KDP vs Lulu vs IngramSpark: self-publishing platforms compared

There is no universal winner. The right platform depends on what you expect the book to do.

KDP's standard print royalty is now 50% or 60% of list price, less printing costs, depending on the book's price and marketplace. Expanded Distribution remains 40%. This is an important change from older comparisons—including the previous version of this article—which simply described the standard KDP paperback royalty as 60%.

Lulu's print economics work differently. For retail distribution, 50% of the retail price goes to the distribution channel; the manufacturing cost is then deducted, and Lulu divides the remaining gross profit 80/20 in favour of the creator. Books bought directly through the Lulu Bookstore do not incur that retail-distribution deduction, so direct sales can generate substantially more per copy.

IngramSpark is not really a simple "royalty percentage" model for print. The author/publisher chooses a list price and wholesale discount, then receives publisher compensation after the relevant print and distribution costs. It is free to open an account and upload a print book or ebook, while ebooks currently earn 85% of the net revenue IngramSpark receives from the retailer.

For most serious business authors, therefore, this is not actually a three-way winner-takes-all decision. KDP and IngramSpark can complement one another extremely well, provided the ISBN and distribution settings are handled correctly. IngramSpark explicitly confirms that authors who own their ISBN can use KDP and IngramSpark together; the major caveat is that KDP Expanded Distribution should not also be used for that edition.

It is tempting to choose a self-publishing platform by looking for the largest royalty percentage. That can be misleading.

For a business author, the value of a book may be much larger than the £2, £4 or £6 earned on the individual copy. A £15 book that wins a consulting engagement, speaking booking or corporate client has done its job regardless of whether the print royalty was £3 or £4. That does not make unit economics irrelevant—it simply means they should be evaluated alongside distribution, presentation and operational reliability.

For print books sold through Amazon's standard KDP distribution, the basic formula is:

(50% or 60% × list price) – printing cost = royalty

The royalty tier now depends on list price. On Amazon.com, for example, a qualifying print book at US$9.99 receives the 60% rate, while US$9.98 falls to 50%; equivalent thresholds vary by marketplace. Expanded Distribution still pays 40% less printing costs, and hardbacks are not eligible for Expanded Distribution.

This changed in 2025, so old advice saying that KDP print books simply earn 60% is now inaccurate. The difference can be significant if you have deliberately priced a short paperback cheaply.

For Kindle ebooks, KDP still provides 35% and 70% royalty options subject to its eligibility rules. A useful 2026 change is that, from 7 July 2026, the Amazon.com price ceiling for the 70% option increased from US$9.99 to US$12.99, with equivalent changes in other marketplaces. In the UK, KDP now gives a £1.77–£12.99 VAT-inclusive range for the 70% option. That gives business authors much more room to price substantial non-fiction ebooks above the old £9.99 ceiling without automatically dropping to the lower royalty plan.

Lulu often gets described as paying "80% royalties", but that shorthand needs explaining.

For a print book sold through Global Distribution, 50% of the retail price first goes to the distribution channel. Manufacturing costs are also removed. Lulu then takes 20% of the resulting gross profit and the creator receives 80%.

Lulu's own worked example makes the difference clear. On a US$21 book with a US$5.50 manufacturing cost, its example produces US$4 of creator revenue when sold through retail distribution but US$12.40 when sold through the Lulu Bookstore.

That makes Lulu particularly interesting where you can drive the sale yourself—for example from an email list, website, course, community or business audience—rather than relying completely on third-party retailers.

IngramSpark gives publishers more control over the trade terms. For US and Canadian markets it currently allows wholesale discounts between 40% and 55%; internationally the range can start at 35%. Ingram describes 53–55% as the conventional range most likely to give bookshops their normal trade discount.

The catch is obvious: a more generous retailer discount leaves less for the publisher. Making the book returnable may improve its attractiveness to physical bookshops, but it also exposes the publisher to the financial risk of returns. Ingram warns that many bricks-and-mortar stores will not order a non-returnable title and that the publisher bears the cost when returnable copies come back.

That is an important business decision, not a box to tick automatically.

If you primarily want your book available to order by bookshops, rather than actively campaigning for significant shelf stock, you may decide that maximising the discount and accepting returns is not worth the margin and cash-flow risk. Conversely, an author running a coordinated trade campaign may regard those terms as part of the cost of getting the book stocked.

One welcome change is that IngramSpark announced free file revisions from 1 January 2026. The old advice that authors have a short grace period and will then be charged for every revised upload is therefore obsolete.

