator of turnover rather than as stress.
• That structure and audience closely match Love's argument that health systems should forecast patient volumes before committing to a site.

Belkin focuses on the clinician workforce — recruitment, retention, locums and other flexible staffing models — drawing on 22 years of client experience.

• Love focuses on patient-demand forecasting and the AI product built around it; their posts describe a founder's account of a customer conversation and market failure rates.
• Follow Belkin for staffing and retention thinking; follow Love for site selection, volume forecasting and vertical AI.

Read original · Original 2

Their first-person, concrete style resembles Love's: a new director role, a specific dinner with named colleagues, and the argument that transformation happens where people, strategy, operations and storytelling meet. Both write plainly, avoid technical vocabulary and argue that the people side of an operational decision determines whether it works.

Brenda S. writes about leadership and workforce experience, not operations or forecasting.

• Their posts offer personal reflection and community-building — a career change and a small-group gathering they run — rather than figures on demand, cost or failure rates.
• Follow Brenda S. for leadership development and team culture; follow Love for the numbers behind a health system's capital and planning decisions.

Read original · Original 2

His post on moving from point solutions to a platform argues that decisions about access, capacity, patient flow and staffing should reinforce one another instead of sitting in silos — the same complaint Love makes about planning done guess by guess.

• His Labor Day post walks through a patient journey — bed preparation, transport, case management — and credits the roles a patient never sees.
• That concrete, plain-language approach matches Love's use of specific examples rather than abstractions.

Johnson writes as an operator rather than a founder: there is no product, no pitch and almost no numbers.

• His strongest material is narrative about how care actually moves through a system.
• Love's posts make arguments about forecasting, backed by figures and a company.
• Follow Johnson for patient flow, operational intelligence and how hospital departments fit together; follow Love for why the forecasting those operations depend on is hard to get right.

Read original · Original 2

Belkin's 11 April 2026 post argues that healthcare delivery organizations cannot treat physician staffing as a reactive function: patient access and revenue stability depend on anticipating clinician supply.

• That means forecasting demand by service line, tracking retirement and attrition, and building pipelines before gaps appear.
• A technology company whose product forecasts coverage gaps, staffing needs, or service-line demand is making Belkin's case, which they have already made publicly from an operator's position after 22 years of client work rather than a vendor founder's.

Belkin's posts focus on clinician supply and staffing pipelines, not patient-volume projections for new sites of care.

• The 11 April 2026 post covers workforce planning, so the one-in-five site failure rate and the 15,000-versus-5,000 urgent-care comparison anchoring Love's argument have no equivalent in Belkin's writing.
• The 28 April 2026 post reports specialty-level burnout figures and argues for locums, PRN, rapid-response, and fractional models, speaking to staffing and recruiting audiences rather than strategy and planning roles.
• Neither post makes an AI argument or shows a buyer's decision being made or changed.
• Ask for a product-focused example if the campaign centres on forecasting volumes with AI; Belkin brings workforce credibility, but not that product narrative.

Examples: Source 1 · Source 2

The 17 March 2026 post describes a dinner of six leaders run with no titles and no presentations, reports what people said about being stretched and what they were building, and invites readers to direct-message for one of ten seats at a San Diego gathering.

• The 11 August 2026 post announces a director role at Kaiser Permanente Care at Home across Southern California and Hawaii and traces a career through workforce wellbeing programmes, cross-functional implementations, sales and benefits, a consulting practice, and authorship.
• For a healthcare technology company seeking an executive roundtable, a dinner series, or a people-and-experience angle to open doors with health system operators, Brenda S. brings a documented convening practice and a first-person account of running operational work inside a health system.

For a campaign about site-of-care planning, Brenda S. is the weaker choice.

• The two posts—a job announcement and a leadership gathering—do not discuss patient-volume projections, site-of-care planning, forecasting methods, analytics, or AI.
• The failure-rate and variance numbers the campaign needs would have to come from another voice.
• The gathering post asks readers to contact Brenda S. directly for seats, making a hosted event a likelier collaboration than a written thought-leadership piece.
• A client that needs the argument itself rather than the room should look elsewhere.

Examples: Source 1 · Source 2

The 22 September 2026 post says health systems assembled technology one point solution at a time and now ask how those capabilities work together.

• It argues that the next phase of digital transformation will be won by connecting operational intelligence across access, capacity, patient flow, staffing, and care delivery rather than accumulating more technology.
• That is the purchase case a platform vendor is making, and Johnson has already published it.
• The 3 September 2026 post grounds the same theme in a personal account of a family member's fall, walking through the case managers, doctors, nurses, environmental services teams, transporters, and administrators whose separate roles a patient experiences as one journey.
• A technology company selling an enterprise platform rather than a single application will find Johnson's published position closer to its own pitch than Love's.

Johnson supplies the why-connect-it half of the campaign, not the how-to-predict half.

• Neither post contains forecasting numbers, failure rates, market examples, or a method for projecting patient volumes; the one-in-five site failure rate and the 15,000-versus-5,000 comparison have no counterpart.
• The 22 September 2026 post names technology and operational intelligence in general terms but does not single out AI.
• The 3 September 2026 post reflects on patient flow rather than planning or analytics.
• A campaign whose hook is AI producing patient-volume projections in seconds would need to pair Johnson with the quantitative case from elsewhere.

Examples: Source 1 · Source 2