What a free reading tool should include

Every pricing page in this field says free, and the word is doing different jobs on different pages. Sometimes it means free for a fortnight. Sometimes it means free to look but not to use. Sometimes it means you are the inventory. And occasionally it means what it says. Before comparing free reading apps, it is worth having a test for the word itself — one we are prepared to fail in public.

The question is not academic for us. We sell a reading tool with a free tier, and further down this page that tier gets the same treatment as everyone else’s — the ceiling measured, the people who hit it named. The field desk’s rule is that judging other products fairly earns you nothing unless the ruler also reaches your own desk. Consider this the demonstration.

Three disguises and a rarity

The trial in disguise is the commonest. Everything works on day one; on day eight or day fifteen the tool you have started to rely on asks for the card, and the “free” on the pricing page turns out to have been a date. There is nothing wrong with trials — software costs money to make — but a trial labelled as a tier is a small dishonesty at the exact moment a company is introducing itself, which is a strange moment to choose.

The feature-gated demo works differently: it never expires, and it never quite functions. The speed slider stops just before the interesting range; the voices are locked but audible in preview; the document formats you actually possess are premium. Search for a free speed reading app and most of what returns is this — a shop window with a working doorbell. Again, the model is legitimate. The question is only whether what remains below the gate is a tool or an advertisement for one.

The ad-funded reader pays its bills with your attention, which in most software is merely tiresome and in reading software approaches self-parody: an instrument for sustained attention, interrupted by design. Ads also imply tracking of some kind, and a reading tool sees what you read. The occasional rarity — the genuinely complete free thing, usually open source, sometimes a labour of love with a donate button — does exist in this field, and deserves naming as the standard the disguises are imitating.

Telling the disguises apart takes about two minutes. Read the pricing page bottom-up, starting from the fine print; look for the words “trial”, “limited” and “premium” in the free column; and ask of every listed free feature whether it is the tool or merely the door to it. If the page will not say plainly what happens on day thirty, assume the answer is the card.

What belongs in the free tier

Here is the argument, stated so it can be disagreed with. A reading tool’s free tier should contain the reading — the actual instrument, at full function, not a throttled sketch of it. It should contain every accessibility control, because charging a dyslexic or low-vision reader extra for the settings that make the tool usable at all is charging a ramp fee at the door. And it should contain the comprehension check, if the tool claims to do anything for comprehension — because a product that charges you to verify its own claim has an obvious conflict of interest.

• The reading itself. Whatever the tool’s core presentation is, the free tier should do it properly: the full speed range, real documents, no watermark on the experience.
• The accessibility controls. Motion reduction, spacing, contrast, voice — anything that determines whether a person can use the tool at all is not a premium feature, by definition.
• The evidence. If the tool makes a claim — faster, better retained, easier — the free tier must include the instrument that could falsify it. Claims you must pay to test are testimonials.

What may fairly cost money

Symmetry requires the other list. Things that cost the maker ongoing money per user — sync across devices, server-side anything — may fairly be priced. Things that accumulate — history, statistics across months — sit reasonably behind a membership, because their value accrues to committed use. What unites the fair list is that none of it decides whether the tool works today; it decides how well the tool compounds.

One category resists the split: features that improve outcomes, like adaptive pacing that ramps difficulty as your recall holds. A case runs either way — such features are ongoing, personalised and expensive to build; they are also, arguably, part of the tool working. We put them on the paid side and note the discomfort rather than resolving it. Reasonable tools will draw this line in different places, and the drawing is the thing to check.

Claims you must pay to test are testimonials.

The gate, stated plainly

Our own line, held to the same light

Now the sceptical reading of our own pricing, which the argument above obliges. The app is free at full function — every pairing mode, every speed, the recall check, the accessibility controls, documents parsed on your device — up to 5,000 words a day. Membership is $20 a month, or $192 a year, after a one-day trial. By our own taxonomy, a daily word ceiling sits nearer the trial-in-disguise family than we would like to admit: it is a meter, and a meter is a gate that moves.

So who actually hits 5,000 words? At an ordinary 250 words per minute, it is roughly twenty minutes of continuous reading. A person running a daily practice session will rarely touch it. A student streaming assigned chapters, a professional feeding reports through the app, anyone using this as their main way of reading rather than a way of practising — they will hit it before lunch, most days. The honest description of our free tier is: complete for practising, a hard ceiling for living in. If your reading day is bigger than twenty minutes of stream, we are asking you to pay.

Two more concessions while the light is on. A one-day trial is short — long enough to test the mechanism, not long enough to test the habit, and habits are what memberships bill against. And free-with-a-meter is still a funnel; we are a business, and the free tier exists partly so the paid tier has an audience. What we can say is what the meter never touches: the accessibility controls, the recall check and the on-device privacy are identical on both sides of the gate, because those were the three things the argument said must never be gated.

Using the taxonomy

The test travels, so take it with you. On any pricing page: find what expires, find what is throttled, find who is paying if you are not, and then locate the gate — the exact line where free stops. A well-placed gate sits between the tool and the conveniences; a badly placed one sits between the reader and the reading, or worse, between a disabled reader and the settings they need. The comparison work in speed-reading apps compared applies this line by line, and what people actually mean when they ask for the best free thing is taken apart in what “best reading app” means.

Free reading apps will keep multiplying, because the word is the best marketing in software and this field is no exception. The word is also, occasionally, true. The way to tell is not the typography of the pricing page but the position of the gate — and the companies worth your eventual money are the ones prepared to describe their own gate in the same tone they describe everyone else’s. That is the standard this magazine signed up for. Hold us to it, starting with the ceiling above.