That is a meaningful improvement for independent authors, although a free revision does not mean an instant one: an amended file still has to move through Ingram's ingestion and review process.

This is where the difference between the three platforms becomes more important than the headline royalty.

KDP's great strength is obvious: it publishes directly into Amazon's own ecosystem.

For an author whose sales will mainly come from Amazon searches, Amazon advertising, an existing audience being sent to Amazon or Kindle sales, that simplicity is difficult to beat.

KDP also provides Expanded Distribution for eligible paperbacks, allowing books to become available through additional distributors. But KDP itself makes clear that Expanded Distribution does not guarantee that a particular bookseller or library will order or list the book. It also requires that the ISBN has not already been submitted for distribution through another service.

That last point is crucial.

If you intend to use IngramSpark for wholesale distribution, leave KDP Expanded Distribution switched off. Trying to send the same ISBN into the book trade through both routes is where conflicts begin. IngramSpark gives the same guidance.

IngramSpark's biggest advantage is not that every Waterstones, Barnes & Noble or independent bookshop will suddenly place your book on a shelf. No self-publishing platform can promise that.

Its advantage is that the book enters Ingram's wholesale ecosystem with conventional publisher metadata, pricing, discount and returns settings. Retailers and libraries that already order through Ingram can therefore find and order the title through an established supply chain.

For business authors this matters more than it might initially appear. A potential client, conference organiser or corporate buyer may not want to order ten copies from the author's personal website. Being able to say that the book is available through normal trade channels can make purchasing easier.

The same applies to other authors whose strategy involves libraries, independent bookshops, schools or specialist retailers.

But available to order is not the same as stocked. Bookshops still make their own commercial decisions, and factors such as demand, professional presentation, wholesale discount, returns status, publicity and evidence that customers actually want the book all matter. Ingram itself notes that many physical stores expect trade-level discounts and returnability.

Lulu can put eligible books into Global Distribution, which sends book data to major bibliographic databases including Nielsen and Bowker and allows bricks-and-mortar bookshops to order titles through Ingram. Lulu says retail availability can take approximately 12 weeks or longer.

That makes the old characterisation of Lulu as simply a moderately paced alternative to KDP too optimistic for a deadline-sensitive retail launch.

There is also an extremely important current rule: Lulu's October 2026 distribution requirements state that titles already listed on Amazon or any other retailer are not eligible for Lulu retail distribution. Lulu also charges a one-off, non-refundable £4.99/€4.99/US$4.99 review fee when a title is first submitted to Global Distribution.

This changes how we would use Lulu.

Lulu remains useful as a printing and direct-selling platform, but we would not regard Lulu Global Distribution as a third parallel distribution layer for a book already running through KDP and IngramSpark. Its current rules make that approach unnecessarily problematic.

There is another practical issue we have encountered with Lulu: its bibliographic feed can contain a good, detailed book description and proper publisher information, yet the resulting retailer listing can still be incomplete. That is particularly frustrating when some fields appear correctly but others disappear downstream.

The distinction here is important. Lulu requires substantive metadata and sends its information into major bibliographic systems, but it does not ultimately control every retailer's product page. Its own guidance says Global Distribution can take 12 weeks or more, so authors should not assume that a retailer-page correction will appear immediately.

For a business author launching around a conference, media appearance or fixed campaign, retailer metadata should be checked weeks in advance rather than the night before launch.

This is probably the most important part of the 2026 update, because getting the ISBN workflow wrong can create exactly the sort of duplicate metadata, failed updates and extra admin that authors want to avoid.

Both KDP and IngramSpark can provide free ISBNs in eligible circumstances, but those ISBNs belong to their respective publishing ecosystems.

IngramSpark explicitly states that its free ISBN cannot be used on KDP, Blurb, BookBaby or another publishing platform, and that moving later to an ISBN you own means setting up the title under a new ISBN. Its free ISBN is currently available to US-based authors/publishers and uses the imprint Indy Pub.

Likewise, a KDP-assigned ISBN is not the ISBN to use when distributing the same edition through IngramSpark. Ingram's transfer guidance expressly says a KDP-free ISBN cannot be transferred.

For a business author building a long-term brand, buying your own ISBN is usually worth it. The publisher/imprint then belongs to you, and the same ISBN can identify the same exact edition as it moves through compatible distribution services. Each materially different format still needs its own ISBN: paperback, hardback and ebook editions should not share one number.

