# Android Business Apps: What Survives a Kenyan Week

> How to judge an Android business app for a Kenyan shop: offline behaviour, M-Pesa reconciliation, eTIMS, battery and what happens when a phone is lost.

Canonical: https://veirahq.com/blog/best-business-apps-android

Android is where most Kenyan business software actually runs, because it is what is already in the shop. The useful question is not which apps are best in the abstract but which ones survive the conditions: a connection that comes and goes, M-Pesa as the default way to pay, [KRA](https://www.kra.go.ke) [eTIMS](https://veirahq.com/blog/what-is-etims) to satisfy, a phone shared between shifts and a battery that has to last one. This sets out what to test before you commit, because the differences show up in week two rather than in the demo.

## What to judge a business app on before you install it

|  | Worth checking | Why it matters here |
| --- | --- | --- |
| Works offline | Whether it functions with no connection and syncs later | Most Kenyan businesses lose the line regularly, and an app that stops is not a tool |
| M-Pesa | Whether it handles Till, Paybill or Pochi natively | An app that cannot see M-Pesa leaves you reconciling by hand, which is the work you installed it to avoid |
| KRA eTIMS | Whether invoices are transmitted, not just produced | A PDF invoice is not an eTIMS invoice, and a VAT-registered business needs the second one |
| Data export | Whether you can take your records out | An app you cannot leave is a decision you make once and live with |
| Real cost | Monthly fee, per-transaction charges and what the free tier excludes | Free apps usually charge somewhere, and it is often per sale |
| Who answers | Whether support is reachable in your hours and your language | An overseas help centre is no use at 7pm on a Friday |

- Test offline behaviour mid-sale, not before opening the app, and check stock moved too.
- If payments and sales live in different apps, a person becomes the integration.
- eTIMS compliance requires certified software; a PIN printed on a receipt proves nothing.
- Individual staff logins are what make the record worth keeping.
- Ask what an export looks like before you load a year of data in.

1. The four conditions that decide it
2. What to test before you commit
3. How the choice goes wrong
4. A worked example: the app that worked until the network did not
5. How Veira handles it
6. Frequently asked questions

## The four conditions that decide it

Connectivity is the first. An app that only works online is not a shop app here; it is a shop app elsewhere that happens to run on an Android phone. The question is not whether it has an offline mode but what that mode does: does it keep selling, does stock still move, and does everything reconcile when the connection comes back.

Payment is the second. If M-Pesa confirmations live in one app and sales live in another, somebody reconciles them by hand every evening, and that person eventually stops. An app that records the payment against the sale it paid for removes the job rather than organising it.

Compliance is the third, and it is binary. Issuing a KRA-compliant tax invoice requires certified software; printing a PIN on a receipt does not count. The fourth is mundane and decides more than the rest: battery, shared use and what happens the day a phone is dropped or stolen.

Want to see it running in your own business first?

## What to test before you commit

Five tests, each doable in an afternoon with a real shop and a real phone.

1. 1Turn the data off mid-saleNot before you open the app -- during a sale. Then take two more sales, receive a delivery, and turn it back on. What you are watching is whether it kept selling, whether stock moved, and whether anything was lost or duplicated on reconnect: [selling offline](https://veirahq.com/blog/offline-pos-kenya).
2. 2Take a real M-Pesa payment and find it againTake a payment, then try to answer &#x27;which sale was this&#x27; from the confirmation code alone, an hour later. If that takes more than a few seconds, you will be doing it by hand every evening: [tracking M-Pesa payments](https://veirahq.com/blog/mpesa-transaction-tracking).
3. 3Ask for the eTIMS certification, in writingNot &#x27;is it eTIMS compliant&#x27; but &#x27;are you a certified integrator, and which control unit route do you use&#x27;. The answer is a fact, not an opinion: [connecting a POS to eTIMS](https://veirahq.com/blog/etims-pos-integration).
4. 4Run a full shift on one chargeScreen on, printing, scanning, with the real brightness a sunlit counter needs. An app that is fine for an hour and flat by four o&#x27;clock is a different product from the one you tested.
5. 5Hand the phone to someone elseCan they sign in as themselves? Can you see who did what afterwards? One shared login means a voided item, a discount or a missing sale belongs to nobody, which is the single most expensive thing on this list.

## How the choice goes wrong

### Judging on the feature list

Every app lists offline mode, M-Pesa and eTIMS. What differs is what each of those words means in practice, and that only shows under the test rather than in the description.

### Choosing by monthly price alone

The fee is the smallest number in the comparison. The real cost is the hour a day someone spends reconciling, or the stock figure nobody trusts.

### One shared login on the shop phone

It is convenient on the first day and unrecoverable on the day something goes missing. Individual logins are what make the record worth having.

### Assuming an app can be swapped later

Your catalogue, your stock history and your sales records live in it. Ask what an export looks like before you load a year of data in, not after.

### Running three apps that do not talk to each other

Sales here, stock there, payments somewhere else, and a person in the middle doing the joining. That person is the integration, and they are the part that breaks.

## A worked example: the app that worked until the network did not

A general store in Kisumu moved to an Android app that did everything asked of it: fast, cheap, clear reports, and staff picked it up in a morning.

Then a network outage ran through most of a Saturday. The app opened, showed a spinner, and would not take a sale. The shop went back to a notebook, served customers all day, and spent Sunday evening typing it in.

Stock was wrong for a fortnight afterwards, because the notebook had captured the money but not the items. Nothing about the app was bad. It had simply never been tested against the one condition that was guaranteed to happen.

Running a shop on memory and paper leaves money on the table: missed sales, stock you cannot account for, and receipts KRA will not accept.

Veira records every sale, tracks stock and keeps you eTIMS-compliant automatically, so the numbers look after themselves.

## How Veira handles it

Veira runs on Android phones and on the Veira handhelds, and selling continues when the connection does not: sales are taken, stock moves, and eTIMS invoices queue and transmit on reconnect rather than being lost or skipped.

Payments are recorded against the sale they paid for, so the evening reconciliation is a check rather than a reconstruction, and each member of staff signs in as themselves so the record names a person.

And the compliance side is not an add-on: Veira is a KRA-certified eTIMS integrator, so the tax invoice is produced as part of the sale rather than by a second device or a second app.

## Frequently asked questions

Judge an Android business app on the Saturday the network drops and the evening someone has to reconcile, not on the feature list. The apps that survive a Kenyan week are the ones that keep working when a condition everybody knows about actually happens.

## Keep reading

[See all Business guides](https://veirahq.com/blog/business)

## Veira for your business

On the Veira side of this: [the app itself, and how it keeps selling offline](https://veirahq.com/download).

[Browse Veira by business type](https://veirahq.com/solutions) · [Who we serve](https://veirahq.com/industries)

Veira files a compliant invoice on every sale and keeps selling when the line drops.