In the UK and Ireland, ISBNs are obtained through Nielsen; in the US the national agency is Bowker. IngramSpark itself recommends purchasing through the appropriate national ISBN agency.

For most business authors, our recommended order is:

ISBN agency → KDP → IngramSpark

There is an important qualification: on KDP, use your own ISBN and do not enable Expanded Distribution.

This recommendation is based on the way the systems currently work. IngramSpark explicitly says that where a title is already available on KDP but has never been in KDP Expanded Distribution, the author can simply upload it to IngramSpark as a new title; no title-transfer process is necessary.

KDP then fulfils Amazon print orders directly, while IngramSpark handles the broader wholesale route. Ingram itself describes precisely this KDP-plus-Ingram configuration and warns not to enable KDP Expanded Distribution because it can conflict with the ISBN entering Ingram's system.

That gives a typical business-book workflow like this:

Own ISBN registered with the national agency → identical metadata prepared → KDP paperback set up with Expanded Distribution OFF → proof and approve Amazon edition → IngramSpark setup using the same paperback ISBN → approve Ingram proof → enable Ingram wholesale distribution.

That does not mean the ISBN becomes "Amazon's" because it was entered on KDP first. If you bought the ISBN yourself, it remains your ISBN. What matters is that the title, author/contributor, format, imprint and other core bibliographic information agree across the ISBN record, KDP and IngramSpark. Ingram repeatedly emphasises the importance of accurate metadata, and KDP's own ISBN process checks publisher information against the ISBN record.

There is also a very practical reason to favour this route for a normal launch: KDP is generally the simpler place to discover an obvious cover, interior or Amazon-detail-page issue before introducing another distribution feed.

This changes if print pre-orders are central to your launch.

KDP does not provide conventional pre-orders for print books. Its scheduled-release function is not the same thing: customers cannot buy the print edition in advance in the way they can pre-order a Kindle ebook. KDP's actual pre-order programme applies to ebooks.

In that situation there is a strategic case for preparing the KDP edition as a draft, getting the ISBN and metadata established, then enabling the IngramSpark distribution route sufficiently early for retailer pre-order data to propagate before making the KDP print edition live around publication.

That is a more complex launch, however, because the Amazon page may first be generated from the Ingram feed. It is not the route we would choose unless a genuine print pre-order period matters to the campaign.

For most business books—where the priority is a dependable Amazon listing plus book-trade availability rather than gathering thousands of advance paperback orders—the cleaner KDP direct + IngramSpark wholesale workflow is easier to manage.

The frustrating part is that an ISBN does not create one magical central record that every retailer reads instantaneously.

Different parties maintain and receive bibliographic feeds, while retailers decide how and when those feeds are incorporated into their own catalogue. Ingram asks authors for extensive metadata—including title, description, contributors, imprint, subject classifications, publication date and pricing—because that information is transmitted to customers, libraries, retailers and wholesalers.

Lulu similarly sends data into Nielsen, Bowker and its retail distribution network, and warns that distribution itself can take 12 weeks or more.

That helps explain a problem we have increasingly encountered around IngramSpark launches: you can have the ISBN record, platform record and retailer record showing slightly different versions of the same information, or an older entry can persist after you believe you have updated it.

The safest practical response is to create one canonical metadata sheet before touching any publishing platform. Freeze the following information wherever possible: ISBN, exact title and subtitle, author/contributor spelling, imprint, edition/format, publication date, description, categories, keywords, page count once final, and recommended retail price. Then copy those values consistently into the ISBN-agency record, KDP and IngramSpark.

That recommendation is an inference from the way these services validate and distribute metadata, but it removes a considerable number of avoidable discrepancies.

And do not make casual changes to the subtitle, imprint or author name in one system during launch week because you think the wording looks slightly better. Metadata consistency is operational infrastructure, not just marketing copy.

This is an area where official timescales and real-world experience need to be separated.

KDP remains the platform we are most comfortable using when a book has to get live relatively quickly, although authors should stop relying on the old "24–48 hours and it'll definitely be there" assumption.

KDP says publishing and update times vary by book and action, with many new-book and detail-page processes taking up to approximately 72 hours and some print changes taking several days across different Amazon marketplaces.

KDP does not provide print pre-orders, while scheduled print releases have their own lock-in period. Kindle ebook pre-orders require the final manuscript before the submission deadline 72 hours ahead of release.

A notable 2026 improvement is greater flexibility for Kindle pre-orders. For pre-orders created from 2 September 2026, KDP allows dates up to 18 months ahead, and the release can be postponed twice provided the eventual date is no more than six months beyond the original.

For customer support, the sensible assumption is still to build contingency into the schedule. KDP has a formal Contact Us process rather than a guaranteed instant-response service; an author with a fixed launch should not treat support as an emergency publishing department.

There are really two very different Lulu timelines.

A normal Lulu project can be set up much more quickly for Lulu's own ecosystem, while physical production itself normally takes additional manufacturing time. Global retail distribution is a different matter: Lulu currently advises approximately 12 weeks or longer for the book to become available through retail partners.

That makes Global Distribution unsuitable for a last-minute decision such as, "The conference is in three weeks; let's get this into every retailer."

Its metadata requirements are also strict, and a non-compliant file or metadata record can be rejected. As of October 2026, Lulu explicitly requires titles, contributors and other information to meet its retail partners' requirements before distribution.

In practice, Lulu is good at providing more than a bare-minimum description and, when you own the ISBN, proper publisher information can flow through the bibliographic chain. The problem is that downstream retailer pages can still omit information or take a long time to catch up. That does not make Lulu a bad platform; it means authors need to distinguish Lulu's own metadata from what a third-party retailer eventually displays.

This is where our view has changed most sharply.

IngramSpark's official guidance says a new print title normally goes through file review in three to five business days before an eProof is issued.

Our recent experience, however, is that authors should leave considerably more headroom than that—particularly if a launch date genuinely matters.

There is corroborating evidence from recent user reports. September 2026 Trustpilot reviews include authors reporting titles remaining "In Processing" for around two or three weeks, support requests going unanswered and, in some cases, much longer support-resolution periods. These are self-selected customer reviews rather than representative performance statistics, so they should not be treated as proof that every IngramSpark job will be delayed. They do, however, show that the problem is not isolated to one author.

This distinction matters: IngramSpark still publishes a three-to-five-business-day typical file-review target, but we would no longer build a launch schedule around the assumption that it will always meet it.

When everything works, the platform's reach is extremely useful. When something does not work—an ISBN mismatch, a title stuck processing, metadata that will not update, an eProof issue—the additional wait for support can become genuinely frustrating, especially close to publication.

There is some good news. The previous article's warning about revision fees after the initial grace period is no longer current: IngramSpark announced that file revisions would be free from 1 January 2026.

So the problem in 2026 is less "a correction will cost me another upload fee" and more "how much time and administration will the correction create?"

For an important business-book launch, we would now work backwards and aim to have the IngramSpark edition effectively finished weeks rather than days before publication. That buffer is an editorial recommendation based on current experiences, not an Ingram-mandated deadline.

For business authors, print quality deserves more attention than it receives in most self-publishing comparisons.

A novel is normally purchased because somebody wants to read the story. A business book may also be placed on a conference seat, sent to a prospective client, photographed for a website, handed to a board member or sold after a keynote. In that context the book is partly a brand object.

That is why we would resist declaring any one platform the universal "best quality" printer.

Print-on-demand books can come from different production facilities, and paper, colour, binding and trim specifications can change the result dramatically. The meaningful comparison is not "Does IngramSpark beat KDP?" in the abstract; it is "Which platform produces this particular book, in this format, to the standard I want?"

IngramSpark provides considerably more trade-production choices around paper, colour and binding than the simplest KDP workflow. Its current setup includes multiple black-and-white and colour interior choices and a range of paperback and hardback configurations, depending on trim size.

KDP's smaller set of choices is often an advantage for first-time authors because there are fewer decisions to make. Lulu also has a strong print-production proposition and can work particularly well for projects where the author is more interested in direct fulfilment or specialist physical formats than conventional Amazon-first publishing.

The previous version of this article characterised IngramSpark as noticeably better in print quality and slightly more expensive. We would soften that claim now. There is no sufficiently reliable universal price or quality difference to justify saying that Ingram is always 30p more expensive or always superior. Printing costs depend on trim, page count, binding, ink, marketplace and changes to each platform's pricing.

The practical approach is much better:

Order a proof or author copy from the actual service you intend to use. Check the cover colour, spine alignment, paper opacity, margins, black density, photographs or charts, binding and overall feel. If the book is going to represent your business in front of high-value clients, spend the extra few pounds to inspect the product rather than relying on an on-screen preview.

Also remember that each format needs to be managed as its own publishing product. Paperback and hardback editions require different ISBNs, and IngramSpark explicitly requires a unique ISBN for each distributed format.

Several pieces of self-publishing advice that were reasonable a few years ago are now either incomplete or wrong.

KDP print royalties are no longer simply 60%. Standard print distribution now pays 50% or 60% depending on list price, with Expanded Distribution at 40%.

KDP's 70% ebook pricing window is wider. Since 7 July 2026, the ceiling on Amazon.com is US$12.99 rather than US$9.99, with corresponding marketplace changes.

Kindle pre-orders became more flexible in September 2026. New pre-orders can be created up to 18 months ahead, with greater postponement flexibility, but the 72-hour manuscript deadline remains important.

KDP requires disclosure of AI-generated content. Its current content rules distinguish AI-generated text, images or translations—which must be disclosed—from AI-assisted work, which does not require the same disclosure. This matters to authors using generative tools for covers, illustrations, translations or manuscript material.

IngramSpark revision charges are no longer the issue they once were. Free revisions arrived from January 2026, making the old "60-day grace period and then pay for every correction" warning obsolete.

But IngramSpark timing deserves more caution, not less. Its stated typical review remains three to five business days, while recent 2026 customer reports describe materially longer waits and slow support. Authors with fixed launches should plan for the possibility of delay.

Lulu Global Distribution is slower and more restrictive than many older comparisons imply. Lulu currently says retail availability may take 12 weeks or longer, charges a one-off distribution-review fee, and says titles already listed on Amazon or another retailer are not eligible for its retail-distribution programme.

Those changes are significant enough that a publishing strategy built from a 2023 or 2024 blog post can now lead an author into the wrong workflow.

For most business authors, our answer in 2026 is not KDP or IngramSpark.

It is:

KDP + IngramSpark, with your own ISBN and clearly separated responsibilities.

Use KDP for direct Amazon print distribution. Use IngramSpark for wider wholesale distribution. Use the same self-owned ISBN for the same paperback edition, and leave KDP Expanded Distribution switched off. IngramSpark expressly supports this arrangement.

Register and finalise the ISBN and core metadata before pushing the book through multiple systems. Then maintain a single master metadata record so the title, subtitle, author name, publisher/imprint, publication date, description and other core information do not gradually diverge.

For a normal launch, our preferred sequence is:

Own ISBN → ISBN-agency metadata → KDP direct → IngramSpark wholesale.

For a launch where a print pre-order is genuinely important, consider preparing KDP without making the print edition live, getting the IngramSpark trade edition distributed earlier, and bringing the direct KDP print listing live close to publication. That approach requires more care because KDP itself does not support conventional print pre-orders.

Lulu remains worth considering, especially when direct sales, fulfilment, a particular print format or Lulu's own storefront suit the project. Its direct economics can be attractive. But we would no longer recommend casually adding Lulu Global Distribution on top of an existing KDP/Ingram setup, because Lulu's current eligibility rules expressly exclude titles already listed with Amazon or another retailer.

For other authors, the balance shifts. An Amazon-led commercial-fiction author may care far more about KDP's Kindle ecosystem than bookshop ordering. A children's author may place colour reproduction and physical format first. An academic or specialist non-fiction author may value Ingram's library and trade supply chain. An author with a large direct audience may find Lulu's direct-sale economics particularly interesting. The "best self-publishing platform" therefore depends on what success means for the book.

But for a business book designed to build authority as well as generate sales, our 2026 ranking is:

Best overall publishing setup: KDP + IngramSpark with your own ISBN. Best for getting an Amazon edition live efficiently: KDP. Best for conventional wholesale and bookshop availability: IngramSpark. Best worth investigating for direct-sale economics and alternative print workflows: Lulu. Platform requiring the biggest schedule buffer at present: IngramSpark, based on the gap between its published review target and recent user experiences.

Most importantly, do not wait until publication week to decide how the platforms will fit together.

A self-published business book is still a publishing project. Treat the ISBN, metadata, proofs, retailer feeds and distribution settings with the same seriousness you would give the manuscript itself. Doing that work early is much less exciting than writing the book—but it is considerably less painful than trying to untangle duplicate ISBN records, outdated retailer information or a title stuck "In Processing" when your launch campaign is already running.

Brandspire's role as a self-publishing partner is therefore not simply about getting a manuscript uploaded. The real job is bringing the editorial, design, metadata, production, distribution and launch decisions together so that the finished book works as a professional business asset rather than simply existing online. That broader positioning is consistent with the purpose of the original article and its call to help authors navigate the complete publishing process.